Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
“The evidence assembled in Biodiversity and Climate Change leaps out at us like a scene from a
3-D movie. It’s warning us—compelling us—to act. Nowhere is the science clearer or the
facts more compelling than in this meticulously researched volume. When conservationists
and scientists of Lovejoy’s and Hannah’s caliber warn that if we don’t press forward faster,
we’ll see greater species extinction, more ocean acidification, more biodiversity loss, more
strains on agriculture and fishing—let me tell you: we need to listen. This book isn’t just a
call to heed the science; it’s a call to citizens everywhere to live up to their responsibilities
and protect this fragile planet we share.”
—John Kerry, United States Secretary of State, 2013–2017
“Mankind’s heedless extraction and pollution of our planet’s resources is tearing apart the
web of natural systems that has sustained our species throughout the long course of human
development. Tom Lovejoy and Lee Hannah have assembled a book that chronicles these
emerging ecological and climatic disasters yet gives hope that we can still help Earth’s sys-
tems heal and blunt the suffering of coming generations.”
—Sheldon Whitehouse, United States Senator for Rhode Island
“Biodiversity and Climate Change: Transforming the Biosphere serves as a comprehensive account of this
greatest of threats to humanity’s future. It will serve both as a textbook and a call to action.”
—From the Foreword by Edward O. Wilson
“An authoritative analysis of the increasing speed and scale of climate-change impacts on
our biodiversity, together with an illuminating set of specific ways to use our biodiversity to
address climate change. A powerful coupling.”
—Christiana Figueres, Executive Secretary of the United Nations Framework Convention on
Climate Change, 2010–2016
“In Biodiversity and Climate Change, the renowned Tom Lovejoy and Lee Hannah blend leading
voices to form a clear case for climate action, highlighting a powerful though underutilized
natural defense: ecosystem restoration.”
—Henry M. Paulson Jr., Chairman of Paulson Institute, and United States Secretary of the
Treasury, 2006–2009
“Biodiversity and Climate Change lays out the latest science on the central challenges of our time.
It is clear, comprehensive, and utterly compelling—an essential addition to the literature.”
—Elizabeth Kolbert, author of The Sixth Extinction: An Unnatural History
“We’ve all heard the clanging alarms about climate change. This book is an extraordinary
scientific portrait, expert, multifaceted, and up-to-date, of how those changes are affecting
biological diversity—upon which we humans depend, of which we are part. Don’t send to
know for whom the bell tolls. It tolls for us.”
—David Quammen, author of The Tangled Tree: A Radical New History of Life
“Lovejoy and Hannah generate a compelling story of the species extinctions that will ac-
company ongoing, rapid changes in Earth’s climate, coupled with the unrelenting pressure
of human population growth.”
—William H. Schlesinger, President Emeritus of the Cary Institute of Ecosystem Studies
“As the global community prepares a post-2020 Global Deal for Nature, this book reminds
us of the importance of ecosystems and nature-based solutions to advance the Paris Agree-
ment, the Sustainable Development Goals, and the 2050 Biodiversity Vision.”
—Dr. Cristiana Paşca Palmer, United Nations Assistant Secretary-General and Executive Sec-
retary of the Convention on Biological Diversity
“Come explore this comprehensive and insightful synthesis that underscores the intimately
interconnected nature of biodiversity and climate change—its past, present, and future.”
—Jane Lubchenco, Oregon State University
“Escape to Mars is not a realistic option for any species. Lovejoy and Hannah’s assessments
will help us chart a feasible pathway to preservation of our remarkable world.”
—James Hansen, Columbia University Earth Institute
Biodiversity and
Climate Change
This page intentionally left blank
Biodiversity
and E DI T E D BY
T H O M A S E . L OV E J OY
& L EE H A NNA H
Climate
Change
Transforming the Biosphere
F O R E WO R D B Y E DWA R D O . W I L S O N
N E W H AV E N & L O N D O N
Published with assistance from Gordon and Betty
Moore, Conservation International, and with assistance
from the foundation established in memory of Calvin
Chapin of the Class of 1788, Yale College.
10 9 8 7 6 5 4 3 2 1
For Betsy, Kata, Annie, Tia, and Jay—
and the generations to come
This page intentionally left blank
Foreword xi
Contents E DWA R D O . W I L S O N
Preface xiii
Acknowledgments xv
PA RT I OV E RV I E W: W H AT I S
C L I M AT E C H A N G E B I O L O G Y ?
PA RT I I W H AT C H A N G E S A R E
W E O B S E RV I N G ?
F O U R Phenological Dynamics
in Pollinator-Plant Associations
Related to Climate Change 42
E R I C P O S T A N D M I C H A E L AV E RY
PA RT I I I W H AT D O E S T H E F O U R T E E N Impacts of Ocean
PA S T T E L L U S ? Acidification on Marine Biodiversity 185
J O A N A . K L E Y PA S
E I G H T A Paleoecological Perspective
on Sudden Climate Change and FIFTEEN Tropical Forests in a
Biodiversity Crises 97 Changing Climate 196
J E F F R E Y PA R K J A M E S E . M . WAT S O N , D A N I E L B .
S E G A N , A N D J O S H UA T E W K S B U RY
N I N E Climate Change, Conservation,
and the Metaphor of Deep Time 114 Case Study 6: Postponing the Amazon
R IC H A R D B. A RONSON Tipping Point 208
D A N I E L N E P S TA D
Case Study 4: The Effects of Sea-Level Rise
on Habitats and Species 125 SIXTEEN Temperate and Boreal
C É L I N E BE L L A R D, Responses to Climate Change 211
C A M I L L E L E CL E RC , A N D L AU R E N B . B U C K L E Y A N D
F R A NCK COU RC H A M P JA N N E K E H IL L E R ISL A M BE R S
xi
xii FOR EWOR D
understand, are nevertheless characteristi- ism HIPPO, with the position of each letter
cally more subjective. “Ecoregions,” which representing an extinction factor and (usu-
are based in part on geographic data, are ally) in the order of its importance: thus,
even more subjective and more difficult to habitat reduction, invasion by nonnative spe-
classify. cies, pollution of habitat, population increase of
How many species still exist on Earth? humanity, and overhunting by excessive fish-
A bit more than two million have been ing, or hunting, or removal for other pur-
discovered, formally named, and given a poses. Climate change is perhaps the most
Latinized name. (The common housefly is important dimension of pollution impact.
denoted, for example, as Musca domestica.) We Finally, how well are conservation proj-
recognize that understanding of this ele- ects doing? In the case of land-dwelling ver-
ment of biodiversity, begun by the Swed- tebrates globally assessed at the time of this
ish naturalist Carl Linnaeus in 1735, is still writing, approximately one-fifth of land-
woefully incomplete except for plants and dwelling vertebrate species are classified as
vertebrates (mammals, birds, reptiles, am- threatened to some degree, and within that
phibians, fishes). The actual number of spe- only one-fifth of the one-fifth have been
cies on the planet, deduced statistically, is stabilized or improved in status.
roughly 10 million. The authors of the present work agree
Of the species known, how fast are spe- overall that climate change is the principal
cies disappearing due to human activity? threat to the survival of biodiversity even in
The estimates range between one hundred its early stages. Biodiversity and Climate Change:
and one thousand times the rate that existed Transforming the Biosphere serves as a compre-
before humanity spread around the world. hensive account of this greatest of threats
And it is accelerating. to humanity’s future. It will serve both as a
What exactly are the causes of extinction? textbook and a call to action.
They are often summarized by the initial-
The field of climate change biology dates to
Preface the mid-1980s, when the first symposium
was held on the subject at the National Zoo
in Washington, DC. The World Wildlife
Fund convened that meeting, and one re-
sult was the book Global Warming and Biodiver-
sity, edited by Robert L. Peters and Thomas
E. Lovejoy and published by Yale University
Press in 1992. At that point, the contribu-
tors and editors were largely confined to
examining how past biological responses
to climate change might predict change to
come. It was nonetheless clear, as one of us
remarked to Mostafa Tolba at a 1987 gather-
ing to help plan the Convention on Biologi-
cal Diversity, that if nothing was done about
climate change, one “could forget about
biological diversity.”
As the field grew and the imprint of cli-
mate change on the planet’s biodiversity
became more apparent, little more than a
decade after that initial meeting it was al-
ready clear that a completely new synthe-
sis of the field was in order. In 2001, we
began several years of gathering the best
authors and writing, a process that culmi-
nated in Climate Change and Biodiversity, which
we coedited and which was published by
Yale University Press in 2005. That volume
highlighted many of the numerous ways
climate change was affecting biodiversity
and biological systems. But change was still
happening, both in the real world and in
our understanding of impacts, faster than
it could be synthesized into print. Ocean
acidification was recognized as a major
global environmental problem by the time
the book appeared in bookstores, but it was
treated in part of only one chapter.
With the passage of another decade, a
truly staggering body of evidence suggested
that the planet was headed for serious bio-
logical trouble, so in 2014 we realized that
a new overview of this centrally impor-
tant topic was once again needed. Indeed,
xiii
xiv PR EFACE
what we first thought of as a second edi- chapters exploring possible biotic conse-
tion very quickly morphed into an entirely quences of change in a variety of systems.
new book. Today, as the world is pushed The evidence in these chapters suggests the
through the critical barrier of 1° centigrade importance of questioning the global target
of global warming, the fingerprints of cli- of limiting climate change to 2° centigrade.
mate change can be seen in biodiversity ev- Many systems will be past tipping points
erywhere. Already some major ecosystems or past acceptable limits of change at that
are experiencing abrupt change, and major point.
portions of the planet’s biology such as the Finally, we and our contributors exam-
Amazon are approaching tipping points. ine the importance of being more proac-
The impact of climate change in unfor- tive in conservation and policy. Where in
tunate synergy with other forms of human- the world can protected areas be placed for
driven change means that humanity is, in climate change? How should the protected-
fact, changing the biosphere. This picture area estate be redesigned to be resilient in the
is laid out well by this new team of lead- face of climate change? How can we man-
ing scholars in the following pages, as are age systems through extraordinary change,
some of the possible forms of adaptation and which policies are needed to facilitate
that might lessen the biological impact. that process? These chapters recognize the
This volume opens with an introduc- positive contribution that biodiversity can
tion to the topic and the relevant climate make; for instance, transforming the bio-
changes. Our chapter authors then look sphere through ecosystem restoration at a
at the world around us and the mounting planetary scale can reduce the amount of
body of evidence that climate change is climate change that biodiversity and hu-
already rewriting biological history. From manity must cope with in the future. The
hundreds of millions of tree deaths to mas- volume concludes by drawing all the pieces
sive mortality in coral reefs, the evidence of together at a planetary scale and asking how
climate change is already all around us, and we can feed the still-increasing billions of
it is writ large. people, maintain robust biodiversity, and
Placing these stunning changes in con- still accommodate change in both.
text requires understanding the past, and This book presents a current assess-
that perspective is provided by an out- ment of knowledge about climate change
standing set of paleoecological research- and biodiversity. More will continue to be
ers, examining both marine and terrestrial learned and understood on an ongoing ba-
realms. Movement has been the hallmark of sis about the interaction between human-
past biotic response to climate change, but driven climate change and the biological
genetic responses may manifest themselves fabric and systems of the planet. It is our
when bottlenecks are tight, as they will be fervent hope that this synthesis will lead to
in the future thanks to human conversion a widespread understanding that our planet
of natural habitats to other uses. works as a linked biological and physical
We then move from the past to the fu- system, and therefore will lead to less cli-
ture. Our learning from models of the fu- mate change and more surviving biological
ture and theory are explored in a series of diversity than would otherwise be the case.
We would like to extend our deepest thanks
Acknowledgments to all those who have contributed to the
creation of this book. The participants in a
special meeting called by the Science and
Technical Advisory Panel of the Global En-
vironment Facility provided inspiration,
including Camille Parmesan, Joshua Tewks-
bury, Gustavo Fonseca, Miguel Araújo, Mark
Bush, Rebecca Brock, Joanie Kleypas, Re-
becca Shaw, and Tom Hammond. Elizabeth
Hiroyasu went above and beyond, oversee-
ing and coordinating much of the produc-
tion of the book. Deepest thanks go to Jean
Thomson Black and Michael Deneen for
their support through the editing process.
We are indebted to Ed Wilson for his
continuing inspiration and insights con-
tained in the foreword to this book. We are
grateful to the Moore Center for Science and
the Climate Change Strategy Team at Con-
servation International for their insights
and support. For hard work on coordinat-
ing figures and assisting with graphics, we
thank Monica Pessino and the rest of the
Ocean o’ Graphics team at the University
of California, Santa Barbara. Funding from
Cary Brown and the Fiddlehead Fund has
supported many aspects of this project. The
National Science Foundation, the National
Science and Engineering Research Council
of Canada, and the Belmont Forum pro-
vided support that enabled major insights
in freshwater biology and ecological theory
presented here. Special thanks to Carmen
Thorndike for all that she does. And finally,
we thank a long list of colleagues, includ-
ing those who contributed chapters, with-
out which this endeavor would have been
impossible.
xv
This page intentionally left blank
PA RT I
Overview:
What Is Climate
Change Biology?
This page intentionally left blank
CHAPTER ONE People are changing the biosphere. Expand-
ing agricultural production, growing cities,
Changing the and increasing resource use and disposal are
among the human activities that are hav-
Biosphere ing profound effects on our planet. Some
of these effects are exerted and felt locally,
THOMAS E. LOVEJOY AND LEE HANNAH whereas others have global reach. Some hu-
man impacts, such as pollution, have been
effectively addressed at local or national
scales, and others, such as depletion of the
ozone layer, have been effectively addressed
through global cooperation. But many of
the effects people are having on the bio-
sphere have not been fully addressed, and
some of these have global consequences of
unprecedented proportion.
Two global environmental problems—
biodiversity loss and climate change—are
the preeminent environmental issues of
this millennium. Biodiversity loss—the
extinction of millions of species—is a very
real consequence of increasing human use
of the planet. Climate change due to green-
house gas emissions threatens to alter the
composition of the atmosphere and the liv-
ing conditions on the surface of the planet
for both people and all other species.
The purpose of this book is to explore
multiple dimensions of the interaction be-
tween climate change and biodiversity. Cli-
mate change biology is an emerging disci-
pline that boasts an explosion of literature
reflecting the importance and abundance of
climate change effects on the natural world.
The great interest in this topic, and the
large number of research findings emerg-
ing about it, combine to make a synthesis
of the whole field invaluable for research-
ers and students, as well as for professionals
in land use and conservation who increas-
ingly need to understand the field in broad
overview.
3
4 OV E RV I E W: W H AT I S C L I M AT E C H A N G E B I O L O G Y ?
We present the fundamentals of cli- streams. The need for integrated response
mate change for biologists in Chapter 2. to climate change and biodiversity has
The headline is that human burning of fos- never been greater, because changes in the
sil fuels for heat, transportation, and other biological world have profound implica-
purposes, combined with releases of other tions for regional and global climate. The
pollutants, is altering the thermal balance of way in which people relate to nature will be
the atmosphere. These pollutants, including forever altered by climate change.
carbon dioxide (CO2), methane, and others, A new synthesis is needed—a manage-
are collectively known as greenhouse gases, ment-relevant analysis of climate change
because they absorb and reradiate outgoing and biodiversity that transcends categories
longwave radiation, effectively warming the and conventions—to arrive at solutions that
Earth’s atmosphere. This warming has ef- provide a robust future for both people and
fects on atmospheric circulation, resulting in ecosystems. We now offer four themes that
a vast number of changes to climate, such as set the stage for the chapters that follow.
changes in precipitation, seasonality, storm
intensity and frequency, and many other fac-
tors, in addition to a pronounced warming of PL A N E TA RY BOU N DA R IE S
the planet’s surface. Increases in atmospheric
concentrations of CO2 and other greenhouse Planetary boundaries represent the “safe op-
gases have been measured beyond question, erating space” for the Earth system (Rock-
and the warming of the atmosphere and its ström et al. 2009). Climate change and
consequences are now being observed. biodiversity are two of the nine planetary
The next twenty chapters address the boundaries defined to date and represent
past, present, and future of biodiversity re- half of the planetary boundaries that are es-
sponse to climate change. Plants and ani- timated to have already been exceeded (the
mals all have climatic conditions that limit others being deforestation and nitrogen de-
their distribution, so this major and com- position). The planetary boundary for bio-
plex change in climate will result in mas- diversity is being exceeded because we are
sive shifts in the distribution of species and losing genetic diversity and robust popula-
ecosystems across the planet. Many shifts tions of various species, which means there
in individual species are already being ob- are ongoing and expected extinctions (Ur-
served. Observations of changes in ecosys- ban et al. 2016; Steffen et al. 2015).
tems and changes in interactions between In many ways, all four of the planetary
species are more complicated and there- boundaries that have been exceeded to date
fore more difficult to document, but they are related to climate change and biodiver-
are now being recorded in many regions. sity. The third exceeded boundary (after
Marine, terrestrial, and freshwater systems climate and biodiversity) is deforestation,
are all being affected, and the conservation a major contributor to both climate change
consequences of these biological changes and biodiversity loss (Steffen et al. 2015).
are of major significance. The final exceeded boundary is disruption
We explore the conservation and pol- of nutrient cycles, especially that of nitro-
icy responses to these challenges in Part V gen. One of the leading causes of nitrogen
(Chapters 22–28). Understanding climate release into the atmosphere is the burning
change biology can be a complex task for of fossil fuels, and species loss is one of the
policy makers and resource managers, par- major consequences of a disrupted nitro-
ticularly when the international response to gen cycle. Nitrogen overuse in agriculture,
climate change and biodiversity loss is seg- especially in developed countries, is lead-
regated into separate international conven- ing to nitrogen runoff impacts on freshwa-
tions and separate national policy-response ter systems and massive dead zones in the
CHANGING THE BIOSPHERE 5
oceans. Biodiversity loss and climate change etal goals. There are many dimensions to
are thus strongly linked to deforestation and adaptation, including maintenance of living
the disruption of nutrient cycling. space in societies challenged by sea-level
In this most global sense, biodiversity rise, incomes for individuals and families
loss and climate change are intimately con- facing adverse impacts of climate change,
nected and affect all people on the planet. It and ecological functioning and services.
is important that we find solutions to these These dimensions are all often addressed
two great environmental problems that in isolation—for instance, through engi-
are commensurate with the global scale of neering responses such as building dams to
the problems and that integrated solutions respond to water shortage—but there are
are pursued, not solutions in isolation. Al- major advantages to addressing the needs
though local and regional solutions are crit- of people and nature in a more integrated
ical for affected populations, true success fashion.
in conserving biodiversity and combating Wherever they live, people have a rela-
climate change can be achieved only when tionship with nature. That relationship may
the results are measurable at a global scale. be close and intimate, as in communities
Extinction rates need to be lowered so that that grow or harvest their own food, or it
they are commensurate with background may be remote and less obvious, as in cit-
rates, and atmospheric CO2 concentrations ies or highly structured economies. What-
need to be rapidly stabilized and probably ever the relationship, people and nature are
gradually decreased thereafter. closely tied together, which involves food
We will see in the chapters that follow (produced locally or grown elsewhere and
that many parts of the solution space in- then transported to where we consume it),
clude both climate and biodiversity ben- living space, recreation, spiritual connect-
efits. For example, reducing deforestation edness, climatic conditions, and clean air
maintains habitats and also reduces CO2 and water.
releases into the atmosphere, while lower- Climate change will alter the relationship
ing greenhouse gas emissions mitigates cli- between people and nature everywhere on
mate change, which reduces the magnitude the planet. People in rural settings may find
of species range shifts and thus the conflict that the food they can grow and where they
between species needs for movement and can grow it are changed, resulting in differ-
human land uses, such as agriculture, that ent patterns of agricultural use and natural
stand in the way of those movements. lands. People in cities may find that there
Although the climate-biodiversity nexus are higher costs for importing fuel for heat-
is deep and affects all people on the planet, ing or cooling, and that supply chains for
this is only the first and highest-order di- goods that are consumed have greatly dif-
mension of the need for integrated re- ferent environmental impacts.
sponse. Integrated responses are needed Nature provides both tangible and in-
that address the global dimension but also tangible benefits that accrue to people near
the relationship of people with nature in lo- and far in complex ways. Nature may have
cal situations, where ecosystems are an ef- benefits for people far away (e.g., sequestra-
fective tool to meet adaptation needs. tion of carbon for climate stabilization, bio-
diversity existence value) at the same time
that it has benefits for people close at hand
PE OPL E A N D NAT U R E A DA PTI NG (e.g., fresh water, tourism revenue) (Kre-
TO GE T H E R men et al. 2000).
This makes understanding the effect of
Adaptation is a response to climate change climate change on the relationship between
that seeks to maintain individual or soci- people and nature complex and important.
6 OV E RV I E W: W H AT I S C L I M AT E C H A N G E B I O L O G Y ?
The effects of climate change occur in an nature understand how the best elements
already-dynamic situation, in which hu- of a relationship with nature can be main-
mans and their activities are intercon- tained to the greatest extent possible and
nected locally, nationally, regionally, and where new relationships with nature will
internationally. Maintaining positive out- be needed because of climate change.
comes for people and ecosystems across all
of these spheres as the climate changes is a
major goal for people and nature adapting E CO SYST E M-BASE D A DA P TATION
together.
For example, communities living with One aspect of people and nature adapting
coral bleaching may need to adjust to both together is the use of ecosystems to help
reduced fisheries output and reduced op- meet human adaptation goals, or ecosystem-
portunities for tourism income. Although based adaptation (Jones et al. 2012). This is
the extent of coral bleaching suggests that a second example of the interaction of solu-
completely maintaining current levels of tions to both climate change and biodiver-
fishery output and tourism income may be sity loss. It is the special case of ecosystems
impossible, a new balance may be struck being a functional part of needed human
in which protecting reefs most able to sur- responses to climate change, with benefits
vive bleaching can help in the recovery of that are usually focused locally or regionally.
neighboring reefs and fisheries following Examples of ecosystem-based adapta-
bleaching while still maintaining tour- tion include coastal protection provided
ism revenues. Actions to create habitats for by mangroves and coral reefs, freshwater
remaining coral reef fish species may be provisioning provided by cloud forests,
important in maintaining fisheries produc- flood control provided by upland habitats,
tion. Life with climate change may look and drought resistance provided by healthy
considerably different for a community that range and natural wetlands. The scope of
is dependent on a coral reef. People’s rela- ecosystem-based adaptation literally covers
tionship with nature may change, through all ecosystems in all regions of the world,
changes in fisheries livelihoods and through even though it is only one small part of peo-
reduced reef extent. But a new relationship ple and nature adapting together.
may maintain some reef fishery, some reef Whereas the mechanisms of ecosystem-
tourism, and a coastal lifestyle with sound based adaptation are direct, the manage-
planning, whereas an unplanned response ment actions needed to maintain or improve
may result in total loss of the reef and its it may be complex (Vignola et al. 2009). For
tangible and intangible benefits. instance, offshore coral reefs provide pro-
Similar scenarios are played out across tection from rising sea levels and increas-
other coastal environments, rangelands, ing storm surges along many coastlines in
forests, lakes, and rivers. For instance, ar- the world. But coral bleaching also affects
eas denuded of mature trees due to climate- these reefs. Reef recovery after bleaching is
driven bark-beetle outbreaks may see tim- critical to the continued provision of coastal
ber revenues, recreational opportunities, protection services. After bleaching, reef re-
and habitat quality decline. A fundamental covery is a race between recolonizing cor-
restructuring of the relationship between als and the algae that grow on the dead reef
people and nature may be required on mul- (McClanahan et al. 2001). Healthy popula-
tiple levels in such settings. tions of fish that eat the algae are required to
Understanding the relationship between help the corals win the race and outcompete
people and nature and how it is affected algae in the battle to recolonize the dead
by climate change can help policy makers, reef. Large parrotfish are among the most
managers, and communities living with effective consumers of algae, but they are
CHANGING THE BIOSPHERE 7
also the targets of fishing. Helping protect together comes from the global race to
enough parrotfish to let corals recolonize protect the world’s last remaining natural
the reef maintains coastal protection and in habitats. As global population increases, the
the long run improves the overall fishery. footprint needed to feed the planet expands.
This example illustrates that sometimes As a result, the amount of natural habitat
there are trade-offs inherent in achieving remaining in the world is progressively dis-
ecosystem-based adaptation benefits. In appearing. The world’s population roughly
this case, a short-term reduction in fishing doubled between 1890 and 1950, doubled
is required to maintain both the long-term again by 1970 and again by 2010 (Gonzalo
health of the overall fishery and coastal pro- and Alfonseca 2016). The space required for
tection (McClanahan et al. 2001). Establish- people and their material needs is rapidly
ing mechanisms for benefits transfer across squeezing out space for natural habitats ev-
time and between beneficiaries can there- erywhere on the planet.
fore be critical to the success of ecosystem- At the same time, the world is moving
based adaptation. to protect natural habitats. The area of na-
Ecosystem-based adaptation approaches tional parks and other protected areas on
are now being combined with engineering the planet has doubled approximately ev-
approaches in “green-gray adaptation.” This ery 40 years, until now about 17 percent
combination allows for the delivery of ben- of the terrestrial surface of the planet is in
efits in the near and long terms, improving some form of protection (Costelloe et al.
utility through time and across generations 2016). Although this protection ensures the
(Bierbaum et al. 2013). An example might existence of remaining natural habitats, it
be the combination of upland forest pro- also reduces the amount of natural habi-
tection with a downstream flood-retention tat available for protection. So the amount
dam. The forest provides flood ameliora- of remaining natural habitat available for
tion benefits for small and medium-sized protection is a function of decreases due to
storms, whereas the retention dam provides habitat destruction to provide for agricul-
added capacity to cope with large storms. ture and other human needs and increases
In this setting, ecosystem-based adaptation in land that is already protected.
can provide part of the adaptation solution Climate change and biodiversity come
while capturing many biodiversity benefits. together in this race to protect remain-
Although ecosystem-based adaptation is ing natural habitat, because these habitats
just one dimension of people and nature are critical for the survival of species at
adapting together, it is a critically impor- the same time that they are critical stores
tant part of that spectrum, with the most of carbon for climate stabilization. When
obvious benefits for people who live in forests or other ecosystems are destroyed
close proximity to ecosystems. As ecosys- for conversion to agriculture or cities, the
tem-based adaptation matures, the benefits carbon contained in the vegetation often
of capitalizing on the nexus of biodiversity enters the atmosphere during burning as-
and climate change will become more ap- sociated with land clearing. Burning con-
parent, and the extension of this principle verts the carbon in vegetation to CO2 in
to broader elements of people and nature the atmosphere, a process that contributes
adapting together will emerge. about one-third of the greenhouse gases
entering the atmosphere each year. So by
protecting natural habitats from destruc-
RU N N I NG OU T OF TIM E tion, we can both avoid the extinction of
species living in those habitats and make a
A third example of mutual benefits from substantial contribution to constraining cli-
addressing biodiversity and climate change mate change.
8 OV E RV I E W: W H AT I S C L I M AT E C H A N G E B I O L O G Y ?
The time remaining to capture these dual tem-based adaptation, and the limited time
benefits is limited, however. Based on cur- for action are all examples of the challenges
rent rates of habitat loss to human uses and and synergies operating at the nexus of cli-
protection of habitat, the estimates are that mate change and biodiversity. Capitalizing
by 2050 or shortly thereafter, remaining on opportunities and responding to those
unprotected primary tropical habitat on the challenges involve science, awareness, pol-
planet will be gone (Hannah et al. 2018). icy, and management action. Action on each
Including secondary, recovering habitats of these fronts is needed to realize a world
extends the amount of time remaining to in which the relationship between people
protect habitat on the planet by 20 to 30 and nature remains positive.
years, but by the end of this century the Science creates the understanding of
window of opportunity for new protection biodiversity-climate relationships needed
will be closed in both primary and second- to generate awareness, policy, and manage-
ary habitats. ment action. The first sections of this book
After opportunities to protect large, land- are devoted to the insights that science has
sparing protected areas have closed, there already generated, from current observa-
will be opportunities to protect smaller tions to understanding of the past, and how
parcels of natural or seminatural habitat in those past and present insights can be ap-
productive landscapes. These land-sharing plied as we manage change. The chapters of
approaches are an important component the later sections look at how we can under-
in preserving biodiversity and responding stand and manage the future.
to climate change, as is land restoration. Many of our existing insights come from
Restoration of ecosystems (Chapter 25) is systems for which data are fortunately avail-
an important way to sequester carbon and able. For instance, our understandings of
provide more habitat for biodiversity. But as climate change responses of birds and but-
human populations continue to grow and terflies are particularly well understood be-
resource use intensifies, increasing pres- cause these taxa are particularly well stud-
sure will be put on productive landscapes ied, allowing for distributional responses to
and degraded lands. The window for these climate change to be tested. Despite a large
types of win-win activities for biodiversity number of biological responses to climate,
and climate change is closing as well un- however, these cases are not necessarily the
less we change our approach to planetary most important ones, either for people or
management. for biodiversity. There is an urgent need
Across conservation mechanisms, the to generate information about particularly
scope for constructive action to have bio- valuable systems and to generate policy and
diversity conservation and climate change management action based on those findings
mitigation and adaptation work together (see Box 1.1). For example, how elephants
will narrow as the century progresses. The respond to climate change is a complex
greatest opportunity to identify manage- mixture of habitat change due to climate
ment actions that help prevent extinctions change, changing human land uses, and an
and biodiversity loss while helping people often-delicate balance in human-elephant
adapt to climate change and reduce green- relations. Illegal hunting for the ivory trade
house gas emissions is therefore right now. may be the major threat to elephants in the
near term, but assuming that the threat of
illegal trade can be addressed, in the long
AC T ION F OR A N I N T E GR AT E D F U T U R E term it is habitat access and connectivity that
are likely to determine the number of el-
Staying within planetary boundaries, peo- ephants that can be sustained on the planet.
ple and nature adapting together, ecosys- Scientific understanding of these and other
CHANGING THE BIOSPHERE 9
Ecosystem experiments can be of ma- climate change will also include changes
jor value in revealing potential climate in precipitation, another experiment
change impacts and mechanisms. With- has involved simulating such changes in
out them we would be left in the defen- grasslands in northern California (Suttle,
sive position of being able to understand Thomsen, and Power 2007).
effects only as they occur rather than an- Other relevant ecosystem experiments
ticipate them with preemptive policies as have endeavored to understand the im-
well as adaptation measures. pacts that higher CO2 levels might have
In many senses ecosystem experi- on ecosystems. Will there or will there
ments came of age with the Hubbard not be a “fertilization” effect, or if there
Brook experiment (Holmes and Likens is one, will it be short-lived because the
2016), although there had been impor- ecosystem will encounter some other
tant precursors, such as the Brookhaven limitation (e.g., a key nutrient)? There
Irradiated Forest and the Luquillo Forest have been a number of such experi-
of Puerto Rico (Odum 1970). ments and some are ongoing. One Oak
The first climate change experiment Ridge experiment showed a short-lived
was initiated by John Harte at the Rocky fertilization response (Norby et al. 2010),
Mountain Biological Laboratory, where whereas others have not. The new free-air
an alpine meadow has been, and con- concentration enrichment (FACE) experi-
tinues to be, artificially warmed (Harte ment in the central Amazon may reveal
and Shaw 1955), revealing changes in the a similar nutrient constraint (in that case
plant community and soil biota. Since from phosphorous).
complex systems are needed to help guide more at stake than just bringing along the
stewardship of the planet through an era of full complement of nature’s bounty with us
unprecedented climate change. as we move into the future. At stake is the
Awareness is needed to complement sci- relationship of all humans with nature, the
entific understanding, both about climate ability to capitalize on ecosystems to help
change itself and especially about how cli- us adapt to climate change, and the ability
mate change and biodiversity interact. Be- to move quickly to realize the remaining
yond accepting the obvious—that climate scope for solutions before it is too late.
change is real and is having real impacts Policy is needed at the interface of these
today—the general public and policy mak- two great environmental concerns, to avoid
ers alike need to be aware that the inter- having climate change biology and its
face of biodiversity and climate change is consequences for people fall between the
complex and has a vitally important role to cracks. Since 1992, the world has had in-
play in shaping people’s relationship with ternational conventions that deal with bio-
nature. The latter sections of this book ex- diversity and with climate change. These
plore what is needed in policy, awareness, international conventions hold separate
and management action. meetings and are often administered by
Climate change biology is not just about separate agencies within individual coun-
the extinction of polar bears, or even just tries. What has been lacking in the policy
about extinction. Although the elevated risk world has been an ability to work between
of extinction due to climate change is a and across the conventions. Mitigation and
major international concern, there is much adaptation of climate change have massive
10 OV E RV I E W: W H AT I S C L I M AT E C H A N G E B I O L O G Y ?
implications for avoiding extinctions and form policy and management actions that
conserving biodiversity, whereas main- maintain a healthy balance between people
taining biodiversity has major implications and nature as the climate changes (Part V).
for managing the planet’s atmosphere and The combined power of these insights is to
helping people adjust their relationships give humanity the opportunity to create a
with nature and use ecosystems to adapt to healthy relationship with nature of multiple
climate change. scales—from planetary boundaries to eco-
Now that fundamental progress has system-based adaptation—that will endure
been made on each issue in its own right, as the climate changes.
it is particularly critical that collaborative
work at the interface of climate change and
biodiversity occur. Just as in science at the R E FE R E NCE S
interface, policy at this interface is begin-
ning to emerge. For instance, the Green Cli- Bierbaum, Rosina, Joel B. Smith, Arthur Lee, Maria Blair,
Lynne Carter, F. Stuart Chapin, Paul Fleming, Susan
mate Fund recognizes ecosystems as one of
Ruffo, Missy Stults, Shannon McNeeley, Emily Wasley,
the four pillars of climate adaptation. and Laura Verduzco. 2012. “A comprehensive review
Managers, scientists, policy makers, and of climate adaptation in the United States: More than
the public are not separate teams that lob before, but less than needed.” Mitigation and Adaptation
questions or insights back and forth to one Strategies for Global Change 18 (3, October): 361–406.
Costelloe, Brendan, Ben Collen, E. J. Milner-Gulland, Ian
another. Rather, they need to form a single
D. Craigie, Louise McRae, Carlo Rondinini, and Em-
integrated team in which scientific under- ily Nicholson. 2016. “Global biodiversity indicators
standing and awareness drive policy that reflect the modeled impacts of protected area policy
supports sound management action. Because change.” Conservation Letters 9 (1, January): 14–20.
of this need for integration, it is critical that Gonzalo, Julio A., and Manuel Alfonseca. 2016. “World
population growth.” World Population: Past, Present & Future
the climate change and biodiversity com-
29.
munities work together for mutual under- Hannah, Lee, Patrick Roehrdanz, Guy Midgley, Jon
standing and for the evolution of awareness, Lovett, Pablo Marquet, Richard Corlett, Brian Enquist,
policy, and management that responds to the and Wendy Foden. “Last call for tropical protected ar-
full scope of the implications climate change eas.” In review.
Harte, John, and Rebecca Shaw. 1995. “Shifting domi-
carries for ecosystems and biodiversity.
nance within a montane vegetation community:
The role of climate change biology is to Results from a climate-warming experiment.” Science
provide an understanding of the responses 267: 876–880.
of the natural world to climate change. In- Holmes, Richard T., and Gene E. Likens. 2016. Hubbard
sights from changes already being observed Brook: The Story of a Forest Ecosystem. Yale University Press.
Jones, Holly P., David G. Hole, and Erika S. Zavaleta.
(Part II) put us on alert about possible dis-
2012. “Harnessing nature to help people adapt to cli-
ruptions to the balance in the relationship mate change.” Nature Climate Change 2 (7): 504.
between people and nature. Understand- Kremen, Claire, John O. Niles, M. G. Dalton, Gretchen
ing how species and ecosystems have re- C. Daily, Paul R. Ehrlich, John P. Fay, David Grewal,
sponded to past climate change, both in and R. Philip Guillery. 2000. “Economic incentives
for rain forest conservation across scales.” Science 288
deep time and in near time (Part III), pro-
(5472): 1828–1832.
vides a window onto how species and eco- McClanahan, T., N. Muthiga, and S. Mangi. 2001. “Coral
systems have responded in a fully natural and algal changes after the 1998 coral bleaching: In-
world, which we must now interpret, and teraction with reef management and herbivores on
landscapes increasingly dominated by hu- Kenyan reefs.” Coral Reefs 19 (4): 380–391.
Norby, Richard J., Jeffrey M. Warren, Colleen M. Iversen,
man land uses. Modeling and experimenta-
Belinda E. Medlyn, and Ross E. McMurtrie. 2010. “CO2
tion to estimate future change (Part IV) can enhancement of forest productivity constrained by
be combined with knowledge and models limited nitrogen availability.” Proceedings of the National
of changes in human production to help in- Academy of Sciences 107: 19368–19373.
CHANGING THE BIOSPHERE 11
Odum, Howard T. 1970. “Summary: An emerging view Suttle, K. B., Meredith E. Thomsen, and Mary E. Power.
of the ecological system at El Verde.” Tropical Rain Forest: 2007. “Species interactions reverse grassland re-
1191–1281. sponses to changing climate.” Science 315: 640–642.
Rockström, Johan, Will Steffen, Kevin Noone, Åsa Pers- Urban, M. C., G. Bocedi, A. P. Hendry, J. B. Mihoub,
son, F. Stuart Chapin III, Eric Lambin, Timothy Len- G. Pe’er, A. Singer, J. R. Bridle, L. G. Crozier, L. De
ton, et al. 2009. “Planetary boundaries: Exploring the Meester, W. Godsoe, and A. Gonzalez. 2016. “Im-
safe operating space for humanity.” Ecology and Society proving the forecast for biodiversity under climate
14 (2): art. 32. change.” Science 353 (6304): p.aad8466.
Steffen, Will, Katherine Richardson, Johan Rockström, Vignola, Raffaele, Bruno Locatelli, Celia Martinez, and
Sarah E. Cornell, Ingo Fetzer, Elena M. Bennett, Pablo Imbach. 2009. “Ecosystem-based adaptation to
Reinette Biggs, et al. 2015. “Planetary boundaries: climate change: What role for policy-makers, society
Guiding human development on a changing planet.” and scientists?” Mitigation and Adaptation Strategies for Global
Science 347 (6223): art. 1259855. Change 14 (8): 691.
CHAPTER TWO Climate is the set of weather conditions pre-
vailing over a region over a period of time.
What Is Climate To be long enough to average across the nat-
ural, short-term variations of these condi-
Change? tions, the period over which the climate is
defined has generally been three decades—
MICHAEL C. MACCR ACK EN roughly a human generation. By contrast,
weather is the instantaneous state of the at-
mosphere; it is the statistical ensemble of
the weather that is used to quantify the state
of the climate. As some say, climate is what
you expect and weather is what you get.
Weather conditions (and so climatic
conditions, being the accumulated set of
weather conditions) are largely determined
by the dynamic interactions of the atmo-
sphere, the oceans, the land surface, and
the glaciers and ice sheets, which are often
collectively referred to as the components
of the climate system. For each location,
climate generally represents the conditions
with which a region is associated, such as
southern California being warm and dry
and the Pacific Northwest coast being cool
and wet. However, scientifically, the cli-
mate of a region also includes the seasonal,
annual, and longer variations and cycling
(e.g., El Niño and La Niña oscillations) that
result from the various internal intercou-
plings and linkages of the components of
the climate system.
In addition to internal variability, the
climate can also be affected by factors con-
sidered external to the climate system. Ex-
ternal factors that are capable of causing
changes in the system’s energy fluxes and
balances (generally referred to as radiative
forcings) include changes in incoming so-
lar radiation due to changes in solar out-
put and/or time-varying changes in the
distance to the Sun, aerosol injections into
the stratosphere resulting from major and
12
W H AT IS CLIM ATE CH A NGE? 13
persistent volcanic eruptions, geological Among the most important positive (or
processes such as the shifting distribution amplifying) feedbacks to consider in these
of continents and mountain ranges, and analyses are water-vapor feedback (whereby
changes in composition of the atmosphere warming leads to an increase in atmo-
due to processes ranging from weather- spheric water vapor that then exerts fur-
ing to sequestration of carbon in swamps ther warming) and snow- and ice-albedo
or submergence of vegetation by sea-level (or reflectivity) feedbacks (whereby warm-
rise. Human activities of various types are ing leads to melting of snow and ice that
now, however, altering the Earth’s energy decreases surface reflectivity that, in turn,
balance in comparable ways. The largest allows for additional absorption of solar
warming influence is resulting from the radiation and so more warming). Counter-
increase in the CO2 concentration due to ing these amplifying processes are ones that
combustion of fossil fuels (i.e., coal, petro- moderate how much change can occur. For
leum, natural gas, oil shale-derived fuels) example, the infrared radiation emitted up-
and deforestation and soil disturbance. In ward from the surface is proportional to the
addition to CO2, human activities have been fourth power of the surface temperature, al-
increasing the atmospheric concentrations though this negative feedback is moderated
of methane (CH4), nitrous oxide (N2O), by a process referred to as the atmospheric
halocarbons (mainly chlorine containing greenhouse effect, whereby the atmosphere
compounds), and other gases that alter the absorbs much of the upward radiation from
Earth’s infrared (heat-related) gases (collec- the surface and lower atmosphere and then
tively, greenhouse gases). Human activities reradiates much of it (presently about 90
have also been adding to the atmospheric percent) back to the surface to induce fur-
loadings of dark aerosols like black carbon ther warming. Interacting together, these
(e.g., from diesel engine exhaust and coal- processes have created the range of climates
fired power plants) that absorb additional across the Earth to which we have become
sunlight and light-colored aerosols such as accustomed and on which society has be-
dust and sulfate (the latter created mainly come dependent.
from sulfur dioxide, SO2, emissions by
coal-fired power plants) that exert a cool-
ing influence by reflecting additional sun- L E S SONS F ROM E A RT H ’S C L IM ATIC
light back to space. Human activities have H ISTORY
also caused changes in land surface charac-
teristics such as deforestation, agriculture, Geological records, chemical and physi-
and urbanization, which can affect surface cal variations in ice cores, and many other
roughness and the rate of evaporation from types of evidence make it clear that the
soils. Earth’s climate has varied greatly over its
The responses of the climate system to history. Indeed, the most important les-
these forcings not only are direct, but the sons from the paleoclimatic record are that
direct responses also can trigger further in- significant climate change can occur, that
teractions and feedbacks among climate these changes occur as a result of identifi-
system components that can amplify or able changes in forcing rather than as a re-
moderate the direct response. Thus, work- sult of random fluctuations, and that large
ing out the net effects of these forcings on changes generally tend to result from accu-
temperature, precipitation, and the weather mulated changes over many millennia.
requires very careful quantitative analysis, An important example of the role of
especially in seeking to learn from what has climate forcings in altering the climate is
happened in the past and projecting for- evident in the glacial-interglacial cycling
ward what could happen in the future. that occurred over the past million years
14 OV E RV I E W: W H AT I S C L I M AT E C H A N G E B I O L O G Y ?
(Berger 2001). Roughly 125,000 years ago, but not experiencing extensive glaciation
the global average temperature during the over this period.
interglacial period before the present one During the Holocene, fluctuations in
was up to ~1ºC–2ºC higher than present, global average temperature have generally
and sea level is estimated to have been 4–8 been less than ~0.5ºC, driven primarily by
meters higher. This several-thousand-year extended periods of volcanic activity, small
warm interval was followed by transition changes in orbital forcing and solar activity,
into an increasingly glacial period that and changes in land cover and agriculture.
lasted ~100,000 years. At its coldest, the Although regional-scale fluctuations have
global average temperature was ~6ºC be- been somewhat larger, the relatively favor-
low present and sea level dropped by ~120 able global climatic conditions that have ex-
meters, exposing large areas of continen- isted well into the twentieth century have
tal shelves around the world, drying out allowed time for the maturing of ecosys-
the Mediterranean Sea, and creating a land tems and the development of agriculture,
bridge between North America and Asia. As communities, and civilizations around the
the global average temperature rose to its world.
present level about 6,000–8,000 years ago,
ice sheets melted and sea level rose at an av-
erage rate of ~1 meter per century for 120 H U M A N-I N DUC E D C L IM AT E C H A NGE
centuries. Large shifts in flora and fauna
occurred throughout this period, demon- The relative steadiness of the climate is now
strating that ecosystems on land and sea are being disrupted because of the changes in
closely tied to the prevailing climate, albeit atmospheric composition caused mainly
perhaps with a lag of centuries, and that by the use of coal, petroleum, and natural
such shifts can occur if climate change is gas to power industrialization and eco-
extended over relatively long periods. nomic growth, and alteration of the land-
These glacial-interglacial changes in the scape to allow for global agriculture. Fig-
climate were driven by cyclic changes in ure 2.1 shows the changes in the global
the Earth’s orbit around the Sun, which re- average surface air temperature from 1880
distributed incoming solar radiation by sea- to 2016 compiled by the National Oceano-
son and latitude in ways then amplified by a graphic and Atmospheric Administration.
wide range of mostly positive feedback pro- Detection-attribution analyses described
cesses, including a natural carbon feedback in IPCC’s Second Assessment Report made
that tended to amplify the orbitally induced it possible to initially conclude that “there
warming. As a result of quite fortuitous was a discernible human influence on the
timing of three important orbital periodici- global climate” (Santer et al. 1995). Analy-
ties (ellipticity of the orbit, tilt of the axis ses of over a dozen measures of climate
of rotation, and time of year of closest ap- change (e.g., changes in atmospheric mois-
proach to the Sun), seasonal and latitudinal ture; the height of the tropopause, which
patterns of incoming solar radiation have is the boundary between the troposphere
been relatively stable over the past several and stratosphere; regional patterns and tim-
thousand years, which is known as the Ho- ing of land and ocean warming) and more
locene epoch, and the astronomically deter- since that initial assessment now support
mined seasonal and global patterns of solar the much stronger conclusion that “it is ex-
radiation would be expected to vary little tremely likely that human influence has been
over the next several tens of thousands of the dominant cause of the observed warm-
years (Mysak 2008)—that is, absent human ing since the mid-20th century” (IPCC
activities, the world might be a bit cooler, 2013, emphasis in original). With human
W H AT IS CLIM ATE CH A NGE? 15
Figure 2.1. Time history (1880–2016) of the change in sonal temperatures swing) is now equal to
global average temperature over time based on the av- about 15 percent of the 6°C warming from
erage of geographically weighted local departures of
annual-average surface air temperature (ºC) from the
the Last Interglacial Maximum to the pres-
20th century baseline; this compilation is maintained ent and about the same fraction of the simi-
by the US National Oceanographic and Atmospheric lar difference between the temperatures of
Administration. The late 19th- and early 20th-century the present and the very warm Cretaceous,
values were depressed by a higher-than-average number when dinosaurs roamed the Earth.
of significant volcanic eruptions; the mid-20th century
was roughly level as greenhouse gas–induced warming
Although international attention has fo-
was offset by an increased loading of sulfate aerosols cused on the change in the global average
and slowly declining solar irradiance; warming since temperature, changes in climate involve
the 1970s is mainly due to the rising concentrations of much more than simply a change in the
greenhouse gases. (Figure courtesy of NOAA.) average temperature (for details, see IPCC
2013; Cramer and Yohe 2014; for the United
forcing now emerging as the dominant States, see USGCRP 2017). Indeed, what
cause of climatic change, the period since people, plants, and wildlife experience is
roughly the mid-twentieth century is now not the mathematical construct that makes
being designated the Anthropocene epoch up the three-decade average; it is changes
(Steffen et al. 2011). in the intensity, occurrence, and patterns of
According to the most recent data, the the weather. As an example, in addition to
global average surface air temperature has recent changes in the average temperature,
increased nearly 1ºC since the mid-nine- the likelihood of warm and cool summers
teenth century (Plate 1), and suggestions over Northern Hemisphere land areas has
of a break in the rate of warming during been changing (Figure 2.2). Quite amaz-
the first decade of the twenty-first century ingly, in response to global warming of only
have proved mistaken (Santer et al. 2014). ~0.5ºC since the mid-twentieth century,
An increase of 1ºC may seem small in con- the likelihood of the very warmest sum-
trast to diurnal and seasonal swings in local mers occurring has increased by more than
temperature on the order of 20ºC in some three standard deviations from a likelihood
regions, but the global warming to date (so of less than 0.1 percent during the period
the baseline around which diurnal and sea- 1951–1980 to currently ~10 percent of the
16 OV E RV I E W: W H AT I S C L I M AT E C H A N G E B I O L O G Y ?
Figure 2.2. The frequency of occurrence of local de- variable that trend statistics are not well es-
partures of June-July-August (summertime) surface tablished, but oceans are warmer, and there
air temperature (in standard deviations from the local
mean value) for land areas in the Northern Hemisphere
appears to be a greater likelihood that the
relative to the mean value for 1951–1980 (generally de- storms that do develop will have higher
scribed as the seasonal climatological value). The distri- peak winds and greater precipitation.
bution of frequencies of occurrence for the period 1951– Changes in mid-latitude weather patterns
1980 closely matches a normal (i.e., bell-shaped curve) illustrate how warming in high latitudes can
distribution (solid line), which is used to subdivide the
set into cool (dark gray), typical (white), and warm
drive shifts in the climate (e.g., Francis and
(light gray) occurrences, each subset having a probabil- Vavrus 2012; Francis and Skific 2015). The
ity of a third. The distribution of local summertime de- variable weather in Northern Hemisphere
partures from the 1951–1980 normal has shifted to the mid-latitudes is created by cold, dense Arc-
right as the three-decade climatology has moved closer tic air masses spreading southward and
to the present. What were once cool summers now oc-
cur much less frequently, and what were once warm and
interacting with warm, moist subtropical
extremely warm summers now occur much more fre- air masses pushing poleward. Convection
quently. (Redrawn from Hansen et al. 2012.) and storms occur where moist air is forced
to rise over colder air. When the equator-
time—so an increase in incidence of such pole temperature gradient is strong, the
extremes by over a factor of a hundred in atmospheric circulation sets up to most ef-
only about six decades (Hansen et al. 2012). ficiently move heat poleward and the Co-
Other aspects of human-induced change riolis effect (i.e., the turning eastward of
are also discernible. Broadly speaking, meridional, or south to north, winds result-
warming in the Arctic has been at least ing from the rotation of the Earth) leads to
twice as large as the global average, causing there being a stronger and more zonal (west
widespread impacts on ecosystems (Arctic to east) jet stream with fewer meanders and
Climate Impact Assessment 2004). Also, the faster west-to-east storm movement. With
dry subtropics have expanded poleward Arctic warming becoming strong and tropi-
and average precipitation has increased over cal warming limited by increased evapora-
mid- and high-latitude land areas. The like- tion, the equator-pole temperature gradient
lihood of intense rainfall has also increased, is decreasing, which is leading to greater
with a greater fraction of precipitation com- meandering of the zonal jet. The larger
ing in such events, causing increased runoff waves in the jet stream allow warm moist
and in turn causing higher river levels and air to push farther north and cold, Arctic air
increased potential for flooding at those to push farther south, increasing the range
times of year when stream flows are high. of possible conditions at any given location
With less precipitation occurring in small and slowing the eastward movement of wet
rain events, periods of high evaporation of and dry weather regimes.
soil moisture are now longer and warmer, For example, in the Great Plains of North
leading to faster and more frequent tran- America (the only continent extending
sitions to very dry and even drought con- from the subtropics to the Arctic without
ditions (USGCRP 2017). The occurrence an east-west mountain range to separate
of hurricanes and typhoons is so low and cold polar and warm, moist subtropical air
W H AT IS CLIM ATE CH A NGE? 17
masses), the shifting weather conditions empirical understanding of how the cli-
allow for warm, more northward penetra- mate system functions and how all of the
tion of moist subtropical air, causing, in system variables and processes interact and
some years, heavier snowpacks and so in- link together to determine the climate that
creased springtime snowmelt runoff in the prevails. Thus, the models follow how so-
northern Great Plains. With its relatively lar radiation changes through the day and
flat geography and a history of less intense season and is reflected and absorbed by the
weather events, river channels have tended atmosphere, clouds, land, ice, and oceans.
to be shallow. Now, with rapid melting of The models also (1) simulate the flow of in-
the heavier snow amounts, major flood- frared (heat) radiation and its dependence
ing has become more frequent and severe. on temperature and atmospheric composi-
Thus, not only is the climate changing, but tion; (2) calculate the motions of the atmo-
the hydrogeography and land cover of re- sphere and oceans, both vertically and hori-
gions are also being affected by the chang- zontally; (3) treat the evaporation of water
ing weather. from the surface, the formation of clouds
and precipitation, and then accumulation of
soil moisture and loss by runoff; (4) treat
PROJ E CTIONS OF GLOBA L the buildup and loss of snow cover on land
WA R M I NG T H ROUGH T H E and the expansion and melting of sea ice on
T W E N T Y-F IR ST C E N T U RY the oceans; and more (see Figure 2.3). For
the atmospheric component of the system,
With global use of coal, petroleum, and the model is virtually the same as weather
natural gas still providing about 80 per- forecast models, but for the oceans, ice, and
cent of the world’s energy, the atmospheric land surface, the models are constructed
concentration of CO2 will continue to in- to include processes that affect cumula-
crease for some time, even with aggressive tive changes through seasons, years, de-
attempts to switch to deriving energy from cades, and longer. Given the influence of
renewable sources. Stopping the ongoing small-scale chaotic behavior, the nonlinear
increase in radiative forcing will require governing equations that are the basis of
CO2 and other greenhouse gas emissions weather forecast models limit the potential
to decrease to near zero (so not just stop- predictability of specific large-scale weather
ping the growth in emissions); returning to situations to only 1–2 weeks. In contrast,
the preindustrial level of radiative forcing model simulations for past multidecadal to
would require not only this but also essen- multicentennial periods (e.g., the past two
tially the removal of roughly as much CO2 centuries) and distant periods in the past
from the atmosphere as has been added. have demonstrated that the climate system
That would include the amount that has models, with forcings prescribed as they
been taken up by the oceans and biosphere, changed over these periods, reasonably re-
because much of this carbon would return produce the climatic conditions that have
to the atmosphere as the atmospheric con- been reconstructed from observations and
centration was lowered from its current geological evidence.
level. In seeking to project how the climate is
Computer models of the climate sys- likely to evolve in the future, scenarios of
tem have been developed to project how the consequences of future societal choices
the climate will respond to the changing in terms of energy-generated emissions and
atmospheric concentrations of CO2 and other factors are used as boundary condi-
other climate-altering gases and aerosols tions for simulations using the state-of-the-
(i.e., particles). These models have been science climate models. Model results then
built to incorporate both theoretical and portray how the climate would be expected
18 OV E RV I E W: W H AT I S C L I M AT E C H A N G E B I O L O G Y ?
Figure 2.3. Schematic view of the components of the IPCC assessments for a description of global
global climate system (bold), their processes and inter- changes and impacts and the most recent
actions (thin arrows), and some aspects that may change
(bold arrows). Presently available models of the atmo-
US and World Bank assessments for more
sphere, oceans, and land, as well as snow and ice sur- detailed descriptions in particular regions
faces, generally divide each domain into quadrilateral (Melillo et al. 2014; USGCRP 2017; World
tiles that are about 100 km to 200 km on a side and sliced Bank 2013, 2014).
thin enough to represent vertical variations in each sys- As shown in Figure 2.4, the global-scale
tem. Variables such as temperature, water content, and
speed of movement are maintained for each tile in each
results indicate that the greater the emis-
climate system component. Flows and fluxes through the sions, the greater the global warming that
boundaries of every tile and across from one component will occur. These simulations generally
of the system to another are calculated for each process, indicate that a prolonged reliance on fos-
then used to update the values of each variable in each sil fuels would lead to the global average
tile to a new time using a time step typically measured
in minutes. With such a fine time step, climate system
temperature reaching 4ºC above its prein-
models are simulating the evolving “weather” of the dustrial level by 2100 and continuing up-
atmosphere-ocean-land-ice system and the results are ward thereafter, whereas a very aggressive
then time- and space-averaged to provide an indication shift to reliance on solar, wind, and other
of how the climate is changing in response to natural renewable energy technologies has the po-
internal fluctuations and the changing values of external
forcings. (Figure 1.1 from IPCC 2001: Climate Change 2001:
tential to limit the increase in the global
The Scientific Basis. Contribution of Working Group I to the Third average temperature peaking at 2ºC–3ºC,
Assessment Report of the Intergovernmental Panel on Climate Change.) depending on how rapidly the conversion
of the energy system occurs.
Emissions of CO2 and other greenhouse
to evolve over coming decades and centu- gases not only cause climate change; many
ries. Because the model results are condi- of these gases also contribute to poor air
tional, being dependent on the scenarios, quality, widespread health effects (particu-
these results are called projections rather larly to women and children as a result of
than predictions. For further details, see the emissions from traditional cooking stoves),
Figure 2.4. Global projections of changes in the likeli- in likelihood of what were (B) the wettest consecutive
hood of important occurrences of moderately extreme five days and (D) the wettest 5% of days during the
weather. Graphs on the left show the time history from period 1986–2005. The projections are based on sce-
1960 to the present (based on observations) and to 2090 narios that remain mainly fossil fuel based (black line,
(based on projections) of the change in likelihood of RCP8.5), transition slowly away from fossil fuels (light
what were the days with daily maximum temperature gray, RCP4.5), and transition rapidly to energy technolo-
among the (A) coldest and (C) warmest 10% of days dur- gies that do not emit CO2 and other greenhouse gases
ing the period 1961–1990. Graphs on the right show the (dark gray). (TFE.9, Figure 1 from Stocker et al. 2013;
time history from 1960 to present (based on observa- IPCC Fifth Assessment Report, Working Group I, Techni-
tions) and to 2090 (based on projections) of the change cal Summary.)
20 OV E RV I E W: W H AT I S C L I M AT E C H A N G E B I O L O G Y ?
crop damage, water pollution, glacial melt- can also reduce near-term warming, and the
ing (due to warming and deposited black United States and China have taken the lead
carbon) and more. There are thus many rea- in doing so. There is thus the potential for
sons to limit climate-changing emissions. nations to make even greater and earlier cuts
Negotiations to cut emissions of CO2 in climate-warming emissions, including
and other long-lived greenhouse gases especially emissions of methane, precursors
have been the focus of international nego- of tropospheric ozone, and black carbon,
tiations through the annual Conference of but there is really very little time for such
the Parties (COP) since the UN Framework actions to be taken because of the warming
Convention on Climate Change (UNFCCC) trajectory already created by past emissions.
was negotiated in 1992, with the most re- As serious as the overall warming would
cent agreement being negotiated in Paris be, the unprecedented pace of the warm-
in 2015. Based on the emissions-reduction ing is at least as serious because it shortens
commitments made in support of the Paris the time for forests and other ecosystems to
Agreement’s goal, the global average tem- keep up with the shifting of their preferred
perature is nonetheless still projected to be climatic zones. In addition to the rapid pace
between 3ºC and 4ºC above its preindustrial of warming, the accelerating pace of long-
level, which is well above the agreement’s term sea-level rise and the lowering of the
aspirational goal of 1.5ºC and 2ºC; scientific pH of ocean waters (i.e., acidification) as
studies suggest that preventing most of the the CO2 concentration rises are projected
adverse impacts will actually require re- to cause increasing inundation of coastlines
turning the increase in global average tem- and spreading disruption of marine ecosys-
perature to no more than 0.5ºC (equivalent tems, respectively.
to a CO2 concentration of 300–350 ppm;
e.g., see Hansen et al. 2016).
So, what more can be done? Greatly im- BE YON D T H E AV E R AGE
proving the efficiency in our use of energy
(which would result in also reducing emis- The focus of international climate change
sions of black carbon) and transitioning negotiations has primarily been on the pro-
to renewable energy sources (solar, wind, jected decade-to-decade changes in global
wave, tidal, water current, etc.) in order to average temperature, but the global aver-
reduce CO2 emissions are the most environ- age is really only a mathematical construct.
mentally benign ways of reducing human- Understandably, the multidecadal average
induced impacts on the climate. Reversing change will vary by latitude, season, and lo-
deforestation, even implementing affores- cal conditions, being greater in high than
tation, and rebuilding soil carbon will also in low latitudes, greater over land than over
be essential to limiting the rise in the atmo- the ocean, and greater at night than dur-
spheric CO2 concentration and later pulling ing the day as a result of greenhouse gases
CO2 back out of the atmosphere. Because absorbing outgoing longwave radiation re-
the atmospheric lifetime of methane is only radiating back toward the surface twenty-
about a decade, as compared to the multi- four hours a day. In the low latitudes, the
millennial persistence of the human-in- temperature increase is smaller than in
duced increase in the CO2 concentration, re- high latitudes because a greater fraction of
ducing leakage of methane from the global the available energy results in evaporation
energy system and its generation by global (and evapotranspiration). With the vertical
agriculture can help slow the pace of climate distribution of CO2-induced radiative forc-
change over the coming few decades (UNEP ing exerting a slight tendency to stabilize
2011; Shindell et al. 2012). Reducing emis- the troposphere, additional energy from
sions of the many halocarbon compounds condensation is needed to power convective
W H AT IS CLIM ATE CH A NGE? 21
storms, which leads to the increasing occur- coming drier. Considering detailed changes
rence of intense precipitation. At the same in the weather is thus going to be essen-
time, warmer oceans are making addi- tial to understanding many biological re-
tional energy available to tropical cyclones sponses. Useful analyses will need to evalu-
(known as typhoons or hurricanes, depend- ate changes in the likelihood of various
ing on the region), powering an increase in weather types, especially including changes
storm intensity and so in damage, which is in the frequency, intensity, and duration of
then further compounded along vulnerable extreme conditions (e.g., heat waves, epi-
coastlines because of sea-level rise. sodes of intense precipitation and drought).
Individual organisms and ecosystems Figure 2.5 shows projections for the twenty-
experience the weather at the particular first century of changes in the frequency of
location that they occupy, making changes what were low-likelihood weather events
in the local weather more relevant than during the mid- to late twentieth century,
changes in long-term average conditions. showing a sharp increase in the occurrence
For example, in areas with variable terrain, of hot days and wet extremes and a sharp
intense rains may cause some areas to flood decrease in the occurrence of what were
while large-scale, flatter areas end up be- considered extremely cold days. The most
recent USGCRP assessment includes sub-
stantial new projections of changes in the
weather (USGCRP 2017).
Baede, A. P. M., E. Ahlonsou, Y. Ding, and D. Schimel. States: The Third National Climate Assessment. U.S. Global
2014. “The climate system: An overview.” In Climate Change Research Program.
Change 2001: The Scientific Basis. Contribution of Working Mysak, Lawrence A. 2008. “Glacial inceptions: Past
Group I to the Third Assessment Report of the Intergovernmental and future.” Atmosphere-Ocean 46: 317–341. https://doi
Panel on Climate Change, ed. J. T. Houghton, Y. Ding, D. .org/10.3137/ao.460303.
J. Griggs, M. Noguer, P. J. van der Linden, X. Dai, K. Santer, Benjamin D., Susan Solomon, Celine Bonfils,
Maskell, and C. A. Johnson, 85–98. Cambridge Uni- Mark D. Zelinka, Jeffrey F. Painter, Francisco Beltran,
versity Press. John C. Fyfe, et al. 2014. “Observed multi-variable
Berger, Andre. 2001. “The role of CO2, sea-level, and signals of late 20th and early 21st century volca-
vegetation during the Milankovitch forced glacial- nic activity.” Geophysical Research Letters 42. https://doi
interglacial cycles.” Geosphere-Biosphere Interactions and Cli- .org/10.1002/2014GL062366.
mate. Cambridge University Press. Santer, Benjamin D., Tom M. L. Wigley, Tim P. Bar-
Cramer, Wolfgang, and Gary W. Yohe. 2014. “Detection nett, and Ebby Anyamba. 1995. “Detection of climate
and attribution of observed impacts.” In Climate Change change and attribution of causes.” In Climate Change
2014: Impacts, Adaptation, and Vulnerability: Part A: Global and 1995: The Science of Climate Change, Intergovernmental Panel on
Sectoral Aspects, Working Group II Contribution to the Fifth As- Climate Change (IPCC), 407–443. Cambridge University
sessment Report of the Intergovernmental Panel on Climate Change, Press.
ed. C. B. Field, V. R. Barros, D. J. Dokken, K. J. Mach, Shindell, Drew, Johan C. I. Kuylenstierna, Elisabetta Vi-
M. D. Mastrandrea, T. E. Bilir, M. Chatterjee, et al., gnati, Rita van Dingenen, Markus Amann, Zbigniew
979–1037. Cambridge University Press. Klimont, Susan C. Anenberg, et al. 2012. “Simultane-
Field, C. B., V. R. Barros, K. J. Mach, M. D. Mastrandrea, ously mitigating near-term climate change and im-
M. van Aalst, W. N. Adger, D. J. Arent, et al. 2014. proving human health and food security.” Science 13:
“Technical summary.” In Climate Change 2014: Impacts, 183–189. https://doi.org/10.1126/science.1210026.
Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Steffen, Will, Jacques Grinevald, Paul Crutzen, and John
Contribution of Working Group II to the Fifth Assessment Report McNeill. 2011. “The Anthropocene: Conceptual and
of the Intergovernmental Panel on Climate Change, ed. C. B. historical perspectives.” Philosophical Transactions of the
Field, V. R. Barros, D. J. Dokken, K. J. Mach, M. D. Royal Society A 369: 842–867. https://doi.org/10.1098/
Mastrandrea, T. E. Bilir, M. Chatterjee, et al., 33–94. rsta.2010.0327.
Cambridge University Press. Stocker, T. F., D. Qin, G.-K. Plattner, L. V. Alexander, S. K.
Francis, Jennifer A., and Stephen J. Vavrus. 2012. “Evi- Allen, N. L. Bindoff, F.-M. Bréon, et al. 2013. “Techni-
dence linking Arctic amplification to extreme weather cal summary.” In Climate Change 2013: The Physical Science
in mid-latitudes.” Geophysical Research Letters (39). https:// Basis. Contribution of Working Group I to the Fifth Assessment
doi.org/10.1029/2012GL051000. Report of the Intergovernmental Panel on Climate Change, ed. T. F.
Francis, Jennifer A., and Natasa Skific. 2015. “Evidence Stocker, D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen,
linking rapid Arctic warming to mid-latitude weather J. Boschung, A. Nauels, Y. Xia, V. Bex, and P. M. Midg-
patterns.” Philosophical Transactions of the Royal Society A 373. ley, 31–115. Cambridge University Press.
https://doi.org/10.1098/rsta.2014.0170. UN Environment Programme (UNEP) and World Me-
Hansen, James, Makiko Sato, and Reto Ruedy. 2012. “Per- teorological Organization (WMO). 2011. Integrated As-
ception of climate change.” Proceedings of the National Acad- sessment of Black Carbon and Tropospheric Ozone. D. Shindell,
emy of Sciences 109. https://doi.org/10.1073/1205276109. chair. Nairobi: UNEP.
Hansen, James, Makiko Sato, Paul Hearty, Reto Reudy, et USGCRP. 2017. Climate Science Special Report: Fourth National
al. 2016. “Ice melt, sea level rise and superstorms: Evi- Climate Assessment, Volume I. Edited by D. J. Wuebbles, D.
dence from paleoclimate data, climate modeling, and W. Fahey, K. A. Hibbard, D. J. Dokken, B. C. Stewart,
modern observations that 2ºC warming could be dan- and T. K. Maycock. U.S. Global Change Research Pro-
gerous.” Atmospheric Chemistry and Physics 16: 3761–3812. gram. https://doi.org/10.7930/J0J964J6.
https://doi.org10.5194/acp-16-3761-2016. World Bank Group. 2013. “Turn down the heat: Climate
Intergovernmental Panel on Climate Change. 2013. Cli- extremes, regional impacts, and the case for resil-
mate Change 2013: The Physical Science. Cambridge Univer- ience.” Washington, DC: World Bank.
sity Press. World Bank Group. 2014. “Turn down the heat: Con-
Melillo, Jerry M., T. C. Richmond, and Gary W. Yohe, fronting the new climate normal.” Washington, DC:
eds. 2014. Highlights of Climate Change Impacts in the United World Bank.
PA RT I I
25
26 W H AT C H A NGE S A R E W E OB SE RV I NG?
Paleo data: document associa- Over the past 490 my, coral reef NA: Vernon (2008)
tions between historical cli- die-off coincided with in-
mate change and ecological creases in CO2, methane, and/
responses or warm temperatures
Experiments: document a Laboratory experiments show Laboratory, greenhouse, and field
significant role of climate in corals bleach under stresses experiments of temperature ma-
species’ biology such as warm temperatures, nipulations show small increases
extreme salinities, and high in temperature increase pheno-
rates of sedimentation: Lesser logical asynchrony between the
(1997), Jones et al. (1998), butterfly and its host plant, driv-
Glynn and D’Croz (1990), ing increased pre-diapause larval
Anthony et al. (2007) mortality and affecting extinc-
tion and colonization dynamics:
Weiss et al. (1988), Hellmann
(2002), Boughton (1999)
Long-term observations: Coral bleaching events consis- >50 years of regular (often yearly)
significant and consistent tently follow warm sea sur- censuses across the species’ range
associations between a cli- face temperature events (e.g., document multiple population
mate variable and a species’ El Niño): Hoegh-Guldberg declines and extinctions fol-
response (1999) lowing drought and increased
variability in precipitation: Singer
(1971), Singer and Ehrlich (1979),
Ehrlich et al. (1980), McLaughlin
et al. (2002a), McLaughlin et al.
(2002b), Singer and Parmesan
(2010). Also early snowmelt (false
springs) and unseasonal frost:
Thomas et al. (1996)
Fingerprints: responses that First observations of mass tropi- Highest levels of population extinc-
uniquely implicate climate cal coral bleaching in 1979, tions along the southern range
change as causal factor concurrent with accelerating boundary and lowest levels of
SST warming: Hoegh-Guld- population extinctions along
berg (1999) northern and high-elevation
range boundaries, uniquely con-
sistent with regional warming
and not with local or regional
habitat degradation and destruc-
tion, urbanization, agricultural
expansion, or plant invasions:
Parmesan (1996), Parmesan
(2003), Parmesan (2005). There
is a significant downward step in
both proportion of population
extinctions and snowpack trends
above 2,400 m in the Sierra Ne-
vada mountain range: Parmesan
(1996), Johnson et al. (1999)
(continued)
28 W H AT C H A NGE S A R E W E OB SE RV I NG?
Note: The stronger the evidence within each line, and the greater the number of different lines of evidence
that support a climate change interpretation, the greater is the confidence in attributing an observed
change in a particular population, species, or system to observed climate change. Effects of climate change
on coral reefs and on a butterfly provide examples of this approach to attribution. (Modified from Parme-
san et al. 2013.)
The ever-increasing number of indi- urchins, limpets) and across diverse ecosys-
vidual studies of climate change impacts tems (from temperate terrestrial grasslands
in natural systems has allowed the emer- to tropical cloud forest, and from coastal
gence of new studies that merge data from estuaries to open ocean). It appears that
many different publications into “meta- (unexpectedly) strong responses to cur-
analyses” able to detect common trends rent warming trends are swamping other,
when all species are brought together into potentially counteractive forces of global
a single analysis. There are now five major change. Thus, the more visible aspects of
global meta-analyses assessing changes in human impacts on species’ distributions,
distributions and abundances of individual such as habitat loss, are not masking the
species in recent decades (Table 3.2). These impacts of climate change.
analyses showed that about half of species Three meta-analyses estimated rate of
(for which long-term data exist) exhibited range boundary changes (Table 3.3). Ter-
significant changes in their distributions restrial meta-analyses differed somewhat,
over the past 20 to 140 years. These changes with Parmesan and Yohe (2003) estimat-
are not random but are systematically in the ing a 6.1 km/decade shift of poleward and
direction expected from regional changes upward range boundaries, and Chen et al.
in the climate. One of the major additions (2011) estimating a 19.2 km/decade shift of
to the literature since Parmesan (2005a) poleward range boundaries. The latter num-
are documentations of response to climate ber was recalculated by Parmesan from the
change in marine species at a global scale supplemental data file, taking only data from
(Poloczanska et al. 2013). the poleward range boundary and excluding
The most remarkable outcome is the data on shifts in centroids to make it com-
consistency of response across diverse spe- parable to Parmesan and Yohe’s analysis. But
cies and regions. Responses have now been these studies are also fundamentally differ-
documented in every taxonomic group for ent in that Chen et al. used the mean shift of
which long-term data exist (e.g., birds, but- all species for a given region as their individ-
terflies, trees, mountain flowers, fish, sea ual data points (a taxa region mean), whereas
Table 3.2. Global fingerprints of climate change impacts across wild species
Changes con-
sistent with
For those with local or regional
Species or stud- long-term climate change,
ies focused on distribution or % of all species
distribution or abundance data, that showed
Species in abundance data: % (n) of species change regard-
N: species and given system % (n) of total in showing signifi- less of type of
Study functional groups (%) study cant change change
Parmesan and 1,598 T: 85.2% 58% (n = 920) 50% (n = 84%a
Yohe 2003 M: 13.5% 460/920)
F: 1.3%
Root et al. 2003 1,468 T: 94% 58% (n = 926) 52% (n = 82.3%a
M: 5.4% 483/926)
F: 0.6%
Rosenzweig et 55 studies (~100– T: 65%b 33% (n = 18 — 90%c
al. 2008 200 species) M: 13%b studies)
F: 22%b
Poloczanska et 857 sp = 1,735 T: 0% 80% (n = 63% (279/446 83%a
al. 2013 sp × trait M: 100% 1,060/1,323 total for
combinationsd F: 0% total distribu- distributions
tions + abun- only)
dances)d
Note: For species in given system, T = terrestrial, M = marine, and F = freshwater. For changes consistent with
local or regional climate change, type of change includes phenological changes. For each data set, a response for
an individual species or functional group was classified as (1) no response (no significant change in the measured
trait over time), or (2) if a significant change was found, the response was classified as either consistent or not
consistent with expectations from local or regional climate trends. Percentages are approximate and estimated
for the studies as analyses across each of the studies may differ. The specific metrics of climate change analyzed
for associations with biological change vary somewhat across studies, but most use changes in local or regional
temperatures (e.g., mean monthly T or mean annual T), with some using precipitation metrics (e.g., total annual
rainfall). For example, a consistent response would be poleward range shifts in warming areas. Probability (P) of
getting the observed ratio of consistent to not consistent responses by chance was <10–13 for Parmesan and Yohe
2003; Root et al. 2003; Root et al. 2005; and Poloczanska et al. 2013; it was <0.001 for Rosenzweig et al. 2008.
Test were all binomial tests against p = 0.5, performed by Parmesan.
a
P < 0.01 × 10–13.
b
Individual species were analyzed by Rosenzweig et al., but data on species not provided in publication—percent-
ages shown are based on numbers of studies.
c
P < 0.001 (from binomial test against random expectation of 50%-50% chance of change in either direction—
either consistent or not consistent with local or regional climate change).
d
For a few species, data were available on multiple traits (e.g., both the leading and the trailing edges of a spe-
cies’ range), such that those species are represented more than once in the analysis. These are the minority, but
as numbers given here are based on species × trait combinations, the totals are greater than the total number of
species for this analysis.
30 W H AT C H A NGE S A R E W E OB SE RV I NG?
Parmesan and Yohe used each individual out the Northern Hemisphere (most from
species as data points. Thus the difference North America and Europe, but a scattering
in estimate of rates of shift may simply stem from Asia and Africa), whereas 76 percent
from differences in how data were com- of the species in Chen et al. were from two
piled. There are two more likely candidates quite northerly countries (the United King-
for these differences: first, Chen et al. (2011) dom and Finland), geographically situated in
included data from the most recent decade, regions that have experienced greater warm-
largely absent from Parmesan and Yohe ing than the global average.
(2003), and the 2000s was the hottest decade More notable is that estimated expan-
on record; hence, it might be expected that sion of leading edges derived from marine
range shifts would have accelerated during systems is several times greater than either
this record hot time period; and second, the terrestrial estimate: at ~72 km/decade (Po-
data from Parmesan and Yohe were distrib- loczanska et al. 2013). This may at first seem
uted more globally than those of Chen et al., counterintuitive, because warming has
with the former including studies through- been less than half as much in the oceans as
Table 3.3. Rates of change in distribution from meta-analyses of marine and terrestrial systems
Note: The number of studies and number of observations (taxonomic or functional groups) from studies are
given, together with a breakdown of studies by realm. The criteria for data inclusion are outlined for each study.
Seabirds, anadromous fish, and polar bears were counted as marine. Wading birds were counted as freshwater
birds. Minimum time span of observations within studies is 19 years, unless stated otherwise. Multispecies stud-
ies include two or more species unless stated otherwise. (Modified from Poloczanska et al. 2013.)
a
<4% of total number of observations from single-species studies.
b
<10% of total number of observations from single-species studies.
c
Recalculated using only leading-edge (cold limit) observations spanning >18 years, and with data after 1990.
RANGE AND ABUNDANCE CHANGE S 31
on land. But a new measure, called veloc- shifts are emerging. For example, recent
ity of climate change (VoCC; Loarie et al. large-scale analyses indicate an increasing
2009), seems to explain this discrepancy. scrub encroachment into Arctic tundra that
Technically, VoCC is a speed measured as is consistent with regional climate change
the ratio of the temporal gradient to the (Serreze et al. 2000; Dial et al. 2016).
spatial gradient, given a particular time
frame and emission scenario. So VoCC
measures the geographic shift of tempera- DI F F E R E N TI ATI NG DI AGNO STIC
ture isotherms through time as the climate PAT T E R NS
warms. A more intuitive way to understand
this is that VoCC is the speed required to Important diagnostic patterns, specific to
move across the surface of the Earth if you climate change impacts, helped implicate
are trying to maintain a constant tempera- global warming as the driver of the observed
ture under a given level of temperature changes in natural systems (defined in Par-
change. Temperatures in the ocean do not mesan and Yohe 2003). These patterns in-
change quickly with geographic distance clude differential responses of cold-adapted
(i.e., they have a shallow temperature gra- and warm-adapted species at the same loca-
dient), whereas on land temperatures can tion and the tracking of decadal tempera-
change quite quickly over space, especially ture swings. For the latter, long time series
in mountainous regions. So, for example, if are essential (>70 years). A typical pattern
the ocean and land both warm by an equal observed in Britain, Sweden, and Finland
amount, a fish must travel much farther was northward shift of the northern range
poleward to maintain the same tempera- boundaries of birds and butterflies during
ture than would a small mammal on the two twentieth-century warming periods
nearby coast (at the same latitude). (1930–1945 and 1975–1999), and southward
The absolute rate of temperature rise has shifts during the intervening cooling period
been about three times greater on land than (1950–1970). There were no instances of the
in the oceans (0.24°C/decade versus 0.7°C/ opposite pattern (i.e., boundaries shifting
decade, respectively; Burrows et al. 2011), southward in warm decades and northward
but VoCC gives a different picture. Recent in cool decades). In total, such diagnostic
global analyses of VoCC estimates are simi- “sign-switching” responses were observed
lar between oceans and land, especially in 294 species by Parmesan and Yohe (2003)
when comparing latitudes where both land and in 24 species by Poloczanska et al. (2013)
and ocean are found (i.e., between 50°S spread across the globe and ranging from
and 80°N), with oceanic isotherms shifting oceanic fish to tropical birds to European
at a rate of 27.5 km/decade and terrestrial butterflies. These specific patterns of biologi-
isotherms shifting at 27.4 km/decade (Bur- cal trends are uniquely predicted by climate
rows et al. 2011). Further, although absolute trends—no other known driver would cause
temperature change and VoCC are both these specific patterns of response to emerge.
stronger at the highest latitudes, VoCC is un-
like absolute temperature change in being
strong in the tropics, particularly in tropical OBSE RV E D C H A NGE S I N I N DI V IDUA L
oceans (Burrows et al. 2011). Not surpris- SPE C IE S A N D I N COM M U N ITIE S
ingly, VoCC metrics provide a better match
to observed range shifts in both marine and In spite of the inherent difficulties in de-
terrestrial systems than does absolute tem- tecting climate change impacts in natural
perature change (Burrows et al. 2011; Poloc- systems, studies showing impacts on indi-
zanska et al. 2013). In regions where VoCC vidual species are numerous (>800 at the
is strongest, indications of whole biome time of Parmesan 2006) and increasing
32 W H AT C H A NGE S A R E W E OB SE RV I NG?
every year. The scales of study have varied COM PL E X ITI E S OF CONSE RVATION
from local, to regional, to continental, and PL A N N I NG W H E N A SYST E M IS
from 20 years to more than 100 years. As- DY NA M IC : A N E X A M PL E F ROM
signing climate change as the cause of the EUPH Y DRYAS EDITH A
observed biotic changes often has a deeper
basis (see Table 3.1), such as a known mech- E. editha as a Sensitive Indicator of Climate
anistic link between climatic variables and Change Impacts
biology of the study species (reviewed by Having been studied for more than 60 years
Parmesan et al. 2000; Easterling, Evans, et by dozens of researchers, Edith’s check-
al. 2000; Easterling, Meehl, et al. 2000; Ot- erspot butterfly presents a model example
tersen et al. 2001; Walther et al. 2002). In- of how knowledge at many different lev-
dividual studies are too numerous to review els can be integrated to answer complex
exhaustively, and there are many examples questions concerning response to climate
in other chapters of this volume; therefore, change. In particular, a great deal is known
I focus here on a few examples that provide about the effects of weather and yearly cli-
insight into mechanistic understanding of mate variability on individual fitness and on
the role of climate change on shaping spe- population dynamics, and there is a long
cies distributions. (For more examples of history of coupling such long-term obser-
distribution and abundance changes, see vations with experimental manipulations
case study 5 in this volume). in laboratories, greenhouses, and natural
Geographic and taxonomic differences populations. This rich knowledge base al-
in rates of response have emerged from lowed for a mechanistic link to be made
the relative wealth of phenological data, between large-scale patterns of distribution
but long-term observations of the biogeo- change in E. editha and long-term climate
graphic dynamics of species are generally trends (over the twentieth century) (Table
too sparse to detect such differences. Nev- 3.1; Parmesan 2005b; Parmesan et al. 2013).
ertheless, a few patterns are emerging. For Here, I provide a brief summary of the
example, Poloczanska et al. (2013) com- how researchers have been able to link ob-
pared dynamics at the leading and trail- served population extinctions to extreme
ing range boundaries for marine species weather and climate events (reviewed in
(not always of the same species) and found Parmesan et al. 2013). In some populations,
leading edges to be expanding nearly five phenological asynchrony is the mechanistic
times as fast as trailing edges were contract- basis for climate sensitivity. Field observa-
ing (see Table 3.3). This asymmetrical pat- tions and experimental manipulations, per-
tern of range boundary shifts qualitatively formed over many decades, have repeatedly
matches that found for European butter- shown that phenological mismatches be-
flies, in which nearly 67 percent of species tween E. editha and annual host plants rou-
expanded their northern range boundaries tinely cause larvae to starve when hosts se-
but <20 percent contracted their southern nesce before the insects are ready to diapause
range boundaries (Parmesan et al. 1999). (Singer 1972; Weiss et al. 1988; Boughton
The reasons for these asymmetrical re- 1999; Hellmann 2002; Singer and McBride
sponses are not clear, but a likely candidate 2012). These mismatches existed prior to
is simply that while range expansion can the onset of current warming but render
occur rapidly (in a single year), population such populations vulnerable to warming
extinctions tend to lag behind deterioration conditions as even a small advancement of
of the environment, especially for long- plant senescence dramatically increases lar-
lived species. This process, termed extinction val mortality (Singer and Parmesan 2010).
debt, is a well-developed concept in the eco- However, shifts in the relative timing of
logical literature (Tilman et al. 1994). insect development and host senescence are
RANGE AND ABUNDANCE CHANGE S 33
not the only mechanisms related to climate son et al. 1999). The regional extinction of
change that have caused population extinc- a suite of E. editha populations occurred at
tions of this species. Observed responses of 2,300 m, caused by frequent false springs
different ecotypes of this butterfly living consistent with this overall shift toward
in different habitat types have been highly earlier snowmelt at this elevation.
diverse (Ehrlich et al. 1980; Thomas et al. In sum, three different mechanisms of
1996), and population extinctions have oc- climate change–caused population extinc-
curred in response to diverse but specific tion were observed in a single butterfly spe-
climatic conditions, particularly extreme cies: shifts in insect and host phenological
weather and extreme climate years. synchrony caused by hot years, different
In late winter at low to middle elevations dormancy responses by plants and insects
(500 m–1000 m) in the southern half of to drought, and direct mortality of both
this species’ range (~29°N to 45°N), many plants and insects caused by a series of false
populations specialize on annual hosts, seeds springs. All three of these extreme climate
normally germinate in response to rainfall, events either have increased, or are ex-
and larvae emerge from their overwintering pected to increase, with anthropogenic cli-
dormant phase (diapause) in response to a mate change (Easterling, Evans et al. 2000).
combination of winter chilling and rainfall. But even though E. editha populations are
There are levels of drought at which the lar- climate sensitive, this butterfly also, ironi-
vae emerge from winter diapause, but the cally, exhibits traits that appear to make it
plants do not germinate: this caused several resilient to changes in climate. First, there
butterfly population extinctions in the 1970s is large interpopulation variation in heri-
(Ehrlich et al. 1980). At higher elevations table behaviors that have a strong influence
(>1,500 m), both seeds and larvae break dor- on microtemperatures of eggs and early-
mancy at snowmelt, so milder winters and stage offspring (Bennett et al. 2015). Female
reduced snowpack have advanced both plant E. editha show interpopulation variation in
and insect activity to the point where both egg placement, exposing their offspring to
are vulnerable to “false springs.” False springs diverse thermal environments, from 3°C
occur when late winter is warm enough that cooler to 20°C hotter than ambient air.
plants and animals respond to this false cue Second, the species’ single flight season is
and emerge from winter dormancy, only placed at different times of year across its
to be exposed to (and often killed by) later range, with flight in midsummer in cool
freezing temperatures as typical winter con- climates and diapause through summer in
ditions prevail. A series of false springs drove hot climates. Further, whereas both traits
a set of E. editha populations at 2,300 m to are highly heritable, the species contains
extinction in the 1980s and 1990s (Thomas substantial genetic variation in these traits;
et al. 1996; Boughton 1999). therefore, modification of both phenology
Anthropogenic climate change has in- and egg placement are possible. This eco-
creased the frequencies of false springs at logical and evolutionary flexibility provide
lower montane elevations throughout the options for populations of this species to
world, where snowpack has been gradually mitigate future changes in regional climate.
declining and melting earlier (Mote et al.
2005; Stewart 2009; Beniston 2012). Specifi-
E. editha quino Has Demonstrated Resilience
cally, in the Sierra Nevada, which contains
to Climate Change Impacts
the highest elevation populations of E. editha,
snowpack dynamics showed a shift at 2,400 The Quino checkerspot, an endangered
m, showing no significant change above subspecies of Edith’s checkerspot, is cur-
2,400 m but becoming thinner and melting rently illustrating remarkable resilience to
two weeks earlier below 2,400 m (John- anthropogenic climate change, despite its
34 W H AT C H A NGE S A R E W E OB SE RV I NG?
Figure 3.1. Map showing timing of first observation of were not well adapted to it (Parmesan et al.
Quino populations relative to elevation. Topography 2015). The moral drawn from this is that
is depicted by higher elevations having a lighter back-
individuals do not need to be preadapted
ground color. A small number of the northernmost ex-
tinct sites lie outside the map area. (From Parmesan et al. to novel environments, or to exhibit rapid
2015.) evolution, in order to colonize beyond his-
torical range boundaries. They merely need
geographical location at the species’ equa- to survive.
torial range limit (Parmesan et al. 2015).
Quino has recently begun colonizing higher
Short-Term Success, but Long-Term
elevations than those historically inhabited
Vulnerability
by this subspecies, shifting from an average
elevation of 360 m throughout most of the Quino highlights the difficulty of plan-
twentieth century to an average of 1,164 m ning for future climate change impacts. In
by the 2010 decade (Figures 3.1 and 3.2). the near term, an upward range shift has
But, because Quino was historically already already been encouraged by adding pro-
at the upper elevational limits of its tradi- tection to areas into which the butterfly is
tional host plants (principally species of an- expanding on its own. However, projected
nual Plantago), to colonize higher elevations distributions indicate that, even in the
it had to switch its principal host to another medium term (next 40 years), Quino likely
species, Collinsia concolor, which inhabited will lose all suitable climate space in the
the mountain regions around Los Angeles region of its current distribution (Figure
and San Diego. There is no evidence that 3.3). For long-term persistence of Quino,
this host switch was driven by evolution- conservation managers will again need to
ary changes, and colonies on the novel host consider assisted colonization. With respect
RANGE AND ABUNDANCE CHANGE S 35
Figure 3.2. Scatter plot of elevations of populations shifts have been observed in many species,
against year that population was first discovered (first with few shifts toward the equator. Species
recorded). Populations were placed into one of three
groups depending on whether they were first discovered
compositions within communities have
in the period 1890–1997, 1998–2009, or 2010–2013. The generally altered in concert with local tem-
mean elevation for each grouping is indicated with a bar. perature rise. At single sites, species from
Elevation and year of first discovery are highly signifi- lower latitudes and elevations have tended
cantly correlated (r = 0.60, df = 116, p < 0.0001). (From to increase in abundance as those residing
Parmesan et al. 2015.)
predominately at higher latitudes and el-
evations suffer local declines. Consistency
to shifting geographic boundaries of spe- across studies indicates that local and re-
cies as they attempt to track climate change, gional trends reflect global trends.
the type of stepping-stone approach we These types of distributional changes,
advocated for Quino is likely to aid other first documented in the 1990s and continu-
endangered species as well. True adaptive ing to be documented at ever increasing
management, with regular and frequent rates, differ idiosyncratically in the exact
reassessments of conservation needs and patterns of colonizations and extinctions
priorities, will become increasingly neces- across regions. Further, not all species re-
sary in a time of rapid climate change. At- spond to climate change by shifting their
tention to real-time dynamics, then, can be distributions. This individualism in species’
used to reduce current uncertainty in future responses to climate change, resulting in
projections, both in climate and in species’ asynchronous range shifts, has led to early
responses, by informing managers as to signs of altered species interactions as spe-
which future (projected) pathway appears cies shift geographically (e.g., the Quino
to be emerging for their species. butterfly as a novel herbivore to Collinisa con-
color). Individual species’ range shifts have
begun to manifest as shifts in major bi-
SU M M A RY omes, particularly in areas of the greatest
magnitude of climate change. Traditional
Global coherency in patterns of biological conservation models are rapidly being
change, a substantial literature linking cli- transformed by an accelerating documenta-
mate variables with ecological and physi- tion of changes in wild systems that have
ological processes, combined with telltale been linked to climatic trends. Although
diagnostic fingerprints allow a causal link many effects have been relatively minor,
to be made between twentieth-century cli- and even benign, the most sensitive species
mate change and biological impacts. The and systems already are being harmed se-
patterns of change alone are evidence of verely (Parmesan 2006). It is clear that the
a climate change signal (reviewed in Par- continuation of observed trends is likely
mesan 2006; Pecl 2017). Poleward range to have profound impacts on wild species
36 W H AT C H A NGE S A R E W E OB SE RV I NG?
Figure 3.3. Current and future projected species distribu- Bennett, Nichole L., Paul M. Severns, Camille Parmesan,
tions for Quino based on climatic niche models. Known and Michael C. Singer. 2015. “Geographic mosaics
historical records are shown with the plus symbol (+). of phenology, host preference, adult size and micro-
Current climate projections (light gray) and future cli- habitat choice predict butterfly resilience to climate
mate projections (dark gray) represent an ensemble of warming.” Oikos 124 (1). https://doi.org/10.1111/
all model outputs, with “presence” indicated by light or oik.01490.
dark gray coloring if a grid cell was estimated as pres- Boughton, David A. 1999. “Empirical evidence for com-
ent in at least 50% of SDM models (i.e., anywhere from plex source-sink dynamics with alternative states in a
50% to 100% of model agreement). (From Parmesan butterfly metapopulation.” Ecology 80: 2727–2739.
et al. 2015.) Burrows, Michael T., David S. Schoeman, Lauren B.
Buckley, Pippa Moore, Elvira S. Poloczanska, Keith M.
Brander, Chris Brown, et al. 2011. “The pace of shift-
and consequently on the ways in which the ing climate in marine and terrestrial ecosystems.” Sci-
conservation community strives to preserve ence 334: 652–655.
biodiversity. Chen, I-Ching, Jane K. Hill, Ralf Ohlemüller, David B.
Roy, and Chris D. Thomas. 2011. “Rapid range shifts
of species associated with high levels of climate
warming.” Science 33: 1024–1026.
R E FE R E NCE S Cramer, Wolfgang, Gray W. Yohe, Maximillian Auff-
hammer, Christian Huggel, Ulf Molau, M. Assunção
Anthony, K. R. N., S. R. Connolly, and O. Hoegh-Guld- Faus da Silva Dias, Andrew Solow, Dáithí A. Stone,
berg. 2007. “Bleaching, energetics, and coral mortal- Lourdes Tibig, Laurens Bouwer, et al. 2014. “Detec-
ity risk: Effects of temperature, light, and sediment tion and attribution of observed impacts.” In Climate
regime.” Limnology & Oceanography 52: 716–726. Change 2014: Impacts, Adaptation and Vulnerability. Contribution
Beniston, Martin. 2012. “Is snow in the Alps receding of Working Group II to the Fifth Assessment Report of the Intergov-
or disappearing?” Wiley Interdisciplinary Reviews—Climate ernmental Panel on Climate Change, ed. Christopher B. Field,
Change 3 (4): 349–358. et al., 979–1037. Cambridge University Press.
RANGE AND ABUNDANCE CHANGE S 37
Dial, Roman J., T. Scott Smeltz, Patrick F. Sullivan, Chris- United States.” Bulletin of the American Meteorological Society
tina L. Rinas, Katriina Timm, Jason E. Geck, S. Carl 77: 279–292.
Tobin, Trevor S. Golden, and Edward C. Berg. 2016. Latif, M., and N. S. Keenlyside. 2009. “El Niño/South-
“Shrubline but not treeline advance matches climate ern Oscillation response to global warming.” Philo-
velocity in montane ecosystems of south-central sophical Transactions of the Royal Society of London B 106 (49):
Alaska.” Global Change Biology 22: 1841–1856. https://doi 20578–20583.
.org/10.1111/gcb.13207. Loarie, Scott R., Phillip B. Duffy, Healy Hamilton, Greg-
Easterling, David R., Jenni L. Evans, P. Ya Groisman, ory P. Asner, Christopher B. Field, and David D. Ack-
Thomas R. Karl, Kenneth E. Kunkel, and P. Ambenje. erly. 2009. “The velocity of climate change.” Nature
2000. “Observed variability and trends in extreme 462: 1052–1057.
climate events: A brief review.” Bulletin of the American McLaughlin, J. F., J. J. Hellmann, C. L. Boggs, and P. R.
Meteorological Society 80: 417–425. Ehrlich. 2002a. “Climate change hastens population
Easterling, David R., Gerald A. Meehl, Camille Parme- extinctions.” Proceedings of the National Academy of Sciences
san, Stanley A. Chagnon, Thomas R. Karl, and Linda 99: 6070–6074.
O. Mearns. 2000. “Climate extremes: Observations, McLaughlin, J. F., J. J. Hellmann, C. L. Boggs, and P.
modeling, and impacts.” Science 289: 2068–2074. R. Ehrlich. 2002b. “The route to extinction: Popula-
Ehrlich, Paul R., Dennis D. Murphy, Michael C. Singer, tion dynamics of a threatened butterfly.” Oecologia 132:
Caroline B. Sherwood, Ray R. White, and Irene L. 538–548.
Brown. 1980. “Extinction, reduction, stability and in- Mote, Phillip W., Alan F. Hamlet, Martyn P. Clark, and
crease: The responses of checkerspot butterfly popula- Dennis P. Lettenmaier. 2005. “Declining mountain
tions to the California drought.” Oecologia 46: 101–105. snowpack in western North America.” Bulletin of the
Fedorov, A. V., and S. G. Philander. 2000. “Is El Niño American Meteorological Society 86 (1): 39–49.
changing?” Science 288: 1997–2002. Ottersen, Geir, Benjamin Planque, Andrea Belgrano,
Glynn, P. W., and L. D’Croz. 1990. “Experimental evi- Eric Post, Phillip C. Reid, and Nils C. Stenseth. 2001.
dence for high temperature stress as the cause of “Ecological effects of the North Atlantic oscillation.”
El Niño-coincident coral mortality.” Coral Reefs 8: Oecologia 128 (1): 1–14.
181–191. Parmesan, Camille. 1996. “Climate and species’ range.”
Hansen J., et al. 2006. “Global temperature change.” Nature 382: 765–766.
Philosophical Transactions of the Royal Society of London B 103 Parmesan, Camille. 2003. “Butterflies as bio-indicators
(39): 14288–14293. of climate change impacts.” In Evolution and Ecology Taking
Hellmann, Jessica J. 2002. “The effect of an environ- Flight: Butterflies as Model Systems, edited by C. L. Boggs, W.
mental change on mobile butterfly larvae and the nu- B. Watt, and P. R. Ehrlich, 541–560. Chicago: Univer-
tritional quality of their hosts.” Journal of Animal Ecology sity of Chicago Press.
71: 925–936. Parmesan, Camille. 2005a. “Biotic response: Range and
Hoegh-Guldberg, O. 1999. “Climate change, coral abundance changes.” In Climate Change and Biodiversity,
bleaching and the future of the world’s coral reefs.” ed. T. Lovejoy and L. Hannah, 41–55. Yale University
Marine Freshwater Research 50: 839–866. Press.
Jones, R. J., O. Hoegh-Guldberg, A. W. D. Larkum, and Parmesan, Camille. 2005b. “Case study: Euphydryas editha.”
U. Schreiber. 1998. “Temperature-induced bleaching In Climate Change and Biodiversity, ed. T. Lovejoy and L.
of corals begins with impairment of the CO2 fixation Hannah, 56–60. Yale University Press.
mechanism in zooxanthellae.” Plant Cell & Environment Parmesan, Camille. 2006. “Ecological and evolutionary
21: 1219–1230. responses to recent climate change.” Annual Reviews of
Lesser, M. P. 1997. “Oxidative stress causes coral bleach- Ecology, Evolution and Systematics 37: 637–669.
ing during exposure to elevated temperatures.” Coral Parmesan, Camille, Michael T. Burrows, Carlos M. Du-
Reefs 16: 187–192. arte, Elvira S. Poloczanska, Anthony J. Richardson,
Johnson, Tammy, Jeff Dozier, and Joel Michaelsen. 1999. David S. Schoeman, and Michael C. Singer. 2013.
“Climate change and Sierra Nevada snowpack.” In In- “Beyond climate change attribution in ecology and
teractions between the Cryosphere, Climate and Greenhouse Gases, conservation research.” Ecology Letters 16 (S1): 58–71.
ed. Martyn Tranter, Richard Armstrong, Eric Brun, https://doi.org/10.1111/ele.12098.
Gerry Jones, Martin Sharp, and Martin Williams, Parmesan, Camille, Terry L. Root, and Michael R. Wil-
256: 63–70. International Association of Hydrologi- lig. 2000. “Impacts of extreme weather and climate
cal Sciences. on terrestrial biota.” Bulletin of the American Meteorological
Kapnick, S., and A. Hall. 2012. “Causes of recent changes Society 81: 443–450.
in western North American snowpack.” Climate Dynam- Parmesan, Camille, Nils Ryrholm, Constantí Stefanescu,
ics 38: 1885–1889. Jane K. Hill, Chris D. Thomas, Henri Descimon, Brian
Karl, T. R., R. W. Knight, D. R. Easterling, and R. G. Huntley, Lauri Kaila, Jaakko Kullberg, Toomas Tam-
Quayle. 1996. “Indices of climate change for the maru, W. John Tennent, Jeremey A. Thomas, and
38 W H AT C H A NGE S A R E W E OB SE RV I NG?
Martin Warren. 1999. “Poleward shifts in geographi- Singer, M. C., and P. R. Ehrlich. 1979. “Population dy-
cal ranges of butterfly species associated with regional namics of the checkerspot butterfly Euphydryas editha.”
warming.” Nature 399: 579–583. Fortschritte der Zoologie 25: 53–60.
Parmesan, Camille, Alison Williams-Anderson, Mat- Singer, Michael C., and Carolyn S. McBride. 2012. “Geo-
thew Mikheyev, Alexander S. Moskwik, and Michael C. graphic mosaics of species’ association: A definition
Singer. 2015. “Endangered Quino checkerspot butterfly and an example driven by plant–insect phenological
and climate change: Short-term success but long-term synchrony.” Ecology 93 (12): 2658–2673.
vulnerability?” Journal of Insect Conservation 19 (2): 185–204. Singer, Michael C., and Camille Parmesan. 2010. “Phe-
Parmesan, Camille, and Gray W. Yohe. 2003. “A globally nological asynchrony between herbivorous insects
coherent fingerprint of climate change impacts across and their hosts: A naturally-evolved starting point for
natural systems.” Nature 421: 37–42. climate change impacts?” Philosophical Transactions of the
Pecl, Gretta T., Miguel B. Araújo, Johann D. Bell, Julia Royal Society of London 365 (155): 3161–3176.
Blanchard, Timothy C. Bonebrake, I-Ching Chen, Stewart, Iris T. 2009. “Changes in snowpack and snow-
Timothy D. Clark, Robert K. Colwell, Finn Danielsen, melt runoff for key mountain regions.” Hydrological Pro-
Birgitta Evengård, et al. 2017. “Biodiversity redistri- cesses 23 (1): 78–94.
bution under climate change: Impacts on ecosys- Thomas, Chris D., Michael C. Singer, and David A.
tems and human well-being.” Science 355 (6332): art. Boughton. 1996. “Catastrophic extinction of popula-
eaai9214. https://doi.org/10.1126/science.aai9214. tion sources in a butterfly metapopulation.” American
Poloczanska, Elvira S., Christopher J. Brown, William Naturalist 148: 957–975.
J. Sydeman, Wolfgang Kiessling, David S. Schoeman, Tilman, David, Robert M. May, Clarence L. Lehman,
Pippa J. Moore, Keith Brander, et al. 2013. “Global im- and Martin A. Nowak. 1994. “Habitat destruction and
print of climate change on marine life.” Nature Climate the extinction debt.” Nature 371: 65–66. https://doi
Change 3 (10): 919–925. .org/10.1038/371065a0.
Przeslawski, Rachel, Inke Falkner, Michael B. Ashcroft, Vernon, J. E. N. 2008. “Mass extinctions and ocean acid-
and Pat Hutchings. 2012. “Using rigorous selection ification: Biological constraints on geological dilem-
criteria to investigate marine range shifts.” Estuarine, mas.” Coral Reefs 27: 459.
Coastal and Shelft Science 113: 205–212. Walther, Gian-Reto, Eric Post, Peter Convey, Annette
Root, Terry L., Jeff T. Price, Kimberly R. Hall, Stephen H. Menzel, Camille Parmesan, Trevor J. C. Beebee, Jean-
Schneider, Cynthia Rosenzweig, and J. Allan Pounds. Marc Fromentin, Ove Hoegh-Guldberg, and Franz
2003. “Fingerprints of global warming on wild ani- Bairlein. 2002. “Ecological responses to recent climate
mals and plants.” Nature 421: 57–60. change.” Nature 416 (6879): 389–395.
Serreze, M. C., J. E. Walsh, F. Stuart Chapin III, T. Oster- Weiss, Stuart B., Dennis D. Murphy, and Raymond R.
kamp, M. Dyurgerov, V. Romanovsky, W. C. Oechel, J. White. 1988. “Sun, slope and butterflies: Topographic
Morison, T. Zhang, and R. G. Barry. 2000. “Observa- determinants of habitat quality for Euphydryas editha.”
tional evidence of recent change in the northern high- Ecology 69 (5): 1486–1496.
latitude environment.” Climatic Change 46: 159–207. Wilkinson, C. R., ed. 2000. Global Coral Reef Monitoring Net-
Singer, Michael C. 1972. “Complex components of habi- work: Status of Coral Reefs of the World in 2000. Townsville:
tat suitability within a butterfly colony.” Science 176: Australian Institute of Marine Sciences.
75–77.
CASE STUDY 1
The Bering and Chukchi Seas encompass an extensive continental shelf to the north and east
of the deep Bering Sea, which is oceanographically an extension of the North Pacific. Sea-
ice cover extends typically south to the boundary between the shallow shelf and the deeper
basin in the Bering Sea from November to June. The Chukchi Sea has much more extensive
ice cover, which melts back to a minimum in September. In recent years, the retreat of sea-
sonal sea ice in late summer has often receded so that there is ice only over the deep Arctic
Ocean (Figure CS1.1).
In the Bering Sea, the shallow shelf can tween St. Matthew Island and St. Lawrence
be divided into two ecosystems. To the Island no significant decreases or, in some
south, a fish-dominated pelagic system cases, modest increases in sea-ice persis-
persists that is a globally important fishery, tence have been observed. It is thought
including Alaskan pollock (Gadus chalcogram- that where sea ice has diminished, such as
mus); snow, tanner, and king crab (Chion- in the northern Bering Sea and the Chuk-
oecetes opilio, Chionoecetes tanneri, and Paralithodes chi Sea, major ecological shifts to a more
camtschaticus); as well as Pacific salmon (five pelagic ecosystem can be expected (Greb-
species in the genus Oncorhynchus). To the meier et al. 2006), as illustrated by Figure
north and east, cold-water temperatures CS1.2. For example, walrus feed on abun-
(~0°C) occur year-round, influenced by dant clam populations in the Chukchi Sea
winter sea-ice formation, and these tem- shelf in the summer and use remnant sea
peratures inhibit reproduction of most fish ice as a resting offshore platform, but sea
species found farther south. A key ecologi- ice is now often persisting only over deeper
cal organizing principle for the northern waters of the Arctic Ocean in late summer,
Bering Sea and the adjoining Chukchi Sea beyond the diving capacity of walruses and
north of Bering Strait is that the shallow at depths where no significant food is pres-
productive waters lead to strong pelagic- ent on the seafloor. Once ice has retreated
benthic coupling to the seafloor and depo- so that it is only over deep water, walrus
sition of fresh chlorophyll coinciding with must come ashore in Alaska and Russia and
the spring bloom. This northern ecosystem swim much farther distances to offshore
is dominated by marine macroinvertebrates feeding grounds (Figure CS1.2).
(e.g., clams, polychaetes, sipunculids, crus- Changes to the northern Bering Sea and
taceans) that feed on the high production Chukchi Sea benthos has been well studied
deposited rapidly to the seafloor, which in over the past several decades (Grebmeier
turn serve as food resources for diving birds et al. 2006), and it is clear that dynamic
and mammals such as gray whales (Eschrich- biomass changes are occurring and that
tius robustus), bearded seals (Erignathus barbatus), decreases in biomass could be having an
eiders (genus Somateria), and walruses (Odo- impact on marine mammal and seabird
benus rosmarus). North of the Bering Strait in populations. In addition to changes in sea
the Chukchi Sea, and between St. Lawrence temperature and the spatial distribution of
Island and Bering Strait, the persistence of
seasonal sea ice has significantly declined Copyright © 2019 by Yale University. All rights reserved.
over the past three decades, although be- ISBN 978-0-300-20611-1.
39
40 W H AT C H A NGE S A R E W E OB SE RV I NG?
Figure CS1.1. Persistence of sea ice (in days per year) for tions southwest of St. Lawrence Island with
the Bering and Chukchi Seas, 2003–2010. Presence of sea polychaetes that may be less vulnerable to
ice is defined as more than a 15% sea-ice concentration
threshold in each pixel analyzed, using the Advanced
seasonal acidification. The area south of St.
Microwave Scanning Radiometer (AMSR-E) satellite Lawrence Island is known to be important
time series. (Analysis and figure courtesy of Karen Frey, foraging areas for spectacled eiders (Somate-
Graduate School of Geography, Clark University. Pre- ria fischeri) and walruses, which heavily de-
pared by Lee Cooper.) pend on bivalves for food.
The possibility that commercial fisheries
the cold-water pool that inhibits northward in the southeastern Bering Sea, which are
migration of many fish, already-seasonally- the largest by biomass in the United States,
persistent low pH in parts of the northern could eventually move northward onto the
Bering Sea are likely to be further affected shallow northern shelf now dominated by
by ocean acidification due to cold tempera- soft, bottom benthic animals is understood
tures and freshwater melt. This may be in- by fisheries managers. For example, the
fluencing the replacement of clam popula- North Pacific Fisheries Management Council,
THE BER ING SE A A ND CLIM ATE CH A NGE 41
Figure CS1.2. It is thought that where sea ice has dimin- possibility of fisheries expansion as the cli-
ished, as in the northern Bering Sea and Chukchi Sea, mate warms, even though limited scientific
major ecological shifts can be expected (Grebmeier et
al. 2006), as shown in the twin diagrams. Less exten-
trawling and sampling of the bottom by US
sive sea ice will lead to lower deposition of sea ice and and Japanese scientists continues to monitor
spring-bloom algae (bottom), and a lower biomass ben- for possible shifts in commercially harvest-
thic community. Fish and zooplankton will increase in able fish and shellfish populations.
importance, and pelagic feeding whales such as orcas
and humpbacks will become more important than sea-
ice-associated bowheads and benthic feeding pinnipeds.
(Prepared by Lee Cooper.) R E FE R E NCE
42
PHENOLOGICA L DY NA MICS IN POLLINATOR-PL A NT AS SOCIATIONS 43
of phenology (Schwartz 2003; Post 2013). In contrast to regional-scale satellite ob-
Hence, phenology may be regarded as the servations that are blind to individual spe-
seasonal timing of expression of life-history cies, studies conducted at local sites have in
traits concerned with growth, reproduc- some instances documented substantially
tion, and survival. greater trends in plant phenology. Among
24 species of perennial plants occurring
at a study site in Mediterranean Spain, the
R E C E N T T R E N DS I N PH E NOLO GY timing of leaf unfolding advanced between
R E L AT E D TO C L IM AT E C H A NGE 1943 and 2003 at a mean rate of –4.8 days
per decade (Gordo and Sanz 2009). How-
The most intuitive biological indicators of ever, rates of advance in 20 percent of these
climate change are trends in phenology. species were as high as –7 to –7.5 days per
Examples of earlier (i.e., advancing) timing decade (Gordo and Sanz 2009). Among 232
of spring events are abundant, but those plant species monitored over a 27-year pe-
most commonly reported concern the tim- riod on the island of Guernsey in the Eng-
ing of plant phenology, including green-up lish Channel, annual first-flowering date ad-
and flowering. Estimates of rates of plant vanced on average by –5.2 days per decade,
phenological advance through time, de- but this rate was as high as –13.7 days per
noted as trends, and rates of plant pheno- decade for the subset of species comprising
logical advance in response to temperature evergreens (Bock et al. 2014). Similarly, cli-
increase or changes in other abiotic factors matic sensitivity in this study averaged –9.7
such as timing of snowmelt, denoted as days of advance in first flowering per de-
climatic sensitivity, vary greatly over scales of gree of warming among all species but was
observation and levels of biological organi- greatest for evergreen species, at –10.9 days
zation (Parmesan 2007). Observations de- of advance per degree of warming (Bock
riving from satellite imagery covering vast et al. 2014).
spatial scales and combining multiple spe- Trends in first-flowering dates since the
cies can produce widely differing rates of early 1930s for 24 species at a site near Mo-
phenological advance, even when obtained honk Lake, New York, in the northeastern
for a single region. For instance, MODIS United States, varied from as low as –0.05
satellite observations over the Svalbard ar- days per decade to –1.5 days per decade,
chipelago in High Arctic Norway revealed with some species, such as painted trillium
no trend in the onset of the plant-growing (Trillium undulatum), displaying significant
season between 2000 and 2013 (Karlsen et delays in the onset of flowering by up to
al. 2014). During the same period, MODIS 1.6 days per decade (Cook et al. 2008). In
satellite data covering four subzones of the contrast to these modest rates of advance,
Yamal Peninsula in Arctic Russia revealed or in some instances delays, record-warm
variation in trends in the start of the plant- spring temperatures in Massachusetts, also
growing season among these subzones in northeastern United States, and in Wis-
from an advance of –5.3 days per decade to consin, in the northern Midwest United
a delay of 18.9 days per decade (Zeng, Jia, States, were accompanied by record-early
and Forbes 2013). Over the entire North- flowering in some species of plants at long-
ern Hemisphere, the timing of onset of the term phenological monitoring sites in both
annual plant-growing season as detected locations (Ellwood et al. 2013). In Massa-
by satellite observations advanced by –5.2 chusetts, flowering occurred 21 days ear-
days per decade between the years 1982 lier during the record-warm spring of 2010
and 1999, but this rate slowed to –0.2 days than during the mid-nineteenth century,
per decade between 2000 and 2008 (Jeong whereas in Wisconsin, flowering occurred
et al. 2011). 24 days earlier in the record-warm spring
44 W H AT C H A NGE S A R E W E OB SE RV I NG?
species (Potts et al. 2010). In part, this may one common species of flowering plant
be attributed to a scarcity of case studies, in the High Arctic, mountain avens (Dryas
because currently only a few investigations sp.) (Tiusanen et al. 2016). Seasonal over-
have focused on testing this hypothesis lap between the phenology of bee pollina-
(Rafferty et al. 2013). Nonetheless, evidence tors and their flowering-plant mutualists
of differential phenological shifts in polli- was also disrupted during an anomalously
nator versus flowering-plant phenology in- warm spring in a mid-latitude alpine sys-
dicates the potential for climate change to tem (Kudo 2014). In the United Kingdom,
disrupt such associations. species of bumblebees currently in decline
In many specialized plant-pollinator as- are those presumed to have narrow cli-
sociations, the flowering phenology of the matic niches (Williams, Araújo, and Ras-
focal plant species typically precedes that mont 2007; Williams, Colla, and Xie 2009),
of the timing of pollinator activity on a whereas, generally, pollinator-plant asso-
seasonal basis, and plant-flowering phe- ciations expected to persist under climate
nology advances more quickly in response change are those that display generalized
to warming than does pollinator phenol- adaptations to climatic conditions and spe-
ogy (Forrest and Thomson 2011; Solga, cies associations (Burkle and Alarcon 2011).
Harmon, and Ganguli 2014; Kudo and Ida A recent and comprehensive examination
2013). In parts of Northern Europe, spe- of the consequences for pollinator species’
cialists and sedentary species of bee and richness of changes in floral plant phenol-
syrphid fly pollinators have declined since ogy focused on associations involving 665
1980, but generalists and highly mobile species of insect pollinators and the phe-
species have not (Potts et al. 2010). Among nological dynamics of 132 species of their
plants in the same areas, insect-pollinated plant hosts in a Mediterranean community
species have undergone greater declines near Athens, Greece (Petanidou et al. 2014).
than have wind- and self-pollinated species This study classified floral resources accord-
(Potts et al. 2010). Whether or not recent ing to few- versus many-flowered plant
climate change has contributed to these species and annual versus perennial plant
patterns, they indicate the consequences of species, and it quantified the numbers of
phenological niche conservatism or narrow insect pollinators visiting each plant species
phenological niche breadth for species un- over 4 years. Pollinators comprised bees,
able to adjust to rapid shifts in environmen- muscoid flies, hoverflies, beetles, wasps,
tal conditions (sensu Bartomeus et al. 2013). butterflies, and true bugs. The numbers of
Some data indicate that specialized pol- pollinator species observed visiting plants
linators are more likely than generalists to were compared to the annual deviation in
suffer resource loss from phenological mis- plant flowering phenology from the 4-year
match with their plant mutualists if they mean to analyze consequences for pollina-
do not keep pace with advancing timing tor richness of advances or delays in flower-
of flowering in response to climate change ing phenology.
(Hegland et al. 2009). For example, an ob- This analysis revealed, at the scale of the
servational study based on 14 years of data entire plant community, a positive associa-
documented a decline in abundance of tion between onset of flowering and pol-
pollinator species of muscid flies and chi- linator richness but a negative association
ronomid midges associated with a shorter between flowering duration and pollinator
flowering season at a High Arctic site in richness (Figure 4.1).
Greenland (Høye et al. 2013). The decline These relationships suggest that earlier
in muscid fly abundance at the site is par- onset of flowering, and prolonged flower-
ticularly troublesome considering that they ing across the community, were associated
are the key pollinator group for at least with visits by fewer species of insect polli-
PHENOLOGICA L DY NA MICS IN POLLINATOR-PL A NT AS SOCIATIONS 47
Figure 4.1. Magnitude and direction of correlations be- bers and then transferring them to outdoor
tween insect pollinator richness and the onset (gray col- plots with plants in an ambient phenologi-
umns) and duration (black columns) of flowering across
all plant species, and individually among few-flowered
cal state (Forrest 2015). Results of such ex-
species, many-flowered species, annuals, and perenni- periments have been mixed, with observa-
als in a Mediterranean community near Athens, Greece. tions of both increased (Rafferty and Ives
(Adapted from Petanidou et al. 2014.) 2011) and reduced (Parsche, Frund, and
Tscharntke 2011) pollinator visitation of
nators. The same associations were evident experimentally advanced plants compared
for few-flowered species, many-flowered to phenologically ambient plants. Multiple
species, annuals, and perennials individu- observational studies have demonstrated
ally (Figure 4.1). However, the associations more rapid phenological advance in flow-
were strongest for few-flowered species, ering plants than in their insect mutualists
and the negative correlation between flow- in response to warming (Iler, Inouye, et al.
ering duration and pollinator richness was 2013; Kudo and Ida 2013; Solga, Harmon,
not significant for perennials (Petanidou and Ganguli 2014; Forrest and Thomson
et al. 2014). Because onset and duration of 2011; Rafferty and Ives 2012; Warren, Bahn,
flowering were negatively correlated for and Bradford 2011). Reduced pollinator ef-
the entire community and each category fectiveness may be expected to result from
of flowering plants individually, the rela- increasing phenological mismatch involv-
tionships between flowering duration and ing specialists in such cases (Rafferty et al.
pollinator species richness may actually be 2013; Rafferty and Ives 2011; Kudo and Ida
driven by the negative response of pollina- 2013). However, integrity of pollination
tor richness to earlier onset of flowering services to plants can be maintained despite
(Petanidou et al. 2014). phenological shifts in response to climate
Evidence deriving primarily from obser- change in associations involving generalist
vational studies illustrates the potential for pollinators or pollinators that advance their
development of phenological mismatches timing apace with changes in flowering
between insect pollinators and flowering phenology (Cleland et al. 2012; Bartomeus
plants as a result of climate change. Ex- et al. 2011; Iler, Inouye, et al. 2013; Solga,
perimental studies on this topic have, by Harmon, and Ganguli 2014; Gilman et al.
contrast, been few to date and primarily 2012).
“temporal transplants” (Forrest 2015). Such At least one observational study has
experiments have focused on advancing shown that temporal overlap between in-
the phenology of plants in growth cham- sect pollinators and their floral resources
48 W H AT C H A NGE S A R E W E OB SE RV I NG?
may remain intact even when species at flowering plants and syrphid fly pollinators
each trophic level respond differentially varied by more than twofold between the
to climate change. This analysis used a 20- earliest and latest dates of annual snowmelt
year record of timing of flowering by nine (Iler, Inouye, et al. 2013).
plant species and timing of emergence at As noted earlier, Høye et al. (2013) docu-
the community level by syrphid flies com- mented declines in the abundance of insect
prising at least 15 flower-visiting species at pollinators resulting from phenological
an alpine site at Rocky Mountain Biologi- mismatch with advancement and spatial
cal Station in Colorado (Iler, Inouye, et al. compression of flowering by the plants they
2013). Flowering phenology at this site visit. Importantly, plants themselves may
was associated strongly with the timing of suffer adverse consequences of advancing
snowmelt and generally preceded timing of flowering phenology that outpaces shifts in
emergence by syrphid pollinators. As a con- the timing of activity by their insect polli-
sequence, earlier annual snowmelt was ac- nators. A review of six studies of plant phe-
companied by earlier flowering across plant nological dynamics comprising between 19
species, and flowering phenology advanced and 385 species each, and of four studies
at a higher rate in association with earlier of phenological dynamics of insect pollina-
snowmelt than did fly emergence (Figure tors comprising between 10 and 35 species
4.2). Similarly, the termination of the flow- each, concluded that species-level responses
ering season advanced more strongly with to climate change are highly individualistic,
earlier snowmelt than did the end of the with few general patterns (Solga, Harmon,
annual period of fly activity (Figure 4.2). and Ganguli 2014).
Because syrphid flies at the study site This megareview highlighted three ex-
generally terminated activity prior to the amples of plant species that experienced ad-
end of the annual flowering season, these
differential rates of phenological advance
in response to earlier snowmelt resulted Figure 4.2. Model coefficients quantifying the rate of
phenological advance in flowering onset, onset of syr-
in greater overall phenological overlap be- phid fly activity, end of flowering, and end of syrphid
tween flowers and syrphid flies in years fly activity in relation to the annual timing of snowmelt
with early compared to late snowmelt (Iler, over a 20-year observational study at Rocky Mountain
Inouye, et al. 2013). Across the range of dates Biological Laboratory, Colorado, United States. Aster-
isks indicate statistical significance (p < 0.05) of taxa ×
observed in this 20-year study, the number
snowmelt coefficients. (Adapted from Iler, Inouye, et al.
of days of phenological overlap between 2013b.)
PHENOLOGICA L DY NA MICS IN POLLINATOR-PL A NT AS SOCIATIONS 49
verse consequences of the loss of pollinator 130 years and compared the rates of ad-
services as a result of mismatches deriving vance in springtime activity by those spe-
from differential phenological responses to cies to published estimates of rates of ad-
climate change. A species of glacier lily (Er- vance in flowering dates by 38 species of
ythronium grandiflorum) in a subalpine meadow plants from the same region. Particularly
in Colorado suffered pollen limitation early noteworthy features of this analysis include
in the flowering season due to an absence the length of the time series for both taxo-
of active bumblebee pollinators that early nomic groups, the temporal and spatial
in the season (Thomson 2010). Similarly, a coherence among the pollinator and flow-
species of star-of-Bethlehem (Gagea lutea) in ering plant data sets, and the restriction of
Japan advanced its flowering phenology in this analysis to plant species visited by the
response to warmer spring temperatures to species of pollinators included in the study
an extent that rendered it unavailable to its (Bartomeus et al. 2011). Across the entire
bee pollinator (Kudo et al. 2004). Finally, time series, the 10 species of bees in this
the flowering plant yan hu suo (Corydalis am- study advanced their emergence phenology
bigua) exhibited reduced seed set at a site in by approximately –0.8 days per decade, but
Japan following an advancement of its flow- the rate of advance has nearly doubled since
ering phenology in response to springtime 1970, indicating a response to accelerated
warming because its bee pollinator did not warming (Bartomeus et al. 2011).
undergo an advancement of its emergence Analyses of pooled data spanning the en-
phenology (Kudo et al. 2004). Although tire time series revealed that the rates of ad-
these examples apply to flowering plants vance of springtime activity by bees and the
with specialist pollinators or plant species timing of flowering by plants in this region
that are visited primarily by a single spe- have largely paralleled each other, suggest-
cies of pollinator, generalist pollinators can ing an absence of evidence of developing
apparently replace other pollinators whose mismatch (Bartomeus et al. 2011). Further-
phenology does not advance in nonspecial- more, the plant species in this comparison
ist associations (Solga, Harmon, and Gan- exhibited greater rates of advancement in
guli 2014). Similarly, generalist pollinators flowering phenology since 1970, and for
that show preference for some plant species both bees and plants, early-active species
may switch their preference to other species displayed more rapid phenological advance-
if they become phenologically mismatched ment than those active later in the season
from their preferred resources (Forrest and (Bartomeus et al. 2011). These observations
Thomson 2011). suggest that a phenological mismatch be-
As is evident in the examples here, phe- tween bee pollinators and flowering plants
nological mismatch is commonly studied in the same general region has not yet de-
as an outcome of differential rates of ad- veloped, but the authors caution that such
vance in springtime events between floral a mismatch could develop with additional
resources and pollinators in general or be- warming in the future (Bartomeus et al.
tween pollinators classified as specialists or 2011).
generalists. In our final example, we review That a mismatch is likely for at least
a study of species-specific phenological dy- some of the species in this study is sug-
namics in paired pollinator-plant associa- gested by a comparison of rates of advance
tions to examine interactions at the species by individual bee species and the plant spe-
level. cies they visit (Figure 4.3). We plotted rates
Bartomeus et al. (2011) analyzed data on of advance in springtime activity by each of
the emergence phenology of 10 species of the 10 bee species in this study versus rates
generalist bees in the northeastern United of advance in flowering phenology only for
States and southeastern Canada spanning those species visited by each of them in the
50 W H AT C H A NGE S A R E W E OB SE RV I NG?
-0.35 -0.25 -0.15 -0.05 0.05 0.15 0.25 gence phenology lagged behind advances
-0.01
0
0.5
1.5
-4
-3
-2
-1
0
1
2
3
-3.5
-2.5
-1.5
-0.5
Uvularia grandifolia Sanguinaria Canadensis
Trillium erectum Sanguinaria Canadensis
Lithospermum canescens
Aquilegia Canadensis Caltha palustris
Geranium maculatum Polygala paucifolia Amelanchier arborea
Dicentra cucullarria Amelanchier arborea Conus florida
O. atriventris
Dicentra cucullarria
Sanguinaria Canadensis
Uvularia grandifolia Antennaria neglecta
Dicentra cucullarria Amelanchier laevis
Caltha palustris
Trillium grandiflorum
Geranium maculatum Uvularia grandifolia
A. carlini
Phlox divaricata
Amelanchier laevis
O. bucephala
Lupinus perennis Aquilegia Canadensis
Amelanchier laevis
Uvularia grandifolia Aquilegia Canadensis
O. lignaria
Prunus virginiana
Dicentra cucullarria
Geranium maculatum
Polygala paucifolia Rosa Carolina
Lupinus perennis Uvularia grandifolia
Lithospermum canescens Caltha palustris
Oxalis violacea
A. miserabilis
O. pumila
Phlox divaricata Amelanchier arborea
nological niches. Phenological niche spe- for large and diverse bee populations (Hymenop-
cialization, whether relative to dependence tera: Apoidea: Apiformes).” Urban Forestry & Urban
Greening no. 20: 448–452. https://doi.org/10.1016/j
on species interactions or to abiotic condi- .ufug.2016.10.015.
tions, is likely to result in a loss of existing Bartomeus, Ignasi, John S. Ascher, Jason Gibbs, Bryan
pollinator-plant associations if advances in N. Danforth, David L. Wagner, Shannon M. Hedtke,
the timing of abiotic conditions favorable and Rachael Winfree. 2013. “Historical changes in
for reproductive activity are not matched by northeastern US bee pollinators related to shared
ecological traits.” Proceedings of the National Academy of Sci-
sufficient shifts in the timing of pollinator ences 110 (12): 4656–4660. https://doi.org/10.1073/
activity. Importantly, this may occur even pnas.1218503110.
if mean abiotic conditions themselves, such Bartomeus, Ignasi, John S. Ascher, David Wagner, Bryan
as seasonal or annual temperature, do not N. Danforth, Sheila Colla, Sarah Kornbluth, and Ra-
change but the timing of critical thermal chael Winfree. 2011. “Climate-associated phenologi-
cal advances in bee pollinators and bee-pollinated
thresholds does change. As well, advances plants.” Proceedings of the National Academy of Sciences
in the timing of pollinator activity, even if 51 (108): 20645–20649. https://doi.org/10.1073/
these keep pace with advances in flowering pnas.1115559108.
phenology, may not ensure maintenance of Bock, A., T. H. Sparks, N. Estrella, N. Jee, A. Casebow, C.
existing pollinator-plant associations if, for Schunk, M. Leuchner, and A. Menzel. 2014. “Changes
in first flowering dates and flowering duration of 232
instance, pollinator advances increase their plant species on the island of Guernsey.” Global Change
early-season exposure to natural enemies or Biology 11 (20): 3508–3519. https://doi.org/10.1111/
adverse temperature fluctuations. Finally, in gcb.12579.
cases of pollinator loss due to phenologi- Both, C., R. G. Bijlsma, and M. E. Visser. 2005. “Climatic
cal mismatches resulting from more rapid effects on timing of spring migration and breeding in
a long-distance migrant, the pied flycatcher Ficedula
advancement in the timing of flowering hypoleuca.” Journal of Avian Biology 5 (36): 368–373.
compared to the timing of pollinator activ- Both, C., S. Bouwhuis, C. M. Lessells, and M. E. Visser.
ity, persistence of the plant species involved 2006. “Climate change and population declines in a
in existing pollinator-plant mutualisms may long-distance migratory bird.” Nature 441 (7089): 81–83.
depend on replacement by pollinator spe- Both, C., M. van Asch, R. G. Bijlsma, A. B. van den
Burg, and M. E. Visser. 2009. “Climate change
cies with broader, more flexible, and hence and unequal phenological changes across four
less specialized phenological niches. Com- trophic levels: Constraints or adaptations?” Jour-
plementary measures to conserve both na- nal of Animal Ecology 1 (78): 73–83. https://doi
tive floral and pollinator diversity should be .org/10.1111/j.1365-2656.2008.01458.x.
undertaken where lacking and maintained Burkle, Laura A., and Ruben Alarcon. 2011. “The future
of plant-pollinator diversity: Understanding interac-
where already in effect, as strategies to buf- tion networks across time, space, and global change.”
fer both taxonomic groups against pheno- American Journal of Botany 3 (98): 528–538. https://doi
logical shifts by the species upon which they .org/10.3732/ajb.1000391.
depend. Programs to foster the proliferation Cleland, Elsa E., Jenica M. Allen, Theresa M. Crimmins,
of urban green spaces and pollinator gar- Jennifer A. Dunne, Stephanie Pau, Steven E. Travers,
Erika S. Zavaleta, and Elizabeth M. Wolkovich. 2012.
dens may also offer particularly affordable “Phenological tracking enables positive species re-
and achievable local actions, with the ancil- sponses to climate change.” Ecology 8 (93): 1765–1771.
lary benefit of raising public awareness of Cook, B. I., E. R. Cook, P. C. Huth, J. E. Thompson, A.
the importance of pollinator conservation Forster, and D. Smiley. 2008. “A cross-taxa phenologi-
for ecosystem services (Banaszak-Cibicka, cal dataset from Mohonk Lake, NY and its relation-
ship to climate.” International Journal of Climatology 10 (28):
Ratynska, and Dylewski 2016). 1369–1383. https://doi.org/10.1002/joc.1629.
Ellwood, E. R., S. A. Temple, R. B. Primack, N. L. Brad-
ley, and C. C. Davis. 2013. “Record-breaking early
R E FE R E NCE S flowering in the eastern United States.” PLOS One 1 (8).
https://doi.org/10.1371/journal.pone.0053788.
Banaszak-Cibicka, W., H. Ratynska, and L. Dylewski. Forrest, J. R. K. 2015. “Plant-pollinator interactions and
2016. “Features of urban green space favourable phenological change: what can we learn about climate
PHENOLOGICA L DY NA MICS IN POLLINATOR-PL A NT AS SOCIATIONS 53
impacts from experiments and observations?” Oikos 1 hypothesis.” In Trait-Mediated Indirect Interactions: Ecological
(124): 4–13. https://doi.org/10.1111/oik.01386. and Evolutionary Perspectives, ed. T. Ohgushi, O. J. Schmitz,
Forrest, Jessica R. K., and James D. Thomson. 2011. “An and R. D. Holt, 508–525. Cambridge University
examination of synchrony between insect emer- Press.
gence and flowering in Rocky Mountain meadows.” Kudo, Gaku. 2014. “Vulnerability of phenological syn-
Ecological Monographs 3 (81): 469–491. https://doi.org/ chrony between plants and pollinators in an alpine
10.1890/10-1885.1. ecosystem.” Ecological Research 4 (29): 571–581. https://
Gilman, R. Tucker, Nicholas S. Fabina, Karen C. Ab- doi.org/10.1007/s11284-013-1108-z.
bott, and Nicole E. Rafferty. 2012. “Evolution of Kudo, Gaku, and Takashi Y. Ida. 2013. “Early onset of
plant-pollinator mutualisms in response to climate spring increases the phenological mismatch between
change.” Evolutionary Applications 1 (5): 2–16. https://doi plants and pollinators.” Ecology 10 (94): 2311–2320.
.org/10.1111/j.1752-4571.2011.00202.x. https://doi.org/10.1890/12-2003.1.
Gordo, O., and J. J. Sanz. 2009. “Long-term temporal Kudo, G., Y. Nishikawa, T. Kasagi, and S. Kosuge.
changes of plant phenology in the Western Mediterra- 2004. “Does seed production of spring ephemer-
nean.” Global Change Biology 8 (15): 1930–1948. https:// als decrease when spring comes early?” Ecological Re-
doi.org/10.1111/j.1365-2486.2009.01851.x. search 2 (19): 255–259. https://doi.org/10.1111/j.1440
Hegland, Stein Joar, Anders Nielsen, Amparo Lazaro, -1703.2003.00630.x.
Anne-Line Bjerknes, and Orjan Totland. 2009. “How Miller-Rushing, A. J., T. T. Hoye, D. W. Inouye, and E.
does climate warming affect plant-pollinator inter- Post. 2010. “The effects of phenological mismatches
actions?” Ecology Letters 2 (12): 184–195. https://doi on demography.” Philosophical Transactions of the Royal Soci-
.org/10.1111/j.1461-0248.2008.01269.x. ety B—Biological Sciences 1555 (365): 3177–3186. https://
Høye, T. T., E. Post, H. Meltofte, N. M. Schmidt, and M. doi.org/10.1098/rstb.2010.0148.
C. Forchhammer. 2007. “Rapid advancement of spring Ollerton, J., H. Erenler, M. Edwards, and R. Crockett.
in the High Arctic.” Current Biology 12 (17): R449–R451. 2014. “Extinctions of aculeate pollinators in Britain
Høye, T. T., E. Post, N. M. Schmidt, K. Trojelsgaard, and and the role of large-scale agricultural changes.” Sci-
M. C. Forchhammer. 2013. “Shorter flowering sea- ence no. 346: 1360–1362.
sons and declining abundance of flower visitors in a Parmesan, Camille. 2007. “Influences of species, lati-
warmer Arctic.” Nature Climate Change 8 (3): 759–763. tudes and methodologies on estimates of pheno-
https://doi.org/10.1038/nclimate1909. logical response to global warming.” Global Change
Iler, A. M., T. T. Hoye, D. W. Inouye, and N. M. Schmidt. Biology 9 (13): 1860–1872. https://doi.org/10.1111/j
2013. “Long-term trends mask variation in the di- .1365-2486.2007.01404.x.
rection and magnitude of short-term phenological Parsche, S., J. Frund, and T. Tscharntke. 2011. “Experi-
shifts.” American Journal of Botany 7 (100): 1398–1406. mental environmental change and mutualistic vs.
https://doi.org/10.3732/ajb.1200490. antagonistic plant flower-visitor interactions.” Perspec-
Iler, A. M., D. W. Inouye, T. T. Hoye, A. J. Miller-Rush- tives in Plant Ecology Evolution and Systematics 1 (13): 27–35.
ing, L. A. Burkle, and E. B. Johnston. 2013. “Mainte- https://doi.org/10.1016/j.ppees.2010.12.001.
nance of temporal synchrony between syrphid flies Petanidou, Theodora, Athanasios S. Kallimanis, Stefa-
and floral resources despite differential phenological nos P. Sgardelis, Antonios D. Mazaris, John D. Pan-
responses to climate.” Global Change Biology 8 (19): 2348– tis, and Nickolas M. Waser. 2014. “Variable flowering
2359. https://doi.org/10.1111/gcb.12246. phenology and pollinator use in a community sug-
Iwasa, Y., and S. A. Levin. 1995. “The timing of life his- gest future phenological mismatch.” Acta Oecologica/
tory events.” Journal of Theoretical Biology 1 (172): 33–42. International Journal of Ecology no. 59: 104–111. https://doi
https://doi.org/10.1006/jtbi.1995.0003. .org/10.1016/j.actao.2014.06.001.
Jeong, S. J., C. H. Ho, H. J. Gim, and M. E. Brown. Plard, F., J. M. Gaillard, T. Coulson, A. J. M. Hewison, D.
2011. “Phenology shifts at start vs. end of grow- Delorme, C. Warnant, and C. Bonenfant. 2014. “Mis-
ing season in temperate vegetation over the North- match between birth date and vegetation phenology
ern Hemisphere for the period 1982–2008.” slows the demography of roe deer.” PLOS Biology 4 (12).
Global Change Biology 7 (17): 2385–2399. https://doi https://doi.org/10.1371/journal.pbio.1001828.
.org/10.1111/j.1365-2486.2011.02397.x. Post, E. 2013. Ecology of Climate Change: The Importance of Bi-
Karlsen, S. R., A. Elvebakk, K. A. Hogda, and T. otic Interactions. Monographs in Population Biology 52.
Grydeland. 2014. “Spatial and temporal variabil- Princeton University Press.
ity in the onset of the growing season on Svalbard, Post, E., and M. C. Forchhammer. 2008. “Climate change
Arctic Norway—Measured by MODIS-NDVI satel- reduces reproductive success of an arctic herbivore
lite data.” Remote Sensing 9 (6): 8088–8106. https://doi through trophic mismatch.” Philosophical Transactions of
.org/10.3390/rs6098088. the Royal Society of London, Series B (363): 2369–2375.
Kerby, J. T., C. C. Wilmers, and E. Post. 2013. “Climate Post, E., J. Kerby, C. Pedersen, and H. Steltzer. 2016.
change, phenology, and the nature of consumer-re- “Highly individualistic rates of plant phenological
source interactions: advancing the match/mismatch advance associated with arctic sea ice dynamics.”
54 W H AT C H A NGE S A R E W E OB SE RV I NG?
Biology Letters (12): 20160332. https://doi.org/10.1098/ Thomann, Michel, Eric Imbert, Celine Devaux, and
rsbl.2016.0332. Pierre-Olivier Cheptou. 2013. “Flowering plants un-
Post, E., S. A. Levin, Y. Iwasa, and N. C. Stenseth. 2001. der global pollinator decline.” Trends in Plant Science
“Reproductive asynchrony increases with environ- 18 (7): 353–359. https://doi.org/10.1016/j.tplants
mental disturbance.” Evolution 55 (4): 830–834. .2013.04.002.
Potts, Simon G., Jacobus C. Biesmeijer, Claire Kremen, Thomson, James D. 2010. “Flowering phenology, fruit-
Peter Neumann, Oliver Schweiger, and William E. ing success and progressive deterioration of pollina-
Kunin. 2010. “Global pollinator declines: trends, tion in an early-flowering geophyte.” Philosophical Trans-
impacts and drivers.” Trends in Ecology & Evolution 25 actions of the Royal Society B—Biological Sciences 1555 (365):
(6): 345–353. https://doi.org/10.1016/j.tree.2010 3187–3199. https://doi.org/10.1098/rstb.2010.0115.
.01.007. Tiusanen, M., P. D. N. Hebert, N. M. Schmidt, and T.
Rafferty, N. E., P. J. CaraDonna, L. A. Burkle, A. M. Iler, Roslin. 2016. “One fly to rule them all—muscid flies
and J. L. Bronstein. 2013. “Phenological overlap of are the key pollinators in the Arctic.” Proceedings of the
interacting species in a changing climate: An assess- Royal Society B—Biological Sciences 1839 (283). https://doi
ment of available approaches.” Ecology and Evolution 3 (9): .org/10.1098/rspb.2016.1271.
3183–3193. https://doi.org/10.1002/ece3.668. Visser, M. E., and C. Both. 2005. “Shifts in phenology
Rafferty, Nicole E., and Anthony R. Ives. 2011. “Effects of due to global climate change: The need for a yard-
experimental shifts in flowering phenology on plant- stick.” Proceedings of the Royal Society B-Biological Sciences 1581
pollinator interactions.” Ecology Letters 14 (1): 69–74. (272): 2561–2569.
https://doi.org/10.1111/j.1461–0248.2010.01557.x. Warren, Robert J., II, Volker Bahn, and Mark A. Brad-
Rafferty, Nicole E., and Anthony R. Ives. 2012. “Polli- ford. 2011. “Temperature cues phenological syn-
nator effectiveness varies with experimental shifts in chrony in ant-mediated seed dispersal.” Global
flowering time.” Ecology 93 (4): 803–814. Change Biology 7 (17): 2444–2454. https://doi.org/
Rasmont, P., and S. Iserbyt. 2012. “The bumblebees 10.1111/j.1365-2486.2010.02386.x.
scarcity syndrome: Are heat waves leading to local Williams, P. H., M. B. Araújo, and P. Rasmont. 2007.
extinctions of bumblebees (Hymenoptera: Apidae: “Can vulnerability among British bumblebee (Bombus)
Bombus)?” Annales de la societé entomologique de France 48 species be explained by niche position and breadth?”
(3–4): 275–280. Biological Conservation 138 (3–4): 493–505. https://doi
Schwartz, M. D. 2003. Phenology: An Integrative Environmental .org/10.1016/j.biocon.2007.06.001.
Science. Kluwer Academic Publishers. Williams, Paul, Sheila Colla, and Zhenghua Xie. 2009.
Solga, Michelle J., Jason P. Harmon, and Amy C. Ganguli. “Bumblebee vulnerability: Common correlates of
2014. “Timing is everything: An overview of pheno- winners and losers across three continents.” Conservation
logical changes to plants and their pollinators.” Natural Biology 23 (4): 931–940. https://doi.org/10.1111/j.1523
Areas Journal 34 (2): 227–234. -1739.2009.01176.x.
Stenseth, N. C., and A. Mysterud. 2002. “Climate, chang- Zeng, H. Q., G. S. Jia, and B. C. Forbes. 2013. “Shifts
ing phenology, and other life history and traits: Non- in Arctic phenology in response to climate and an-
linearity and match-mismatch to the environment.” thropogenic factors as detected from multiple satellite
Proceedings of the National Academy of Sciences 21 (99): 13379– time series.” Environmental Research Letters 8 (3). https://
13381. https://doi.org/10.1073/pnas.212519399. doi.org/10.1088/1748-9326/8/3/035036.
CHAPTER FIVE The most distinguishing feature of our
planet is the presence of liquid water cov-
Coral Reefs: ering 71 percent of its surface. This ocean
nurtured the beginnings of life, and today
Megadiversity Meets it is essential for its maintenance and future.
The ocean performs a vital regulatory role,
Unprecedented stabilizing planetary temperature, regulat-
ing the gas content of the atmosphere, and
Environmental determining the climate and weather sys-
tems on which life depends. In many ways,
Change the ocean represents the heart and lungs of
our planet, and life as we know it would not
OVE HOEGH-GULDBERG be possible without its presence.
Unfortunately, the latest science tells us
that we are starting to change the ocean in
fundamental ways that threaten its ability
to provide the ecosystem goods and ser-
vices on which much of humanity lives.
These changes have been highlighted in the
latest Intergovernmental Panel on Climate
Change (IPCC), which included several
new chapters on the potential response of
the ocean to rapid changes in global cli-
mate (Hoegh-Guldberg et al. 2014; Pörtner
et al. 2014). These chapters have established
a consensus that concludes that unprec-
edented and rapid changes are occurring
in ocean temperature, volume, structure,
chemistry, and therefore life itself.
The ocean plays a crucial role in stabi-
lizing planetary conditions, mainly due to
its size and enormous capacity to absorb
carbon dioxide (CO2) and heat with mini-
mal change in temperature. As a result, the
ocean has absorbed 93 percent of the extra
energy from the enhanced greenhouse ef-
fect, and approximately 30 percent of the
CO2 generated by human activities. These
changes have been essential in reducing
the rate at which warming has manifested
itself over the 150 years since the Indus-
trial Revolution, when human activities
55
56 W H AT C H A NGE S A R E W E OB SE RV I NG?
in which carbonate reef systems were pres- whether or not carbonate coral reefs ex-
ent and others, referred to as “gaps,” where ist. Areas such as the eastern Pacific, where
coral reefs were absent as significant reef upwelling waters at the equator are rich in
builders for millions of years (Veron 2008). CO2 and low in pH and carbonate, do not
Notably, reef gaps were typically associated have significant carbonate reef development
with the five major mass extinction events (Manzello et al. 2008). However, aragonite
linked to high CO2 concentrations that built saturation states of 3.3 or higher are gener-
up over thousands, if not millions, of years ally associated with coral reef communities
in the Earth’s atmosphere (Berner and Ko- that can form significant calcium carbon-
thavala 2001). As a result of these changes, ate structures (Hoegh-Guldberg et al. 2007;
Scleractinian corals and many other organ- Kleypas, McManus, et al. 1999).
isms underwent significant decreases in di- The contribution of Scleractinian corals
versity and abundance (Veron 2008). to reef building is strongly dependent on an
Coral reefs are dominated by Scleractin- ancient symbiosis with dinoflagellate pro-
ian corals that form extensive ecosystems in tists from the genus Symbiodinium (Muscatine
tropical regions (warm-water coral reefs), 1990; Muscatine and Porter 1977; Stanley
as well as ecosystems in deeper, low-light and Fautin 2001). The dinoflagellates are
habitats (so-called mesophotic coral reefs located within the gastrodermal tissues of
(Bongaerts et al. 2011) and in deep (down corals, where they continue to photosyn-
to 3000 m) locations (cold-water coral thesize (as any free-living dinoflagellate
reefs; Freiwald et al. 2004). Below 200 m would), providing as much as 95 percent
depth there is so little light that photosyn- of the organic carbon produced by the host.
thesis is no longer possible, and hence reefs In return, the coral host supplies inorganic
are made of corals that do not form symbio- nitrogen and phosphorus to the dinoflagel-
ses with single-celled dinoflagellates. All of lates, compounds that are usually at low
these coral communities, however, attract concentrations in the clear tropical and
and provide habitat for significant ecosys- subtropical waters in which corals and their
tems, many of which are very important dinoflagellates exist (Muscatine and Porter
for fisheries and human livelihoods. Given 1977). This mutualistic symbiosis is highly
space constraints, the rest of this chapter fo- efficient and leads to the internal recycling
cuses on warm-water coral reefs. of inorganic compounds between primary
Today, most warm-water coral reefs are producer and consumer. Corals also feed on
found in sunlit and low-nutrient coastal wa- particles suspended in the water column,
ters between the latitudes 30°N and 33°S. adding yet another source of nutrients and
Tropical and subtropical waters that accom- energy to the otherwise oligotrophic waters
modate warm-water coral reefs are also typical of most coral reefs. The capacity for
saturated with respect to the calcium and polytrophic behavior has been important to
carbonate ions necessary for skeletal forma- the success of corals, and hence coral reefs,
tion by corals and other organisms. The ra- over tens of millions of years of existence
tio of calcium carbonate to the saturation (Stanley and Fautin 2001).
constant of aragonite (the principal form of As a result of access to large quantities
calcium carbonate in coral skeletons) is a of organic carbon and energy, corals are
critical variable (Caldeira and Wickett 2003; able to precipitate large amounts of calcium
Kleypas, Buddemeier, et al. 1999). If too carbonate and build up significant skeletal
low, calcium carbonate deposition (calcifi- structures. These skeletal structures remain
cation) is unable to keep up with calcium after corals have died and are glued together
carbonate erosion and dissolution (decal- by the calcifying activities of other organ-
cification). The ratio of these two broad isms such as encrusting calcareous red al-
ecological processes ultimately determines gae. Other organisms, such as calcifying
CORAL REEFS 61
invertebrates, plankton, and plants, also Hoegh-Guldberg et al. 2007). These pro-
contribute significant amounts of calcium jections are borne out by a growing list
carbonate to reef systems as they develop. of long-term studies that are reporting a
Over time, these limestone-like deposits serious contraction in reef-building cor-
build up three-dimensional reef structures als over the past 30–50 years (Bruno and
and, over the longer term, form islands Selig 2007; De’ath et al. 2012; Gardner et al.
and coastal barrier systems. The three-di- 2003). Given the central role of massive cor-
mensional nature of coral reefs provides a als to reefs, this is also associated with the
complex habitat for the tens of thousands of loss of the three-dimensional structure of
species that live in and around typical coral coral reefs. In addition, mesocosm studies
reefs. like that of Dove et al. (2013) reveal rapid
loss of three-dimensional structure as cor-
als disappear under the pressure of future
LO C A L A N D GLOBA L O C E A N concentrations of atmospheric CO2.
WA R M I NG A N D AC IDIF IC ATION There is little doubt to many experts
that the greatest threat to coral reefs and
Coral reefs have experienced relatively small the hundreds of millions of people that de-
shifts in temperature and seawater chem- pend on them is rapid ocean warming and
istry over the past 420,000 years at least acidification, due to the vulnerability of the
(Hoegh-Guldberg et al. 2007) (Figure 5.1B). mutualistic symbiosis between Symbiodin-
When compared to the current rapid pace ium and reef-building corals (Gattuso et al.
of change in a number of environmental 2014; Hoegh-Guldberg et al. 2014; Pörtner
parameters, even the shifts between ice age et al. 2014). When rapid changes occur, or
and interglacial periods resulted in relatively conditions are pushed outside the environ-
small changes over much longer periods of mental envelope in which corals normally
time. As a consequence, the world’s largest exist, the symbiosis begins the breakdown
continuous reef system, the Great Barrier (Glynn 1993, 2012; Hoegh-Guldberg 1999;
Reef, has waxed and waned over thousands Hoegh-Guldberg and Smith 1989). This
of years as sea levels have risen and fallen. results in mass exodus of the brown di-
These changes are not directly comparable noflagellates out of the coral host’s tissues,
to the changes we are seeing today because leaving the coral host alive but bleached
they occurred over much longer periods of white in appearance (hence the term “coral
time, whereas current change is thought to bleaching”; see Plate 2).
be occurring faster than at any other time Around 1980, populations of corals be-
in the past 65 million years, if not 300 mil- gan to bleach across large areas of the world’s
lion years (Hönisch et al. 2012; IPCC 2013; tropical regions, with no precedent in the
Pachauri et al. 2014). These observations go scientific literature. Over time, information
for both local (e.g., pollution, overexploi- began to accumulate that suggested that
tation of key organisms; Burke et al. 2011) these large-scale changes to coral reefs are
and global (e.g., ocean warming and acidi- associated with short periods of time when
fication) sources (Hoegh-Guldberg 2014; sea temperatures were 1°C–2°C above the
Hoegh-Guldberg et al. 2014). long-term summer maxima for a particular
As human populations have increased, region (Hoegh-Guldberg 1999). Since that
coastal areas have been modified, and the time, the frequency and intensity of coral
demand for local resources has grown ex- bleaching has increased, with the phenom-
ponentially. As much as 75 percent of coral enon being reported in most tropical and
reefs are threatened, with as much as 95 subtropical regions of the world. Although
percent in danger of being lost by 2050 some coral reefs have recovered from mass
(Burke et al. 2011; Hoegh-Guldberg 1999; coral bleaching and mortality, many reefs
62 W H AT C H A NGE S A R E W E OB SE RV I NG?
have not (Baker et al. 2008). These types of the Great Barrier Reef can probably take a
studies have confirmed the conclusions of few heavy punches, but if the ability of its
modeling studies based on the thermal sen- corals to calcify and grow is compromised,
sitivity of coral reefs and underlying mech- then it is not surprising that the impact of
anisms (Donner et al. 2007; Donner et al. disturbances (i.e., loss of corals) will grow
2005; Frieler et al. 2013; Hoegh-Guldberg over time. These negative trends are occur-
1999). Studies of bleaching and recovery ring even though the Great Barrier Reef is
show that even with aggressive action on considered to have very low human pres-
CO2 emissions from the burning of fossil sures on it as compared to reefs in other
fuels and deforestation (e.g., the Paris Agree- regions, and it has been managed by a well-
ment in December 2015; Davenport 2015; resourced and modern marine park system
Schellnhuber et al. 2016), only 10 percent that has often been held up as the bastion of
of today’s corals will survive (Donner et al. good practice.
2005; Frieler et al. 2013; Hoegh-Guldberg These heavy punches keep coming,
1999; Hoegh-Guldberg et al. 2014). How- though. As 2015 finished—the warmest
ever, given the importance of coral reefs year on record—tropical waters across the
in terms of biological diversity, ecosystem planet began to exhibit exceptional warm-
services, and human well-being, saving the ing trends once again. This included the
remaining 10 percent of coral reefs will de- Great Barrier Reef, which exhibited sea
liver benefits in terms of the regeneration surface temperatures between 1°C and 3°C
of coral reefs under a stabilized climate, and higher than the long-term summer av-
hence should be a priority of leaders and erage in the first few months of 2016. By
policy makers (Pachauri et al. 2014). late March, large sections of the northern
It is very important to appreciate the Great Barrier Reef were exhibiting mass
challenge ahead. In 2012, a group of sci- coral bleaching on a scale never seen be-
entists at the Australian Institute of Marine fore (Hoegh-Guldberg and Ridgway 2016;
Science published the results of a long-term Hughes et al. 2016). In some cases, more
study that rigorously investigated changes than 90 percent of corals on reefs were
in the distribution and abundance of reef- bleached; in another case, less bleaching
building corals on the Great Barrier Reef. was reported. There was a clear gradient
Their conclusions were startling. Not only in heat stress and the extent of mass coral
had the growth rates of at least one group bleaching over the 2,300 km range of the
of reef-building corals across the Great Bar- Great Barrier Reef, with reefs in the south-
rier Reef decreased significantly since 1990 ern third seeing much lower temperatures
(De’ath et al. 2009); the amount of coral on and hence less mass coral bleaching. The
the Great Barrier Reef had declined by 50 resulting mortality of corals was also ex-
percent since 1983 (De’ath et al. 2012). A ceptional. By October and November, the
range of factors were included, but it is clear northern sector of the Great Barrier Reef
when considering the three main factors (about 800 km) was exhibiting an average
(i.e., cyclones, starfish predation, and ocean loss of 67 percent of coral cover on sample
warming) that the demise of the Great Bar- of 60 reefs (Hughes et al. 2016).
rier Reef is a complex combination of both These results indicate that projections
local and global factors. Central to this ob- made by scientists almost two decades ago
servation is that the ecological changes un- are coming to pass, with the clear likeli-
der way on the Great Barrier Reef probably hood that more frequent and more intense
involve synergies and antagonisms that in- bleaching events not only are a reality but
teract to produce an ecosystem that is losing also clearly threaten the very existence of
resilience even while experiencing a greater the world’s largest continuous coral reef.
number of impacts. Like any prizefighter, Following on from events in Pacific, this
CORAL REEFS 63
coral bleaching has affected a wide range of conditions sufficiently for megadiverse
other regions, including the Maldives, Ha- ecosystems such as coral reefs to meet this
waii, Japan, Southeast Asia, and many other challenge, atmospheric CO2 must be con-
countries, resembling the events associated strained well below the 450 ppm/2°C limit
with the 1998 El Niño (Hoegh-Guldberg (Frieler et al. 2013), which requires meeting
1999; Hoegh-Guldberg and Ridgway 2016), the goals of the Paris Agreement as quickly
albeit with greater intensity and mortality. as possible (Davenport 2015; Schellnhuber
et al. 2016). As emphasized by the events
of 2016 on the Great Barrier Reef and the
CONC LUSION: R E DUC I NG T H E R AT E world’s other coral reefs, we have prob-
OF GLOBA L C H A NGE TO Z E RO A N D ably underestimated the sensitivity of the
ACTI NG DE C ISI V E LY ON LO C A L world’s biological diversity and ecosystems
FACTOR S to climate change, further emphasizing the
need for us to move even more rapidly to
There is no doubt about the importance the goals of the Paris Agreement.
and interconnectedness of the ocean to all The second major action that we need
aspects of life on our planet. Without the is to rapidly reduce the influence of local
ocean functioning as it does today, life stresses such as declining water quality,
would be very different, if not impossible. overexploitation, pollution, and the physi-
Nonetheless, we are in the process of play- cal destruction of marine ecosystems such
ing catchup in terms of understanding the as coral reefs. In addition to the direct ef-
impacts of global change on the ocean, fects of these factors on the distribution and
whether from local or global sources. With abundance of healthy coral reefs, there is
the latest scientific consensus revealing that the significant contribution that these lo-
the ocean is undergoing changes that are cal factors make to reducing the resilience
unprecedented in 65 million years, if not of coral-dominated ecosystems to impacts
300 million years (Hönisch et al. 2012; IPCC and disturbances. Therefore, it is very im-
2013), there must be equally unprecedented portant that we confront and solve issues
action at a global level to solve the serious such as unsustainable coastal deforestation
problems that face our watery world. In and agriculture, as well as reforming the
many ways, we know what the solutions management of other stressors from the
are; it is a pathway to action that we now physical destruction of coral reefs to the
need to define and act on, and quickly. management and reduction of overexploi-
In terms of strong action on this issue, it tation of fisheries and other reef uses. Direct
comes down to two major issues. The first action of this type has the potential to build
is that we need to rapidly stabilize global the resilience of coral reefs and to buy im-
conditions, whether they be rising temper- portant time for the global community as it
ature, changing ocean chemistry, or other struggles to rapidly reduce emissions of CO2
aspects of climate. Without stabilization, and other dangerous greenhouse gases.
the ability of organisms and ecosystems to
adapt will be severely compromised, if not
entirely impossible. As has been discussed R E FE R E NCE S
elsewhere, the ability for adaptation to play
a role in minimizing the impacts of a rap- Baker, Andrew C., Peter W. Glynn, and Bernhard Riegl.
2008. “Climate change and coral reef bleaching: An
idly changing climate comes down to the
ecological assessment of long-term impacts, recovery
stabilization of conditions, so that genetic trends and future outlook.” Estuarine Coastal and Shelf Sci-
processes can catch up with the rate of ex- ence 80 (4): 435–471.
ceptional environmental change (Hoegh- Berner, Robert A., and Zavareth Kothavala. 2001. “GEO-
Guldberg 2012). To stabilize the climate CARB III: A revised model of atmospheric CO2 over
64 W H AT C H A NGE S A R E W E OB SE RV I NG?
Phanerozoic time.” American Journal of Science 301 (2): Frieler, K., M. Meinshausen, A. Golly, M. Mengel, K.
182–204. Lebek, S. D. Donner, and O. Hoegh-Guldberg. 2013.
Bongaerts, P., T. C. L. Bridge, D. Kline, P. Muir, C. Wal- “Limiting global warming to 2 degrees C is unlikely
lace, R. Beaman, and O. Hoegh-Guldberg. 2011. “Me- to save most coral reefs.” Nature Climate Change 3 (2):
sophotic coral ecosystems on the walls of Coral Sea 165–170.
atolls.” Coral Reefs 30 no. 2: 335. Gardner, Toby A., Isabelle M. Côté, Jennifer A. Gill,
Bruno, John F., and Elizabeth R. Selig. 2007. “Regional de- Alastair Grant, and Andrew R. Watkinson. 2003.
cline of coral cover in the Indo-Pacific: timing, extent, “Long-term region-wide declines in Caribbean cor-
and subregional comparisons.” PLOS One 2 (8): e711. als.” Science 301 (5635): 958–960.
Burke, Lauretta, Kathleen Reytar, Mark Spalding, and Al- Gattuso, J. P., P. G. Brewer, O. Hoegh-Guldberg, J. A.
lison Perry. 2011. Reefs at Risk Revisited. World Resources Kleypas, H. O. Pörtner, and D. N. Schmidt. 2014.
Institute. “Cross-chapter box on ocean acidification.” In Climate
Caldeira, Ken, and Michael E. Wickett. 2003. “Oceanog- Change 2014: Impacts, Adaptation, and Vulnerability. Part A:
raphy: Anthropogenic carbon and ocean pH.” Nature Global and Sectoral Aspects. Contribution of Working Group II to
425 (6956): 365. the Fifth Assessment Report of the Intergovernmental Panel of Cli-
Cinner, Joshua E., Timothy R. McClanahan, Tim M. Daw, mate Change, ed. C. B. Field, et al., 129–131. Cambridge
Nicholas A. Graham, Joseph Maina, Shaun K. Wilson, University Press.
and Terence P. Hughes. 2009. “Linking social and eco- Glynn, P. W. 1993. “Coral reef bleaching: Ecological per-
logical systems to sustain coral reef fisheries.” Current spectives.” Coral Reef 12 (1): 1–17.
Biology 19 (3): 206–212. Glynn, Peter W. 2012. “Global warming and widespread
Costanza, Robert, Rudolf de Groot, Paul Sutton, Sander coral mortality: Evidence of first coral reef extinc-
van der Ploeg, Sharolyn J. Anderson, Ida Kubiszewski, tions.” In Saving a Million Species, ed. Lee Hannah, 103–
Stephen Farber, and R. Kerry Turner. 2014. “Changes 120. Island Press/Center for Resource Economics.
in the global value of ecosystem services.” Global Envi- Hoegh-Guldberg, Ove. 1999. “Coral bleaching, climate
ronmental Change 2: 152–158. change and the future of the world’s coral reefs.” Ma-
Costello, Mark John, Marta Coll, Roberto Danovaro, rine and Freshwater Research 50 (8): 839–866.
Pat Halpin, Henn Ojaveer, and Patricia Miloslavich. Hoegh-Guldberg, Ove. 2012. “The adaptation of coral
2010. “A census of marine biodiversity knowledge, reefs to climate change: Is the Red Queen being out-
resources, and future challenges.” PLOS One 5 (8). paced?” Scientia Marina 76 (2): 403–408.
Davenport, Coral. 2015. “Nations approve landmark cli- Hoegh-Guldberg, Ove. 2014. “Coral reefs in the Anthro-
mate accord in Paris.” New York Times, December 12. pocene: Persistence or the end of the line?” Geological
De’ath, Glenn, Katharina E. Fabricius, Hugh Sweatman, Society, London, Special Publications 395 (1): 167–183.
and Marji Puotinen. 2012. “The 27-year decline of Hoegh-Guldberg, Ove, and John F. Bruno. 2010. “The
coral cover on the Great Barrier Reef and its causes.” impact of climate change on the world’s marine eco-
Proceedings of the National Academy of Sciences 109 (44): systems.” Science 328 (5985): 1523–1528.
17995–17999. Hoegh-Guldberg, Ove, Peter J. Mumby, Anthony J. Hoo-
De’ath, Glenn, Janice M. Lough, and Katharina E. Fabri- ten, Robert S. Steneck, Paul Greenfield, Edgardo Go-
cius. 2009. “Declining coral calcification on the Great mez, C. Drew Harvell, et al. 2007. “Coral reefs under
Barrier Reef.” Science 323 (5910): 116–119. rapid climate change and ocean acidification.” Science
Donner, Simon D., Thomas R. Knutson, and Michael 318 (5857): 1737–1742.
Oppenheimer. 2007. “Model-based assessment of the Hoegh-Guldberg, O., and T. Ridgway. 2016. “Coral
role of human-induced climate change in the 2005 bleaching comes to the Great Barrier Reef as record-
Caribbean coral bleaching event.” Proceedings of the Na- breaking global temperatures continue.” The Conversa-
tional Academy of Sciences 104 (13): 5483–5488. tion, March 21, 2016.
Donner, Simon D., William J. Skirving, Christopher M. Hoegh-Guldberg, Ove, and G. Jason Smith. 1989. “The
Little, Michael Oppenheimer, and Ove Hoegh-Guld- effect of sudden changes in temperature, light and
berg. 2005. “Global assessment of coral bleaching and salinity on the population density and export of zoo-
required rates of adaptation under climate change.” xanthellae from the reef corals Stylophora pistillata and
Global Change Biology 11 (12): 2251–2265. Seriatopra hystrix.” Journal of Experimental Marine Biology and
Dove, Sophie G., David I. Kline, Olga Pantos, Florent Ecology 129 (3): 279–303.
E. Angly, Gene W. Tyson, and Ove Hoegh-Guldberg. Hoegh-Guldberg, O., et al. 2014. “The ocean.” In Climate
2013. “Future reef decalcification under a business-as- Change 2014: Impacts, Adaptation, and Vulnerability. Part B: Re-
usual CO2 emission scenario.” Proceedings of the National gional Aspects. Contribution of Working Group II to the Fifth As-
Academy of Sciences 110 (38): 15342–15347. sessment Report of the Intergovernmental Panel on Climate Change,
Freiwald, André, Jane Heldge Fosså, Anthony Grehan, ed. V. R. Barros, et al., 1655–1731. Cambridge Uni-
Tony Koslow, and J. Murray Roberts. 2004. Cold-Water versity Press.
Coral Reefs: Out of Sight—No Longer out of Mind. Cambridge: Hönisch, Bärbel, Andy Ridgwell, Daniela N. Schmidt,
UNEP-WCMC. Ellen Thomas, Samantha J. Gibbs, Appy Sluijs, Richard
CORAL REEFS 65
Zeebe, et al. 2012. “The geological record of ocean Pachauri, Rajendra K., Myles R. Allen, Vicente R. Barros,
acidification.” Science 335 (6072): 1058–1063. John Broome, Wolfgang Cramer, Renate Christ, John
Hughes, T. P., B. Schaffelke, and J. Kerry. 2016. “How A. Church, et al. 2014. Climate Change 2014: Synthesis Report.
much coral has died in the Great Barrier Reef’s worst Contribution of Working Groups I, II and III to the Fifth Assessment
bleaching event?” The Conversation, November 29, 2016. Report of the Intergovernmental Panel on Climate Change. IPCC.
IPCC. 2013. Climate Change 2013: The Physical Science Basis. Poloczanska, Elvira S., Christopher J. Brown, William
Contribution of Working Group I to the Fifth Assessment Report J. Sydeman, Wolfgang Kiessling, David S. Schoeman,
of the Intergovernmental Panel on Climate Change. Cambridge Pippa J. Moore, Keith Brander, et al. 2013. “Global im-
University Press. print of climate change on marine life.” Nature Climate
Kleypas, Joan A., Robert W. Buddemeier, David Archer, Change 3 (10): 919–925.
Jean-Pierre Gattuso, Chris Langdon, and Bradley N. Poloczanska, E. S., O. Hoegh-Guldberg, W. Cheung, H.
Opdyke. 1999. “Geochemical consequences of in- O. Pörtner, and M. Burrows. 2014. “Cross-chapter box
creased atmospheric carbon dioxide on coral reefs.” on observed global responses of marine biogeogra-
Science 284, no: 5411: 118–120. phy, abundance, and phenology to climate change.”
Kleypas, Joan A., John W. McManus, and Lambert A. In Climate Change 2014: Impacts, Adaptation, and Vulnerability.
B. Meñez. 1999. “Environmental limits to coral reef Part A: Global and Sectoral Aspects. Contribution of Working Group
development: Where do we draw the line?” American II to the Fifth Assessment Report of the Intergovernmental Panel
Zoologist 39 (1): 146–159. of Climate Change, ed. C. B. Field, et al., 123–127. Cam-
Kroeker, Kristy J., Rebecca L. Kordas, Ryan Crim, Iris E. bridge University Press.
Hendriks, Laura Ramajo, Gerald S. Singh, Carlos M. Pörtner, Hans-Otto, David M. Karl, Philip W. Boyd, Wil-
Duarte, and Jean-Pierre Gattuso. 2013. “Impacts of liam Cheung, Salvador E. Lluch-Cota, Yukihiro Nojiri,
ocean acidification on marine organisms: Quantify- Daniela N. Schmidt, et al. 2014. “Ocean systems.” In
ing sensitivities and interaction with warming.” Global Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part
Change Biology 19 (6): 1884–1896. A: Global and Sectoral Aspects. Contribution of Working Group II
Manzello, Derek P., Joan A. Kleypas, David A. Budd, to the Fifth Assessment Report of the Intergovernmental Panel on
C. Mark Eakin, Peter W. Glynn, and Chris Langdon. Climate Change, 411–484. Cambridge University Press.
2008. “Poorly cemented coral reefs of the eastern Reaka-Kudla, Marjorie L. 1997. “Global biodiversity of
tropical Pacific: Possible insights into reef develop- coral reefs: A comparison with rainforests.” In Biodi-
ment in a high-CO(2) world.” Proceedings of the National versity II: Understanding and Protecting Our Biological Resources,
Academy of Sciences 105 (30): 10450–10455. ed. M. L. Reaka-Kudla, and D. E. Wilson, 551. Joseph
Moberg, Fredrik, and Carl Folke. 1999. “Ecological Henry Press.
goods and services of coral reef ecosystems.” Ecological Schellnhuber, Hans Joachim, Stefan Rahmstorf, and
Economics 29 (2): 215–233. Ricarda Winkelmann. 2016. “Why the right climate
Muscatine, L. 1990. “The role of symbiotic algae in car- target was agreed in Paris.” Nature Climate Change 6 (7):
bon and energy flux in reef corals.” Ecosystems of the 649–653.
World 25: 75–87. Stanley, George D., and Daphne G. Fautin. 2001. “The or-
Muscatine, Leonard, and James W. Porter. 1977. “Reef igins of modern corals.” Science 291 (5510): 1913–1914.
corals: Mutualistic symbioses adapted to nutrient- Veron, J. E. N. 2008. A Reef in Time: The Great Barrier Reef from
poor environments.” Bioscience 27 (7): 454–460. Beginning to End. Harvard University Press.
CHAPTER SIX Biologists have long been interested in the
responses of lineages and biotas to grand
Genetic Signatures changes in climates. Wallace (1880) asso-
ciated the cold and warm cycles of “gla-
of Historical and cial epochs” to patterns of distributional
change and the extinction of mammals in
Contemporary Europe and North America. Adams (1905)
expanded on this theme by envisioning
Responses to whole biotas in North America expanding
their distributions from five ice-age “biotic
Climate Change preserves” (“refugia” in modern terminol-
ogy) as continental glaciers receded. Haffer
BR ETT R. RIDDLE (1969) envisioned avian diversification in
the Amazonian region during Pleistocene
cycles of forest expansion and contraction.
Of course, we now know that climate
change is the “norm” at different tempo-
ral and spatial scales, including multiple
glacial-interglacial cycles over the past 2.5
million years (Ma) and a progressive shift
from a generally warmer to cooler Earth
beginning about 12 mya associated with
Antarctic reglaciation (Zachos et al. 2001).
Extant populations, species, and higher taxa
represent the survivors of a long history of
distributional, diversification, and adaptive
responses to shifting climates—and the sig-
natures of any or all of these histories are
conceptually decipherable in the genetic
variation recorded in different partitions of
a genome.
Whereas the earliest attempts to decipher
genetic variation within and among natu-
ral populations relied on sampling varia-
tion in the nuclear genome with techniques
such as variation in chromosome morphol-
ogy (Dobzhansky et al. 1966) and protein
electrophoresis (e.g., Lewontin and Hubby
1966), the ability to reconstruct the genetic
signatures of historical responses increased
markedly over three decades ago, when
phylogeography was introduced as a disci-
66
GENETIC SIGNATUR E S OF R E SPONSE S TO CLIM ATE CH A NGE 67
pline that bridged a theoretical and empiri- tions on contemporary landscapes, influ-
cal gap between phylogenetics above the enced more by ecological attributes such
species level and population genetics below as habitat connectivity or fragmentation,
it (Avise 2000). The sorts of responses that metapopulation structure, and recent adap-
subsequently have been documented in- tive evolution than by historical dynamics.
clude distributional dynamics (range con- This recognition generated the discipline
traction and expansion, range shifting) and of landscape genetics (Manel et al. 2003)
population subdivision, resulting in restric- and has expanded subsequently with access
tions in gene flow (genetically differenti- to genetic markers that evolve much more
ated populations) or cessation of gene flow quickly than even the organelle markers
(diversification of a single ancestral lineage that motivated the growth of phylogeog-
into two or more evolutionarily distinct lin- raphy (e.g., microsatellites; restriction site
eages). Phylogeography originally sampled associated DNA sequencing, or RADSeq;
genetic variation in natural populations and other single nucleotide polymorphisms
by bypassing the complex mixture of se- [SNPs] methods). It is not uncommon to
lected, neutral, and relatively slowly evolv- recognize that one must understand popu-
ing variation in the nuclear genome in fa- lation structure in terms of both historical
vor of much smaller and simpler organelle and contemporary forces (Epps and Keyg-
genomes—mitochondrial DNA (mtDNA) hobadi 2015; Rissler 2016), yet differences
in animals and fungi and chloroplast DNA between phylogeography and landscape ge-
(cpDNA) in plants (Avise 2000). Organelle netics have not been obvious in all studies,
genomes sampled mostly sex-biased (ma- although attempts have been made to de-
ternal) gene flow, and ancestral lineages lineate the two in terms of spatiotemporal
could be “captured” through interspecific scales (Figure 6.1), optimal genetic markers,
gene flow, but these disadvantages were analytical approaches, and ecological versus
somewhat countered by the advantageous evolutionary arenas (Epps and Keyghobadi
quality of evolving and “fixing” new mu- 2015; Rissler 2016).
tations in populations more rapidly than Phylogeography and landscape genet-
did much of the nuclear genome. Although ics alike have entered new phases with in-
early studies tended to be ad hoc and not corporation of next-generation sequencing
statistically rigorous, changes in these at- (NGS) platforms (Thomson et al. 2010).
tributes elevated the discipline to one that On the one hand, these approaches offer
tested alternative hypotheses (e.g., zero, new opportunities for phylogeographers
one, or multiple Pleistocene refugia) within to add a nuclear DNA component to stud-
a statistically defensible context (Hickerson ies in the form of phylogenomics (DeSalle
et al. 2010). A smaller set of phylogeogra- and Rosenfeld 2013) that matches in reso-
phy studies are concerned with responses lution that of organelle DNA and to fish
of groups of codistributed species, whether out genes involved with divergence and a
they simply occur together in a community history of adaptive evolution (e.g., Brewer
or region or form more intimate biological et al. 2014; Jezkova et al. 2015; Schield et
and ecological associations (e.g., pollinators al. 2015). On the other hand, perhaps the
and plant hosts). These kinds of studies fall most interesting form of contemporary re-
under the umbrella of comparative phylo- sponse, adaptive evolution to current land-
geography (Avise et al. 2016), and they offer scapes and climates, can now be investi-
the potential to track the responses of par- gated at a candidate gene level through the
tial or whole biotas to climate change. emerging potential of landscape genomics
Biologists have recognized that genetic (e.g., Sork et al. 2016). Additionally, several
diversity could be a product of gene flow newer techniques are increasing capacity to
and selection within and among popula- add an ancient DNA component to studies
68 W H AT C H A NGE S A R E W E OB SE RV I NG?
Figure 6.1. A depiction of the generally perceived rela- 2015). I close with several additional obser-
tionship between phylogeography or comparative phy- vations regarding the potential of landscape
logeography and landscape genetics within the over-
arching realm of biogeographic research along three
genomics, perhaps in combination with
scales—temporal, spatial, and organismal. (From Rissler phylogeography, to address the potential of
2016.) populations and species to respond to cli-
mate change through ongoing population-
by broadly sampling across the mitochon- level processes, including adaptive evolu-
drial and nuclear genomes of fecal samples tion. I concentrate primarily on nonmarine
and museum specimens (McCormack et al. biodiversity, although a number of the is-
2016; van der Valk et al. 2017). sues discussed are relevant in the marine
Rissler (2016) argued that with the in- realm as well (e.g., Bowen 2016).
creasing power of genetics to blur lines be-
tween spatiotemporal scales, nonadaptive
versus adaptive variation, and ecological T H E R E L E VA NC E OF A
versus evolutionary contexts, phylogeog- PH Y LO GE O GR A PH IC PE R SPE CTI V E
raphy and landscape genetics represents
a continuum of questions within bioge- Phylogeographic studies increasingly have
ography (Table 6.1). Nevertheless, in the incorporated data from nDNA for good
remainder of this chapter I concentrate theoretical and empirical reasons (e.g., Ed-
largely on the phylogeographic component wards and Bensch 2009)—often along with
of this continuum because it is the approach organelle DNA, for a multiple genome per-
that captures an integrated signature of re- spective. Organelle-DNA-based studies are
sponses to a history of responses to climate still employed frequently, to good advan-
change, both within species and among tage, in many phylogeography studies, for
closely related species—at spatial and tem- reasons including ease of preparation and
poral scales that often are commensurate analysis and relatively low cost, which al-
with delineation of biotic reserves, biodi- lows for the maximization of numbers of
versity hotspots, and large-scale range dy- populations and individuals that can be
namics. This scale is receiving increasing sampled (Riddle 2016); although techni-
recognition as critical for establishing a cal advances continue to lower the cost of
spatial and temporal evolutionary frame- nDNA approaches such as RADSeq (An-
work within which to design studies at drews et al. 2016) and other SNPs methods
the landscape scale (Epps and Keyghobadi (Leaché et al. 2017). Moreover, the signal
GENETIC SIGNATUR E S OF R E SPONSE S TO CLIM ATE CH A NGE 69
Table 6.1. Temporally scaled continua among several topics and questions associated with historical biotic
responses to climate change, addressed with genetic approaches
Time frame
(years before Dominant geo- Biogeographic Analytical Overarching
present) Forcing events graphic processes context realm question
10,000– Multiple glacial- Inter- and Phylogeography Inter- and How do species
1,000,000 interglacial intraspecific intraspecific and assem-
cycles range dynam- phylogenetics blages of species
ics; refugia; become isolated,
community or connected, and
biota assembly- interact at con-
disassembly tact zones; what
patterns of biotic
diversification
result from gla-
cial-interglacial
climate cycles?
1,000–10,000 Holocene Intraspecific range Phylogeography; Intraspecific How do recent
warming- dynamics; landscape phylogenet- landscapes
cooling community genetics ics; popula- interact with
episodes; assembly-disas- tion genetics paleoclimates
anthropogeni- sembly to isolate
cally modified populations and
landscapes facilitate or re-
strict gene flow
among them?
0–1,000 Ongoing warm- Shifting popula- Landscape Population How do current
cool phases; tion structure; genetics genetics and recent land-
anthropogeni- gene flow scapes interact
cally modified dynamics with climate,
climates and etc., to facilitate
landscapes or restrict gene
flow among
populations?
(at least in animal mtDNA) of variation is Within the broader purview of conserva-
usually stronger than it is for most nDNA tion genetics, phylogeography is one com-
data sets, allowing for establishment of a ponent of a triad of past, present, and fu-
first approximation of genetic architecture ture priorities that have different objectives
that can be treated as an hypothesis to be and use different approaches (Bowen 2016).
addressed with nDNA for concordance Under Bowen’s (2016) framework, the past
across genomes, episodes of past hybrid- is explored in a phylogenetic context—how
ization, and so on (Barrowclough and Zink many species or lineages, how different they
2009). However, increasing access to NGS are from one another, and the like (often the
approaches to sampling the nuclear ge- units that conservation biology has used,
nome could motivate the transition of phy- rather contentiously, to delineate evolution-
logeography from predominantly mtDNA arily significant units; Casacci et al. 2014);
and cpDNA to a more inclusive genomic the present is concerned with the ecology
framework in the foreseeable future (Ed- of population genetic variation, herein dis-
wards et al. 2015). cussed as “landscape genetics” (again, often
70 W H AT C H A NGE S A R E W E OB SE RV I NG?
considered in conservation biology as man- the latest transition from glacial to intergla-
agement units); and the future is the realm cial climates (Riddle 2016), often referenc-
of topics such as dispersal potential among ing the latest Quaternary—that is, since the
geographically distributed populations and Late Glacial Maximum, LGM, about 25 kya
range shifting under changing climates, ex- (thousands of years ago). However, a focus
plored within a phylogeographic context. only on this single episode of climate tran-
Although I agree in general with Bowen’s sition, though pronounced, would be an
(2016) perspective, I view the role of phy- oversimplification in considering the influ-
logeography in conservation genetics as ence of historical climate change on patterns
incorporating an historical framework in of distribution, diversification, and adaptive
two ways. First, this is an approach that of- evolution. For example, in a review of com-
ten has revealed “cryptic lineages” that are parative phylogeography in continental bio-
the product of past divergence, unknow- tas (Riddle 2016), a substantial portion of
able prior to phylogeographic-scale studies studies focused on Neogene, or both Neo-
that are then made available for assessment gene and Quaternary time frames, thereby
through phylogenetics analyses (Riddle not only covering that portion of Earth his-
and Hafner 2006). Second, much of em- tory trending from generally warmer to
pirical phylogeographic research has been cooler climates over the past few millions
concerned with the reconstruction of past of years (Zachos et al. 2001) but also within
range dynamics, for example, into and out that expanded time frame, requiring atten-
of refugia, which forms an important basis tion to Neogene geological transformations
for trying to predict responses to future cli- (e.g., plateau and mountain uplift, closure
mate change (Sanchez-Ramirez et al. 2015). of the Isthmus of Panama) that are coupled
How relevant has phylogeography been with large-scale climate changes.
within the context of climate change and Even a relatively short interval of time,
biodiversity research? A topics search of the the Pleistocene-Holocene transition, should
Web of Science database on June 12, 2016, not be interpreted as a single event with sta-
using “phylogeograph*” returned 21,036 ble climates before and after. The Younger
citations, confirming the overall popularity Dryas represented an episode of recooling
of this approach; while a search using both following several thousand years of warm-
“phylogeograph*” and “climate change*” ing, starting about 12.9 kya (Renssen et al.
returned 1,131; another using “phylogeo- 2015) and ending about 11.5 kya. Other
graph*” and “biodiversity” returned 1,475; Holocene anomalies include a cooling
and another using “phylogeograph*,” “bio- event 8.2 kya (Morrill and Jacobsen 2005)
diversity,” and “climate change*” yielded and Mid-Holocene “warm period” in the
235 citations. In the same search sequence Northern Hemisphere about 6 kya (Bartlein
as above, if “phylogeograph*” is replaced et al. 2011). Smaller and more recent events
with “comparative phylogeograph*” the such as the Medieval Warm Period of the
numbers returned are 151, 171, and 36, re- ninth to thirteenth centuries and the Little
spectively, indicating a detectable interest in Ice Age of the fifteenth to nineteenth cen-
the topic at a codistributed lineages scale. turies illustrate the point that lineages and
biotas must frequently respond to vari-
able climates, although responses to these
GE O GR A PH IC R E SPONSE S TO smaller and temporally much more recent
PAST C L IM AT E C H A NGE events likely would be different in kind
and at smaller spatial scales and therefore
A large body of phylogeographic work is are not likely to register through a typical
focused temporally on the responses of lin- phylogeographic approach. Rather, land-
eages and biotas to climate change during scape genetics (Epps and Keyghobadi 2015;
GENETIC SIGNATUR E S OF R E SPONSE S TO CLIM ATE CH A NGE 71
Rissler 2016) might become a more useful include, for example, montane habitat is-
framework for measuring response within lands (Toussaint et al. 2013), land-bridge
these more recent time frames and typically islands (Papadopoulou and Knowles 2015),
smaller spatial scales. and deserts (Riddle and Hafner 2006). In-
Frequently, populations and species have deed, the same contact zone investigated in
avoided extinction through changes in geo- detail for western diamondback rattlesnake
graphic distribution that tracked shifting dynamics is also a point of contact for multi-
habitats. Often, this sort of geographic dy- ple arid-adapted species (Riddle 2016), sug-
namic is envisioned as a range contraction gesting a biota-wide pattern of responses to
into a refugial phase (Haffer 1969; Arenas historical climate change with the potential
et al. 2012), followed by range expansion as to elucidate taxonomic, phylogenetic, and
preferred habitats expand within a different ecological attributes contributing to how a
climatic regime (Hewitt 2004; Mona et al. particular lineage has responded in the past
2014). Should an ancestral lineage be sub- (Pyron and Burbrink 2007).
divided into more than a single refuge, the Nevertheless, the most immediate im-
microevolutionary processes of genetic drift pacts from current and future climate
and selection could act as drivers of diver- change likely will occur over shorter time
sification, from the production of genetic frames than generally thought to be neces-
structure among populations to the genera- sary to foster generation of new evolution-
tion of evolutionarily distinct lineages and ary lineages and species. Therefore, it seems
perhaps even new species (Damasceno et appropriate to emphasize the lessons from
al. 2014). Genes may be exchanged again history that focus more directly on impacts
once previously isolated populations come from the redistribution of populations,
into contact at contact zones (if envisioned species, and assemblages of species on the
as happening between multiple taxa, called landscape (range contraction, perhaps into
suture zones; Remington 1968). Contact refugia, and expansion; range shifting).
zones are places that have been investigated Indeed, these sorts of responses have been
for evidence that complete reproductive popular topics for investigation—a topics
isolation was established during refugial search of the Web of Science database on
isolation. For example, Schield et al. (2015) June 28, 2016, using “phylogeograph*,”
used a combination of mtDNA and nDNA “pleistocene,” and “refug*” returned 1,250
RADSeq in the western diamondback rat- studies. A search replacing the first term
tlesnake (Crotalus atrox) to corroborate earlier with “comparative phylogeograph*” re-
findings that populations were historically turned 185 studies, suggesting interest in
isolated into eastern and western popula- the question of whether codistributed lin-
tions long enough to be recognizable as eages respond in a concerted fashion by
distinct mtDNA evolutionary lineages. By sharing refugia (Hewitt 2004), which does
contrast, whereas the nDNA data supported not always occur in a simple way (Stewart
an episode of historical isolation, they dem- et al. 2010; Pelletier et al. 2015).
onstrated ongoing gene flow following sec- Variants of this common theme are pos-
ondary contact. Should such a process of tulated refugial distribution during the Last
refugial isolation or range expansion occur Interglacial (LIG) rather than glacial (LGM)
across multiple glacial-interglacial cycles, stage, predicted by habitat preferences
it has the potential to be Pleistocene “spe- (Stewart et al. 2010; Latinne et al. 2015);
cies pumps,” generators of new biodiversity consequences of microrefugia on popula-
detectable at the level of distinct evolution- tion structure, diversification, and range
ary lineages, an idea that harkens back to dynamics under climate change (Hannah
the tropical forest refugia model of Haffer et al. 2014; Mee et al. 2014); relative dis-
(1969) but was subsequently expanded to tributional stasis during the LGM (Jezkova
72 W H AT C H A NGE S A R E W E OB SE RV I NG?
et al. 2011); and predictions that stable re- and invertebrates into the western Great Ba-
fugial areas represent genetic diversity sin, United States, from historically late gla-
hotspots, perhaps worthy of focal attention cial “stable” (or refugial) populations to the
as high-priority conservation management south in the Mojave Desert. They contrasted
areas (Wood et al. 2013). Most of the recent mtDNA and RADseq sequencing to produce
such studies employ a species distribution nDNA SNP data sets for two codistributed
modeling (SDM) procedure to create an kangaroo rat species, one a sand substrate
independent model of predicted range dy- specialist with patchy distributions (Dipodo-
namics that is then available for evaluation mys deserti) and the other a substrate gener-
through phylogeographic analysis. For ex- alist with more connected populations (D.
ample, Pelletier et al. (2015) used hindcast merriami). On the basis of well-established
models of predicted species distribution at models that predicted that there would be
21 kya in three congeneric species of sala- a loss of genetic diversity in recently ex-
mander from northwestern North America panded populations (Hewitt 2004), with
to support an inference that the geographic loss more extreme in D. deserti due to smaller
range of only one of the three species likely effective populations with less gene flow
contracted into multiple glacial-age refugia, between them. First, SDM modeling sup-
driving evolution of distinct lineages detect- ported the prediction that the ranges of
able genetically among extant populations. both species were contracted into a south-
Scaling up of such studies could be used at a ern (Mojave Desert) distribution during the
codistributed species scale to identify range LGM (perhaps not surprising given that
dynamics associated with putative biodi- much of their expanded ranges were part
versity hotspots. of a large Pleistocene lake at that time). Sec-
A generalized consensus on historical ond, the more rapid and extreme loss of ge-
distributional responses to climate change netic diversity in the substrate specialist was
has not appeared, but it is entirely possible confirmed for mtDNA. Third, loss of nDNA
that each “system” (e.g., lineage or codis- variation mirrored mtDNA qualitatively but
tributed lineages, biome, landscape, biogeo- was quantitatively not as extreme, perhaps
graphic region) requires independent eval- because either effective population sizes of
uation to establish a baseline for predicting organelle DNA are smaller or because of
future distributional responses. Neverthe- sex-biased dispersal. Thus, organelle DNA
less, refined analytical tools are demonstrat- may be a more sensitive marker recording
ing promise in the ongoing growth of ap- range expansions out of Pleistocene refugia,
proaches to modeling the historical ranges but it may not accurately reflect adaptive
of intraspecific populations and locations of niche evolution in the nuclear genome.
refugia (Rosauer et al. 2015).
In what ways does ecology matter in
determining a particular response to past N IC H E E VOLU T ION U N DE R
climate change and in explaining idiosyn- C L I M AT E C H A NGE
cratic responses to the same underlying
climatic transitions? Angert and Schemske Do niches evolve as populations and spe-
(2005) addressed this question experimen- cies respond to climate change? Tradition-
tally with reciprocal transplants of two ally, biogeographers and evolutionary bi-
species of monkey flower (Mimulus cardinalis ologists have framed this question through
and M. lewisii) and presented evidence for re- the dichotomy of “stay at home and evolve
duced fitness in each species at range mar- new niches” or “move and retain niches”
gins. Jezkova et al. (2015) took advantage of as habitats shift across space (Weeks et al.
a postulated post-LGM route of northward 2014). Further, much thinking about cli-
range expansion of arid-adapted vertebrates matically driven niche evolution has fo-
GENETIC SIGNATUR E S OF R E SPONSE S TO CLIM ATE CH A NGE 73
cused on an interspecific or higher taxon and respiratory conditions along elevation
scale, often at time frames deeper than gradients (Lemay et al. 2013).
the latest Pleistocene to Holocene shifts
(Cooney et al. 2016). However, of greater
relevance to the issue of responses to ongo- R E SPONSE S TO CON T E M POR A RY
ing and future climate change would be a C L I M AT E C H A NGE
focus on niche evolution, or stasis, as in-
traspecific populations shift distributions Climate change during the Anthropocene
(Wüest et al. 2015; Lancaster et al. 2016). A is the broad time frame of focal interest
relevant question here is this: how rapidly in attempting to predict consequences of
might we expect populations to undergo human-mediated activities for biodiversity.
adaptive evolutionary responses to newly Yet developing an empirical context for the
encountered climatic regimes, whether role of contemporary landscapes on genetic
they stay at home or move? Modeling can structure within and between populations
contribute to establishing a theoretical can be challenging owing to the inherent
framework, exploring parameters includ- lag between the time of a relatively recent
ing landscape structure (smooth gradient but historical change in landscape con-
vs. patchy), genetic architecture (many loci figuration—either through climate change
with small phenotypic effect vs. few with alone or with some combination of physi-
large effect), average dispersal distances, cal habitat alteration. Given the recency of
and interactions among all these variables many of these changes, extant measures of
(Schiffers et al. 2014). genetic diversity across a landscape may not
Empirical evidence will continue to reflect current configurations but rather re-
come from genomic studies that can iden- tain a time-lag signature of past conditions.
tify genes under selection, but if focused on One important emphasis in the ongoing
the last major glacial (LGM) to interglacial development of a landscape genetics that
transition, we may not expect to have suf- has relevance for conservation management
ficient resolution in traditional Sanger DNA (Keller et al. 2015) is the development of
sequencing techniques to be able to find ways to sort historical from contemporary
that gene or suite of genes driven to muta- influences on genetic structure and gene
tion and fixation of new genotypes under flow, and the building of a cross-referenc-
selection across such a short time frame. ing conceptual and analytical environment
New opportunities arise, however, through between phylogeography and landscape ge-
the use of NGS approaches as applied to netics will be important for doing so (Epps
both DNA and RNA (transcriptome) as- and Keyghobadi 2015; Rissler 2016).
sessments of genome evolution (Brewer et As landscape genetics progresses into an
al. 2014). Focusing on a phylogeographic NGS genomics framework, opportunities
context, larger amounts of genome-wide to investigate more explicitly the relation-
data than provided through popular sub- ship between adaptive evolution and local
sampling procedures such as RADSeq might selection regimes should increase. Sork et
be required to detect selection in natural al. (2016) summarize work on climate-as-
populations, particularly if linkage disequi- sociated candidate genes in the California
librium is low (Edwards et al. 2015). One endemic oak Quercus lobata, including those
such study identified SNPs from transcrip- associated with bud burst and flowering,
tome data and the mtDNA NADH gene in growth, osmotic stress, and temperature
the American pika (Ochotona princeps), a focal stress; 10 out of 40 candidate genes indi-
taxon for investigating the effects of climate cated spatially divergent selection. Exten-
change in western North America, suggest- sions of this work should include much
ing local adaptation to different thermal larger samples of SNPs drawn from NGS
74 W H AT C H A NGE S A R E W E OB SE RV I NG?
approaches to sampling the genome (Sork the elevation ranges of Mimulus cardinalis and M. lewisii.”
et al. 2016). Evolution 59 (8): 1671–1684.
Arenas Miguel, Nicholas Ray, Mattias Currat, and Lau-
rent Excoffier. 2011. “Consequences of range contrac-
tions and range shifts on molecular diversity.” Molecu-
L E S S ONS : E X TINCTION R E SI L I E NC E lar Biology and Evolution 29 (1): 207–218.
A N D E X TINCTION R ISK Avise, John C. 2000. Phylogeography: The History and Formation
of Species. Harvard University Press.
Avise, John C., Brian W. Bowen, and Francisco J. Ayala.
An important issue with regard to future 2016. “In light of evolution X: Comparative phylo-
biodiversity is whether and with what de- geography.” Proceedings of the National Academy of Sciences
gree of accuracy can inferences about his- 113 (29): 7957–7961.
torical and contemporary responses to cli- Barrowclough, George F., and Robert M. Zink. 2009.
mate change be used as a guide to predict “Funds enough, and time: mtDNA, nuDNA and the
discovery of divergence.” Molecular Ecology 18 (14):
responses under rapidly changing climates. 2934–2936.
Herein, I provided an overview of several Bartlein, P. J., S. P. Harrison, S. Brewer, S. Connor, B. A.
ways in which phylogeographers have de- S. Davis, K. Gajewski, J. Guiot, T. I. Harrison-Prentice,
veloped capacity to decipher genetic sig- A. Henderson, O. Peyron, I. C. Prentice, M. Scholze,
natures of what can be called extinction re- H. Seppa, B. Shuman, S. Sugita, R. S. Thompson, A.
E. Viau, J. Williams, and H. Wu. 2011. “Pollen-based
silience—those distributional, demographic, continental climate reconstructions at 6 and 21 ka: A
and adaptive properties that have buffered global synthesis.” Climate Dynamics 37 (3–4): 775–802.
populations, species, and biotas from ex- Bowen, Brian W. 2016. “The three domains of conserva-
tinction within time frames that have wit- tion genetics: Case histories from Hawaiian waters.”
nessed profoundly changing climates. What Journal of Heredity 107 (4): 309–317.
Brewer, Michael S., Darko D. Cotoras, Peter J. P. Croucher,
is required is the ongoing development of and Rosemary G. Gillespie. 2014. “New sequencing
connections between these insights and an technologies, the development of genomics tools, and
emerging framework for being able to pre- their applications in evolutionary arachnology.” Journal
dict extinction risk as populations, species, of Arachnology 42 (1): 1–15.
and biotas are required to track changing Casacci, L. P., F. Barbero, and E. Balletto. 2014. “The
‘evolutionarily significant unit’ concept and its ap-
climates into the future. A combined effort plicability in biological conservation.” Italian Journal of
that incorporates, for example, NGS and Zoology 81 (2): 182–193.
more traditional genomics, SDMs, explicit Cooney, Christopher R., Nathalie Seddon, and Joseph A.
linkages between landscapes and genetics, Tobias. 2016. “Widespread correlations between cli-
and connections between genotypic and matic niche evolution and species diversification in
birds.” Journal of Animal Ecology 85 (4): 869–878.
phenotypic evolution in an explicit climatic Damasceno, Roberto, Maria L. Strangas, Ana C. Car-
niche context—at population, species, and naval, Miguel T. Rodrigues, and Craig Moritz. 2014.
biotic scales—promises to advance under- “Revisiting the vanishing refuge model of diversifica-
standing of the future of biodiversity under tion.” Frontiers in Genetics 5: 1–12.
climate change. DeSalle, Rob, and Jeffrey A. Rosenfeld. 2013. Phyloge-
nomics: A Primer. Garland Science.
Dobzhansky, T., W. W. Anderson, and O. Pavlovsky.
1966. “Genetics of natural populations: XXXVIII.
R E FE R E NCE S Continuity and change in populations of Drosophila
pseudoobscura in western United States.” Evolution 20 (3):
Adams, Chas C. 1905. “The postglacial dispersal of the 418–427.
North American biota.” Biological Bulletin 9 (1): 53–71. Edwards, Scott, and Staffan Bensch. 2009. “Looking
Andrews, Kimberley R., Jeffrey M. Good, Michael R. forwards or looking backwards in avian phylogeog-
Miller, Gordon Luikart, and Paul A. Hohenlohe. 2016. raphy? A comment on Zink and Barrowclough 2008.”
“Harnessing the power of RADSeq for ecological Molecular Ecology 18 (14): 2930–2933.
and evolutionary genomics.” Nature Reviews Genetics 17: Edwards, Scott V., Allison J. Shultz, and Shane Campbell-
81–92. Staton. 2015. “Next-generation sequencing and the
Angert, A. L., and D. W. Schemske. 2005. “The evolution expanding domain of phylogeography.” Folia Zoologica
of species’ distributions: Reciprocal transplants across 64: 187–206.
GENETIC SIGNATUR E S OF R E SPONSE S TO CLIM ATE CH A NGE 75
Epps, Clinton W., and Nusha Keyghobadi. 2015. “Land- elements from bird museum specimens.” Molecular
scape genetics in a changing world: Disentangling Ecology Resources 16: 1189–1203.
historical and contemporary influences and inferring Mee, Jonathan A., and Jean-Sébastien Moore. 2014. “The
change.” Molecular Ecology 24 (24): 6021–6040. ecological and evolutionary implications of microre-
Haffer, Jürgen. 1969. “Speciation in Amazonian forest fugia.” Journal of Biogeography 41 (5): 837–841.
birds.” Science 165 (3889): 131–137. Manel, Stéphanie, Michael K. Schwartz, Gordon Luikart,
Hannah, Lee, Lorraine Flint, Alexandra D. Syphard, Max and Pierre Taberlet. 2003. “Landscape genetics: Com-
A. Moritz, Lauren B. Buckley, and Ian M. McCullough. bining landscape ecology and population genetics.”
2014. “Fine-grain modeling of species’ response to Trends in Ecology & Evolution 18 (4): 189–197.
climate change: Holdouts, stepping-stones, and mi- Mona, Stefano, Nicolas Ray, Miguel Arenas, and Lau-
crorefugia.” Trends in Ecology & Evolution 29 (7): 390–397. rent Excoffier. 2014. “Genetic consequences of habitat
Hewitt, G. M. 2004. “Genetic consequences of climatic fragmentation during a range expansion.” Heredity 112
oscillations in the Quaternary.” Philosophical Transactions (3): 291–299.
of the Royal Society of London Series B—Biological Sciences 359 Morrill, Carrie, and Robert M. Jacobsen. 2005. “How
(1442): 183–195. widespread were climate anomalies 8200 years ago?”
Hickerson, M. J., B. C. Carstens, J. Cavender-Bares, K. A. Geophysical Research Letters 32 (19): L19701.
Crandall, C. H. Graham, J. B. Johnson, L. Rissler, P. F. Papadopoulou, Anna, and L. Lacey Knowles. 2015. “Ge-
Victoriano, and A. D. Yoder. 2010. “Phylogeography’s nomic tests of the species-pump hypothesis: Recent
past, present, and future: 10 years after.” Molecular Phy- island connectivity cycles drive population diver-
logenetics and Evolution 54 (1): 291–301. gence but not speciation in Caribbean crickets across
Jezkova, Tereza, Viktoria Olah-Hemmings, and Brett R. the Virgin Islands.” Evolution 69 (6): 1501–1517.
Riddle. 2011. “Niche shifting in response to warm- Pelletier, Tara A., Charlie Crisafulli, Steve Wagner,
ing climate after the last glacial maximum: inference Amanda J. Zellmer, and Bryan C. Carstens. 2015.
from genetic data and niche assessments in the chisel- “Historical species distribution models predict species
toothed kangaroo rat (Dipodomys microps).” Global Change limits in western Plethodon salamanders.” Systematic
Biology 17 (11): 3486–3502. Biology 64 (6): 909–925.
Jezkova, Tereza, Brett R. Riddle, Daren C. Card, Drew R. Pyron, R. Alexander, and Frank T. Burbrink. 2007. “Hard
Schield, Mallory E. Eckstut, and Todd A. Castoe. 2015. and soft allopatry: Physically and ecologically medi-
“Genetic consequences of postglacial range expansion ated modes of geographic speciation.” Journal of Biogeog-
in two codistributed rodents (genus Dipodomys) depend raphy 37 (10): 2005–2015.
on ecology and genetic locus.” Molecular Ecology 24 (1): Remington, Charles L. 1968. “Suture-zones of hybrid in-
83–97. teraction between recently joined biotas.” In Evolution-
Keller, Daniela, Rolf Holderegger, Maarten J. van Strien, ary Biology, ed. T. Dobzhansky, M. K. Hecht, and W. C.
and Janine Bolliger. 2015. “How to make landscape Steere, 321–428. New York: Plenum Press.
genetics beneficial for conservation management?” Renssen, Hans, Aurélien Mairesse, Hugues Goosse,
Conservation Genetics 16 (3): 503–512. Pierre Mathiot, Oliver Heiri, Didier M. Roche, Kerim
Lancaster, Lesley T., Rachael Y. Dudaniec, Pallavi Chau- H. Nisancioglu, and Paul J. Valdes. 2015. “Multiple
han, Maren Wellenreuther, Erik I. Svensson, and causes of the Younger Dryas cold period.” Nature Geosci-
Bengt Hansson. 2016. “Gene expression under ther- ence 8 (12): 946–980.
mal stress varies across a geographical range expan- Riddle, Brett R. 2016. “Comparative phylogeography
sion front.” Molecular Ecology 25 (5): 1141–1156. clarifies the complexity and problems that drove A. R.
Leaché, Adam D., and Jamie R. Oaks. 2017. “The util- Wallace to favor islands.” Proceedings of the National Acad-
ity of single nucleotide polymorphism (SNP) data in emy of Sciences 113 (29): 7970–7977.
phylogenetics.” Annual Review of Ecology, Evolution, and Sys- Riddle, Brett R., and David J. Hafner. 2006. “A step-wise
tematics 48: 69–84. approach to integrating phylogeographic and phylo-
Lemay, Matthew A., Philippe Henry, Clayton T. Lamb, genetic biogeographic perspectives on the history of
Kelsey M. Robson, and Michael A. Russello. 2013. a core North American warm deserts biota.” Journal of
“Novel genomic resources for a climate change sensi- Arid Environments 66 (3): 435–461.
tive mammal: characterization of the American pika Rissler, Leslie J. 2016. “Union of phylogeography and
transcriptome.” BMC Genomics 14 (1): 311. https://doi landscape genetics.” Proceedings of the National Academy of
.org/10.1186/1471-2164-14-311. Sciences 113 (29): 8079–8086.
Lewontin, Richard C., and Jack L. Hubby. 1966. “A mo- Rosauer, Dan F., Renee A. Catullo, Jeremy VanDerWal,
lecular approach to the study of genic heterozygosity Adnan Moussalli, and Craig Moritz. 2015. “Lineage
in natural populations. II. Amount of variation and range estimation method reveals fine-scale endemism
degree of heterozygosity in natural populations of linked to Pleistocene stability in Australian rainforest
Drosophila pseudoobscura.” Genetics 54 (2): 595–609. herpetofauna.” PLOS One 10 (5): e0126274.
McCormack, John E., Whitney L. E. Tsai, and Brant C. Schield, Drew R., Daren C. Card, Richard H. Adams, Ter-
Faircloth. 2016. “Sequence capture of ultraconserved eza Jezkova, Jacobo Reyes-Velasco, F. Nicole Proctor,
76 W H AT C H A NGE S A R E W E OB SE RV I NG?
Carol L. Spencer, Hans-Werner Herrmann, Stephen P. Van der Valk, Tom, Frida Lona Durazo, Love Dalén, and
Mackessy, and Todd A. Castoe. 2015. “Incipient spe- Katerina Guschanski. 2017. “Whole mitochondrial
ciation with biased gene flow between two lineages of genome capture from faecal samples and museum-
the western diamondback rattlesnake (Crotalus atrox).” preserved specimens.” Molecular Ecology Resources 17:
Molecular Phylogenetics and Evolution 83: 213–223. e111–e121.
Schiffers, Katja, Frank M. Schurr, Justin M. J. Travis, Anne Wallace, A. R. 1880. Island Life, or the Phenomena and Causes of
Duputié, Vincent M. Eckhart, Sébastien Lavergne, Insular Faunas and Floras. Macmillan.
Greg McInerny, et al. 2014. “Landscape structure and Weeks, Andrea, Felipe Zapata, Susan K. Pell, Douglas C.
genetic architecture jointly impact rates of niche evo- Daly, John D. Mitchell, and Paul V. A. Fine. 2014. “To
lution.” Ecography 37 (12): 1218–1229. move or to evolve: Contrasting patterns of intercon-
Sork, Victoria L., Kevin Squire, Paul F. Gugger, Stephanie tinental connectivity and climatic niche evolution in
E. Steele, Eric D. Levy, and Andrew J. Eckert. 2016. ‘Terebinthaceae’ (Anacardiaceae and Burseraceae).”
“Landscape genomic analysis of candidate genes for Frontiers in Genetics 5: 1–20.
climate adaptation in a California endemic oak, Quercus Wood, Dustin A., Amy G. Vandergast, Kelly R. Barr, Rich
lobata.” American Journal of Botany 103 (1): 33–46. D. Inman, Todd C. Esque, Kenneth E. Nussear, and
Stewart, John R., Adrian M. Lister, Ian Barnes, and Love Robert N. Fisher. 2013. “Comparative phylogeogra-
Dalen. 2010. “Refugia revisited: Individualistic re- phy reveals deep lineages and regional evolutionary
sponses of species in space and time.” Proceedings of the hotspots in the Mojave and Sonoran Deserts.” Diversity
Royal Society B—Biological Sciences 277 (1682): 661–671. and Distributions 19 (7): 722–737.
Thomson, Robert C., Ian J. Wang, and Jarrett R. Johnson. Wüest, Rafael O., Alexandre Antonelli, Niklaus E. Zim-
“Genome-enabled development of DNA markers for mermann, and H. Peter Linder. 2015. “Available cli-
ecology, evolution and conservation.” Molecular Ecology mate regimes drive niche diversification during range
19 (11): 2184–2195. expansion.” American Naturalist 185 (5): 640–652.
Toussaint, Emmanuel F. A., Katayo Sagata, Suriani Sur- Zachos, James, Mark Pagani, Lisa Sloan, Ellen Thomas,
bakti, Lars Hendrich, and Michael Balke. 2013. “Aus- and Katharina Billups. 2001. “Trends, rhythms, and
tralasian sky islands act as a diversity pump facilitating aberrations in global climate 65 Ma to present.” Science
peripheral speciation and complex reversal from nar- 292 (5517): 686–693.
row endemic to widespread ecological supertramp.”
Ecology and Evolution 3 (4): 1031–1049.
CASE STUDY 2
Natural populations of salmon (Oncorhychus and Salmo) are widely distributed throughout the
Northern Hemisphere. In addition to their ecological, social, and cultural importance, a
number of Atlantic and Pacific salmon stocks support significant commercial, recreational,
and Aboriginal (First Nation) fisheries. Consequently, reasonable long-term records of catch
or abundance exist for many salmon populations, and these provide convincing evidence
that most populations have fluctuated between periods of high and low abundance for hun-
dreds and perhaps thousands of years. Although some geographic differences are evident,
the widespread synchronous changes in salmon abundance between periods of low and
high productivity appear to be closely associated with regional or global changes in climate
(Noakes and Beamish 2011). However, each species has exhibited distinct responses to cli-
mate change, with some species thriving while others struggle to survive and avoid extinc-
tion within the same climate regime. This apparent paradox highlights the unique nature
of salmon, as they are available in almost unlimited numbers through aquaculture or via
hatchery production while at the same time listed as an endangered species within the same
geographic region or even within the same river system. The current state of global warm-
ing is no different, with some species and stocks of salmon in the North Pacific exhibiting
very high abundance and other stocks threatened with extinction.
77
78 W H AT C H A NGE S A R E W E OB SE RV I NG?
been affected differently by climate change. could influence production for these spe-
Pink (O. gorbuscha) and chum (O. keta) salmon cies, the large number of hatchery fish pro-
spend relatively little time in freshwater, duced and released for these species should
and wild populations of each are heavily (and is argued or assumed to) compensate
supplemented by extensive hatchery pro- for any such losses. There is some evidence
grams throughout the Pacific. Coho (O. to support a decrease in marine survival for
kisutch), Chinook (O. tshawytscha), and Atlantic these species as the reason for their decline
(S. salar) typically spend two or more years in abundance, although factors associated
in their natal rivers or streams (or hatch- with the fitness and survival of hatchery-
ery) before migrating to the ocean, where produced fish may also contribute to the
they will rear for another year or more be- observed decline.
fore returning to spawn. Sockeye (O. nerka)
typically spend two years in rearing lakes
or streams before migrating to sea to live DISC US SION
for another two years. While spawning
channels are occasionally used to enhance Some salmon species and populations (par-
sockeye production, there are very few ticularly more northerly stocks) have thrived
hatcheries releasing artificially reared sock- as a result of the recent changes in climate.
eye, which is a significant difference from
the other commercially harvested salmon
species. Figure CS2.1. All-nation abundance of sockeye, pink, and
There are many factors affecting the sur- chum salmon in the North Pacific recent catches are
nearly 1 million tons. Prefishery abundance of North
vival and growth of salmon. A substantial American 2SW (sea winter) Atlantic salmon (solid
increase in Pacific salmon production oc- line) and Chinook salmon catches in the Strait of Geor-
curred in 1977, coincident with a shift in gia, British Columbia. Atlantic salmon abundance and
regional climate indexes reflective of higher Strait of Georgia Chinook salmon catches have each de-
productivity (Figure CS2.1). The increase creased from several hundred thousand fish in the late
1970s to about 100,000 and 25,000 fish, respectively,
was especially apparent for pink and chum in 2013. Sources: ICES (2007) Report of the Working
salmon, although some of the increase Group on North Atlantic Salmon (WGNAS), 11–20 April
could be attributed to large-scale enhance- 2007, Copenhagen, Denmark. ICES CM 2007/ACFM-13,
ment efforts for these species. The increase 378p and the Pacific Salmon Commission (2014) Joint
for sockeye was more modest, with larger Chinook Technical Committee Annual Report of Catch
and Escapement for 2013, TCCHINOOK (14)-2: 239p.
increases in production associated with North Pacific Anadromous Fish Commission, http://
more northerly stocks. Production for these www.npafc.org/new/science_statistics.html. (Courtesy
three species has remained at high levels of Donald Noakes.)
(nearly 1 million tons) since 1977 (per-
haps due to hatchery production for pink
and chum), although sockeye salmon pro-
duction has decreased since the early 1990s
(Figure CS2.1). The same is not true for
Atlantic, Chinook, and Coho salmon. The
abundance (and/or survival) of Atlantic and
Chinook has declined almost steadily since
the late 1970s, from about 900,000 Atlan-
tic and 700,000 Chinook to about 100,000
Atlantic and 25,000 Chinook salmon (Fig-
ure CS2.1), despite very significant salmon
enhancement efforts, including the use
of hatchery fish. While freshwater factors
CLIM ATE CH A NGE A ND SA L MON POPUL ATIONS 79
Others have undergone very significant important species worthy of protection for
declines and remain at low levels of abun- future generations.
dance. With our limited understanding of
complex ecosystem-level changes, a more
careful approach to fisheries management R E FE R E NCE
and conservation will be required in the
future. There is a higher public expectation Noakes, Donald J., and Richard J. Beamish. 2011. “Shift-
ing the balance: Towards sustainable salmon popula-
with respect to salmon, because important
tions and fisheries of the future.” In Sustainable Fisheries:
stocks are often located in close proxim- Multi-Level Approaches to a Global Problem, ed. W. W. Taylor,
ity to large population centers. Salmon are A. J. Lynch, and M. G. Schechter, 23–50. Bethesda,
an iconic and ecological and economically MD: American Fisheries Society.
CHAPTER SEVEN I N T RODUCTION
80
R APID BROAD-SCALE ECOSYSTEM CHANGE S 81
affected region. Rapid broad-scale changes cate that up to one-half of the global land
triggered by climate can include megafires, area could be highly vulnerable to biome
drought-triggered tree die-off, and associ- change (Gonzalez et al. 2010; Settele et al.
ated pest and pathogen outbreaks (Logan 2014). Projected vegetation changes suggest
et al. 2003; Breshears et al. 2005; Safranyik the following: the temperate mixed forest
et al. 2007; Berner et al. 2017), and hurri- biome and the boreal conifer forest biome
canes and wind-throw events (IPCC 2012, show the highest vulnerability as a fraction
2014). These rapid broad-scale changes can of total biome area; the tundra and alpine
rapidly alter other factors such as microcli- biome and the boreal conifer forest biome
mate (Royer et al. 2011; Villegas et al. 2017; are most vulnerable in total land area; and
Zemp et al. 2017), which in turn can af- tropical evergreen broadleaf forest and des-
fect numerous other species and associated ert biomes show the lowest vulnerability in
biodiversity. Many examples of broad-scale most cases (Gonzalez et al. 2010).
changes are documented in the paleoecol- Many types of rapid broad-scale changes
ogy literature (Settele et al. 2014), although of sufficient magnitude to affect biome
the temporal resolution at which those boundaries have already occurred recently
events can be resolved is relatively coarse across the globe (Figure 7.1; Gonzalez et
(often centuries or longer). Contemporary al. 2010; Settele et al. 2014). For terres-
events have highlighted that broad-scale trial ecosystems, the combined effects of
changes can occur rapidly (years or less; drought and heat and their associated ef-
Figure 7.1; see Breshears et al. 2005, Gonza- fects on pests and pathogens (e.g., Logan
lez et al. 2010; Stephens et al. 2014; Settele et et al. 2003) have resulted in some of the
al. 2014; Hartmann et al. 2018). These rapid most pronounced broad-scale ecosystem
broad-scale changes will have important changes. Regional drought, which periodi-
consequences for biodiversity beyond the cally affects most systems, under the hotter
more commonly considered direct impacts conditions associated with climate change
of climate change (e.g., species physiology, (together referred to as “global-change-type
phenology, and distribution; see Chapters drought” [Breshears et al. 2005] or more
1, 3, and 4) or indirect effects, by which simply “hotter drought” [Allen et al. 2015]),
changes in one species can also affect other can result in increased plant water stress
species (Cahill et al. 2013) up to the point that leads to widespread mortality, par-
of causing species coextinctions (Koh et al. ticularly for trees (Allen et al. 2010; Figure
2004). The focus of this chapter is to alert 7.1). The time to mortality during drought
readers to recent and projected rapid eco- for tree species continues to increase with
system changes that are expected to have temperature, with lethal drought events be-
potential consequences for biodiversity at coming ever more frequent (Adams et al.
ecosystem, landscape, and regional scales. 2017). Increased water stress in trees, espe-
cially in dominant and codominant species,
can result in greater vulnerability to mega-
E X A M PL E SYST E MS W IT H fires (Millar and Stephenson 2015), as well
V U L N E R A BIL IT Y TO R A PI D BROA D - as increasing probability of pest and patho-
SC A L E C H A NGE gen outbreaks (e.g., bark beetles; Logan et
al. 2003) and of mortality directly from the
Climate change is already shifting the dis- drought itself (Breshears et al. 2005; Allen
tribution of biomes across elevations and et al. 2015). Such regional-scale droughts
latitudes throughout the world, especially are of particular concern in the Amazon,
in temperate, tropical, and boreal areas where tree die-off and fire feedbacks, if
(IPCC 2014). Predicted changes in climate sufficient in magnitude, could have pro-
and modeled vegetation projections indi- found impacts on biodiversity (Brando et
82 W H AT C H A NGE S A R E W E OB SE RV I NG?
Figure 7.1. Climate-driven projected changes in vegeta- fog-belt locations and characteristics, put-
tion often are represented as maps of spatial change (A, ting fog-dependent species at risk and leav-
B). Implicit and potentially overlooked in such presen-
tations is the underlying widespread tree mortality (C,
ing them with limited migration options
D) that necessarily would occur with many such pro- (Foster 2001). Warming is altering funda-
jected changes to produce biome-level shifts in vegeta- mental structural characteristics in boreal
tion. Left panels of projected vegetation shifts are from and tundra systems, where loss of perma-
Gonzalez et al. (2010), using the MC1 dynamic global frost changes ecosystem structure (Settele
vegetation model. (A) Modeled potential vegetation un-
der observed 1961–1990 climate. (B) Modeled potential
et al. 2014). Presented here are examples of
vegetation under projected 2071–2100 climate, where systems that are vulnerable to rapid broad-
any of nine climate scenario combinations of general scale changes triggered by climate change.
circulation model and emissions pathway drives pro-
jected vegetation change. Displayed biomes, in panels A
and B, from poles to equator: ice (IC), tundra and al- Evergreen Temperate, Semiarid,
pine (UA), boreal conifer forest (BC), temperate conifer Coniferous Woodlands
forest (TC), temperate broadleaf forest (TB), temperate
mixed forest (TM), temperate shrubland (TS), temper- Piñon pine (Pinus edulis) mortality occurred
ate grassland (TG), desert (DE), tropical grassland (RG), across the US Southwest in response to a
tropical woodland (RW), tropical deciduous broadleaf
forest (RD), tropical evergreen broadleaf forest (RE).
hot drought and concomitant bark beetle
Right panel photographs show (C) Pinus edulis mortality (Ips confusus) outbreaks at the start of the
under way in a southwestern US woodland (October millennium (2002–2004; Breshears et al.
2002), and (D) the same view after dead P. edulis trees 2005; Figure 7.1 and Plate 3). Piñon pine
have dropped needles but trunks remain standing (May is a foundational species that provides re-
2004). (Photos by C. D. Allen from Breshears et al. 2009.)
sources either directly or indirectly to a di-
verse set of individual species, populations,
al. 2014). Megafires—crown fires at scales and communities (e.g., Trotter et al. 2008).
unprecedented historically (and perhaps Biodiversity at multiple scales is likely to be
largely unprecedented prehistorically)— greatly reduced in conjunction with major
can kill trees within hours to days and reductions in such foundational species. Ex-
can drastically alter landscape structure at periments with potted plants in glasshouses
broad scales (Stephens et al. 2014). In up- or in the field with in situ plants, as well
land tropical systems where biodiversity as correlations between climate conditions
is still high, warming climate can change and tree mortality, all highlight the vul-
R APID BROAD-SCALE ECOSYSTEM CHANGE S 83
nerability of piñon pine to warmer tem- less susceptible to beetle invasion (Kula-
perature and associated increases in vapor kowski et al. 2012) and in which stands ex-
pressure deficit during drought (Allen et al. periencing substantial mortality due to bee-
2015). Even though piñon-juniper (Juniperus tles are less likely to experience crown fires
monosperma) woodlands can have a relatively (Simard et al. 2010). However, although pa-
sparse amount of tree cover premortality, leoecological evidence suggests disturbance
loss of just one of the two codominant tree influences lodgepole pine ranges, patterns
species is sufficient to greatly alter site mi- of moisture availability also play a key role
croclimate (Royer et al. 2011). Piñon mor- (Minckley et al. 2011) and are likely to shift
tality across the southwestern United States in response to climate change (Bradford et
in 2002 led to conversion of piñon-juniper al. 2014).
woodland to a juniper woodland savanna
(Clifford et al. 2011), and for the 12 years
Boreal-Tundra Ecosystems
after that mortality event, piñon recruit-
ment has been low enough to suggest that Boreal and tundra ecosystems are vast, and
the woodland vegetation could be replaced although their biodiversity can be relatively
by savanna or grassland vegetation (Red- low, they are potentially vulnerable to rapid
mond et al. 2015). broad-scale changes triggered by climate.
Climate models predict that boreal and tun-
dra ecosystems will change dramatically
Evergreen Temperate, Coniferous,
and rapidly as temperatures rise via changes
Montane Forests
in permafrost thaw rates, increased fire fre-
A mountain pine beetle outbreak, which quency, and increased shrub cover (IPCC
began in the mid-1990s, caused widespread 2012, 2014). Interactions among permafrost
mortality in lodgepole pine (Pinus contorta) thaw, wildfires, and shrub expansion are
stands throughout the western United States expected to collectively trigger rapid broad-
and Canada (Raffa et al. 2008). Lodgepole scale changes (Figure 7.2; Settele et al. 2014).
pine, which is broadly distributed in west- Reductions in biodiversity are expected in
ern North America from the Yukon Ter- at least some cases for tundra (Callaghan,
ritory to Baja, California (Safranyik et al. Tweedie, et al. 2011). For example, warmer
2007), is a classic fire-dependent species temperatures on Disko Island, Central West
with high levels of cone serotiny (Minck- Greenland were associated with net species
ley et al. 2011). Mountain pine beetles (Den- loss at a fell-field, even though warmer tem-
droctonus ponderosae) are a well-recognized and peratures had little effect on species losses
pervasive disturbance agent in lodgepole and gains in a nearby herb-slope commu-
pine forests (Safranyik et al. 2007). The nity (Callaghan, Christensen, et al. 2011).
mid-1990s outbreak appears to be unprec- Boreal species diversity increased in Yukon,
edented over at least the past century (Raffa Canada during a 42-year period over which
et al. 2008), likely due to unusually hot and temperatures increased by 2°C (Danby et al.
dry conditions that induce stress in the host 2011), highlighting the potential for rapid
trees, similar to the stress driving change changes as temperatures continue to warm.
in piñon-juniper woodlands (Allen et al. Mosses and lichens may be replaced by
2015). The severity of the recent outbreak vascular plants and their litter (Callaghan,
may have also been influenced by abnor- Tweedie, et al. 2011), although some stud-
mally high and extensive host abundance, ies suggest they may not respond to warm-
possibly as a result of several decades of fire ing and under some conditions bryophyte
suppression (Raffa et al. 2008). Fire and in- biomass could actually increase (Hudson
sect outbreaks can drive negative feedbacks and Henry 2009). Projected climate change
in which young, recently burned stands are scenarios raise concern that transitions in
84 W H AT C H A NGE S A R E W E OB SE RV I NG?
Figure 7.2. Tundra-boreal biome shift. Earth-system mod- lated to climate and global change, such as
els predict a northward shift of Arctic vegetation with drought, fire, and land use (Brando et al.
climate warming, as the boreal biome migrates into what
is currently tundra. Observations of shrub expansion in
2014). Experimental results indicate that
tundra, increased tree growth at the tundra-forest tran- the interaction between drought and fire
sition, and tree mortality at the southern extent of the could lead to a near-future dieback of the
boreal forest in recent decades are consistent with model Amazon Region forest, with associated bio-
projections. Vegetation changes associated with a biome diversity loss in one of the most species-
shift, which is facilitated by intensification of the fire re-
gime, will modify surface energy budgets, and net eco-
rich areas on the planet (see Case Study 6,
system carbon balance, permafrost thawing, and meth- Figure CS6.1). Hot droughts in association
ane emissions, with net feedbacks to additional climate with human-caused fires are predicted to
change. (Figure 4-10 and figure 4-8 from IPCC 2014.) significantly reduce Amazon forest area in
this century (Bush et al. 2008), possibly
boreal-tundra ecosystems and decline in even in coming decades (Settele et al. 2014).
their associated biodiversity will accelerate Projected changes to climate are expected
as conditions continue to warm, potentially to exacerbate forests already degraded by
triggering changes even more rapidly than land-use change, which has already had a
have occurred to date (Settele et al. 2014). significant impact on biodiversity. These
changes include grasses replacing woody
plants and potentially a rapid transition to a
Lowland Tropical Forests—The
savanna state that supports significantly less
Amazon Region
species (Silvério et al. 2013). Feedbacks be-
Large areas of wet tropical forests, includ- tween vegetation and the atmosphere sug-
ing the Amazon Region, have undergone gest frequent extreme drought events could
significant changes linked to processes re- potentially destabilize significant parts of
R APID BROAD-SCALE ECOSYSTEM CHANGE S 85
forest in the Amazon region (Zemp et al. climate conditions, or “climate velocity,”
2017). in rates of reproduction, dispersal area, and
establishment to take advantage of transient
windows of suitable habitat (IPCC 2014; Fig-
Tropical Montane Ecosystems
ure 7.3). Tracking climate velocity depends
Tropical montane ecosystems are uniquely on reproduction, dispersal, germination,
diverse, containing extremes in topogra- and/or establishment rates, which them-
phy with pronounced climatic gradients selves depend on species-specific character-
(Bruijnzeel 2004). Steep climatic gradients istics such as generation intervals and seed
in tropical montane ecosystems typically size. Many plant species and communities
constrain species distribution; numerous are unlikely to be able to track the climate
species occur within relatively narrow al- velocity (Settele et al. 2014), particularly, for
titudinal bands. Sensitive to small changes example, in flatter—rather than steeper—
in climate, tropical montane ecosystems are terrains, where the dispersal rate needed to
prone to rapid broad-scale changes. Spe- track changes in climate is higher than ex-
cies within narrow elevational bands are pected for many species. Gradients in eleva-
vulnerable to changes in land use that in- tion (mountains) have been used as a proxy
clude transformation of natural ecosystems for expected changes over long latitudinal
into, for example, agricultural lands. These distances of flat terrain (Jump et al. 2009;
changes can lead to shifts in species distri- Settele et al. 2014), although larger distances
butions, constrain migration, and exacer- make this proxy problematic (Jump et al.
bate biodiversity loss (Magrin et al. 2014). 2009).
For example, many species thrive in cloud The impacts on biodiversity are poten-
forests, depend on fog interception, and are tially greater when ecosystem changes are
particularly vulnerable to climate change. broad scale because species’ abilities to track
Permanent or seasonal contact with fog is climate can be overwhelmed. Consequently,
necessary for survival, and small changes in rapid broad-scale ecosystem changes re-
temperature can dramatically alter humid- quire consideration of not only local im-
ity, fog, and cloud density threatening sur- pacts of climate change but also broad-scale
vival (Villegas et al. 2008). Consequently, impacts across a region. The consequences
climate change and related disturbances to biodiversity and associated ecosystem
to hydroclimatic variables that determine services also depend on how patchy the
the presence of fog can have significant remnant areas of an impacted ecosystem
implications for biodiversity in these sys- are in response to changing climate (López-
tems (Bruijnzeel 2004); and consequently, Hoffman et al. 2013). Additionally, biodi-
increased rates of species extinction are ex- versity in one region could potentially be
pected (Magrin et al. 2014). affected by broad-scale ecosystem changes
in another, nonadjacent region through
“ecoclimatic teleconnections,” whereby lo-
I NSIGH TS , R E SE A RC H N E E DS , A N D cal vegetation change in one location, such
M A NAGE M E N T C H A L L E NGE S as drought-induced tree die-off, results in
climatic effects not only locally but also as
Of the many ways in which climate change transmitted through atmospheric circula-
can and will have an impact on biodiversity, tion, potentially affecting vegetation else-
rapid, broad-scale changes in ecosystems where (Garcia et al. 2016; Stark et al. 2016;
pose a particular challenge as to whether Swann et al. 2018). The scope of types of
species can track changes in climate. Spe- impacts also needs to be expanded when
cies that cannot adapt to local changes will considering the biodiversity consequences
be required to track the pace of changing of rapid broad-scale changes in ecosystems.
86 W H AT C H A NGE S A R E W E OB SE RV I NG?
Figure 7.3. Maximum speeds at which species can move some cases to avoid substantial loss of spe-
across landscapes (based on observations and models; cies and ecosystem services at broad scales
vertical axis on left), compared with speeds at which
(Settele et al. 2014; Allen et al. 2015; Mil-
temperatures are projected to move across landscapes
(climate velocities for temperature; vertical axis on lar and Stephenson 2015). Management
right). Human interventions, such as transport or habitat actions, such as assisted migration, can in-
fragmentation, can greatly increase or decrease speeds of crease the inherent capacity of ecosystems
movement. White boxes with black bars indicate ranges and their constituent organisms to adapt to
and medians of maximum movement speeds for trees,
a different climate and can reduce the risk
plants, mammals, plant-feeding insects (median not es-
timated), and freshwater mollusks. For RCP2.6, 4.5, 6.0, of local extinctions and undesirable im-
and 8.5 for 2050–2090, horizontal lines show climate pacts on ecosystem function (Settele et al.
velocity for the global-land-area average and for large 2014). In many cases, species will have to
flat regions. Species with maximum speeds below each migrate across unfavorable habitats to reach
line are expected to be unable to track warming in the
suitable climates or areas of lower land-use
absence of human intervention. (Figure SPM from IPCC
2014.) pressure, including protected areas (IPCC
2014). When habitat fragmentation affects
When species groups that greatly influ- the migration of key species or when the
ence ecosystem structure are negatively af- rate of climate change is faster than natural
fected, other factors such as consequences migration rates, management actions such
of changing microclimate must be consid- as assisted migration may increase the abil-
ered in addition to the direct impacts of a ity of the system to maintain its structure
climate event or trend. and function by tolerating such changes
Management of biodiversity will be (Hoegh-Guldberg et al. 2008). Manage-
challenged given that the capacity for ment actions for assisted migration include
ecosystems to adapt to the projected rates maintaining or improving existing habitat,
and magnitudes of climate change in the maintaining or improving migration corri-
twenty-first century may be insufficient in dors, and directly translocating species or
R APID BROAD-SCALE ECOSYSTEM CHANGE S 87
genetically distinct populations within a sequoias (Sequoiadendron gigantean) in western
species (Hoegh-Guldberg et al. 2008). Key- US national parks might become a prior-
stone species such as pollinators and other ity under extreme drought even though
dominant species such as trees are impor- current land management practices would
tant to manage under changing climate to generally preclude that option (Grant et al.
preserve biodiversity and to maintain the 2013). In conclusion, land managers and
provision of ecosystem services (Zarnetske policy makers should consider the suite of
et al. 2012). Climate change will present potential impacts on biodiversity and asso-
many natural resource management chal- ciated ecosystem services, rather than single
lenges that will require land managers and species-by-species cases, when attempting
policy makers to work together to combat to preserve biodiversity and maintain the
the impacts of a changing climate on eco- provision of ecosystem services, especially
system function and biodiversity. under rapid broad-scale climate-driven
The ability to anticipate, predict, and changes that have already begun to occur
manage biodiversity under rapid broad- and are expected to increase in the future.
scale ecological change overlaps with more
general considerations about biodiversity
responses to climate change and requires ACK NOWLE DGME NTS
considering the following. First, there is
need for greater recognition of the risks of This publication was supported by the Na-
rapid broad-scale change and their associ- tional Science Foundation through Mac-
ated consequences. For example, western rosystems Biology (NSF EF-1340624, EF-
US forests were generally not viewed as 1550756 and EF-1550756) and Critical Zone
highly vulnerable in the early 1990s, but Observatories (Santa Catalina Mountains
within the past two decades have under- and Jemez River Basin; NSF EAR-1331408),
gone enormous broad-scale rapid change Philecology Foundation, Fort Worth for Bio-
through forest die-off and wildfire (Allen et sphere 2, Arizona Agricultural Experiment
al. 2015). Second, an improved understand- Station, the U.S. Geological Survey’s Eco-
ing is needed of the secondary impacts of systems and Climate and Land Use Change
broad-scale ecosystem change, such as for- mission areas, through the USGS Western
est die-off effects on microclimate change Mountain Initiative project; and Colciencias
being propagated elsewhere through eco- (Programa de investigación en gestión del
climatic teleconnections (Garcia et al. 2016; riesgo asociado a cambio climático y am-
Stark et al. 2016; Swann et al. 2018). Third, biental en cuencas hidrográficas—convoca-
to account for these impacts, conservation toria 543/2011). Any use of trade, product,
strategies are needed that allow for “bet or firm names is for descriptive purposes
hedging” for protected areas (Davison et only and does not imply endorsement by
al. 2011). Fourth, improved methods of the U.S. Government.
assessing broad-scale change are needed
during early stages to enable feasible rapid
responses (Hartmann et al. 2018). Fifth, R E FE R E NCE S
broad-scale changes may particularly am-
plify the need to consider assisted migra- Adams, H. D., G. A. Barron-Gafford, R. L. Minor, A. A.
tion (Hoegh-Guldberg et al. 2008). Sixth, Gardea, L. Patrick Bentley, D. J. Law, D. D. Breshears,
depending on the patchiness of the change, N. G. McDowell, and T. E. Huxman. 2017. “Tempera-
ture response surfaces for mortality risk of tree spe-
scientists may wish to consider whether
cies with future drought.” Environmental Research Letters
microclimate refugia can be established 12: 115014.
within an impacted area. For example, the Allen, Craig D., David D. Breshears, and Nate G. Mc-
watering of iconic groves of ancient giant Dowell. 2015. “On underestimation of global vulner-
88 W H AT C H A NGE S A R E W E OB SE RV I NG?
ability to tree mortality and forest die-off from hotter yon-juniper woodland: Climate-change-type drought
drought in the Anthropocene.” Ecosphere 6 (8): art. 129. resets successional clock.” Ecosystems 14 (6): 949–962.
Allen, Craig D., Alison K. Macalady, Haroun Chen- Danby, Ryan K., Saewan Koh, David S. Hik, and Larry
chouni, Dominique Bachelet, Nate McDowell, Michel W. Price. 2011. “Four decades of plant community
Vennetier, Thomas Kitzberger, et al. 2010. “A global change in the alpine tundra of southwest Yukon, Can-
overview of drought and heat-induced tree mortality ada.” AMBIO: A Journal of the Human Environment 40 (6):
reveals emerging climate change risks for forests.” For- 660–671.
est Ecology and Management 259 (4): 660–684. Davison, Jennifer E., David D. Breshears, Willem J. D.
Bradford, John B., Daniel Schlaepfer, and William K. Van Leeuwen, and Grant M. Casady. 2011. “Remotely
Lauenroth. 2014. “Ecohydrology of adjacent sage- sensed vegetation phenology and productivity along a
brush and lodgepole pine ecosystems: The conse- climatic gradient: On the value of incorporating the
quences of climate change and disturbance.” Ecosystems dimension of woody plant cover.” Global Ecology and Bio-
17 (4): 590–605. geography 20 (1): 101–113.
Brando, Paulo Monteiro, Jennifer K. Balch, Daniel C. Foster, Pru. 2001. “The potential negative impacts of
Nepstad, Douglas C. Morton, Francis E. Putz, Michael global climate change on tropical montane cloud for-
T. Coe, Divino Silvério, et al. 2014. “Abrupt increases ests.” Earth-Science Reviews 55 (1): 73–106.
in Amazonian tree mortality due to drought-fire in- Garcia, E. S., A. L. S. Swann, J. C. Villegas, D. D. Bres-
teractions.” Proceedings of the National Academy of Sciences 111 hears, D. J. Law, S. R. Saleska, S. C. Stark. 2016. “Syn-
(17): 6347–6352. ergistic ecoclimate teleconnections from forest loss
Breshears, David D., Neil S. Cobb, Paul M. Rich, Kevin in different regions structure global ecological re-
P. Price, Craig D. Allen, Randy G. Balice, William H. sponses.” PLOS One 11: e0165042.
Romme, et al. 2005. “Regional vegetation die-off in Gonzalez, Patrick, Ronald P. Neilson, James M. Lenihan,
response to global-change-type drought.” Proceedings of and Raymond J. Drapek. 2010. “Global patterns in the
the National Academy of Sciences of the United States of America vulnerability of ecosystems to vegetation shifts due
102 (42): 15144–15148. to climate change.” Global Ecology and Biogeography 19 (6):
Breshears, David D., Travis E. Huxman, Henry D. Ad- 755–768.
ams, Chris B. Zou, and Jennifer E. Davison. 2008. Grant, Gordon E., Christina L. Tague, and Craig D. Allen.
“Vegetation synchronously leans upslope as climate 2013. “Watering the forest for the trees: An emerg-
warms.” Proceedings of the National Academy of Sciences 105 ing priority for managing water in forest landscapes.”
(33): 11591–11592. Frontiers in Ecology and the Environment 11 (6): 314–321.
Bruijnzeel, Leendert Adriaan. 2004. “Hydrological Hartmann, Henrik, Catarina F. Moura, William R. L.
functions of tropical forests: Not seeing the soil for Anderegg, Nadine K. Ruehr, Yann Salmon, Craig D.
the trees?” Agriculture, Ecosystems & Environment 104 (1): Allen, Stefan K. Arndt, David D. Breshears, Hendrik
185–228. Davi, David Galbraith, et al. 2018. “Research frontiers
Bush, M. B., M. R. Silman, C. McMichael, and S. Saatchi. for improving our understanding of drought-induced
2008. “Fire, climate change and biodiversity in Ama- tree and forest mortality.” New Phytologist 218 (1): 15–28.
zonia: A late-Holocene perspective.” Philosophical Trans- Hoegh-Guldberg, Ove, Lesley Hughes, Sue McIntyre, D.
actions of the Royal Society B—Biological Sciences 363 (1498): B. Lindenmayer, C. Parmesan, Hugh P. Possingham,
1795–1802. and C. D. Thomas. 2008. “Assisted colonization and
Cahill, Abigail E., Matthew E. Aiello-Lammens, M. Cait- rapid climate change.” Science 321: 325–326.
lin Fisher-Reid, Xia Hua, Caitlin J. Karanewsky, Hae Hudson, James M. G., and Greg H. R. Henry. 2009. “In-
Yeong Ryu, Gena C. Sbeglia, et al. 2012. “How does creased plant biomass in a High Arctic heath commu-
climate change cause extinction?” In Proceedings of the nity from 1981 to 2008.” Ecology 90 (10): 2657–2663.
Royal Society B: rspb20121890. IPCC. 2012. Managing the Risks of Extreme Events and Disasters to
Callaghan, Terry V., Torben R. Christensen, and Elin J. Advance Climate Change Adaptation: A Special Report of Working
Jantze. 2011. “Plant and vegetation dynamics on Disko Groups I and II of the Intergovernmental Panel on Climate Change,
Island, West Greenland: Snapshots separated by over ed. C. B. Field, V. Barros, T. F. Stocker, D. Qin, D. J.
40 years.” AMBIO: A Journal of the Human Environment 40 Dokken, K. L. Ebi, M. D. Mastrandrea, K. J. Mach,
(6): 624–637. G.-K. Plattner, S. K. Allen, M. Tignor, and P. M. Midg-
Callaghan, Terry V., Craig E. Tweedie, Jonas Åkerman, ley. Cambridge University Press.
Christopher Andrews, Johan Bergstedt, Malcolm G. IPCC. 2014. “Climate change 2014: Impacts, adaptation,
Butler, Torben R. Christensen, et al. 2011. “Multi- and vulnerability. Part A: Global and sectoral aspects.”
decadal changes in tundra environments and ecosys- In Contribution of Working Group II to the Fifth Assessment Re-
tems: synthesis of the International Polar Year-Back port of the Intergovernmental Panel on Climate Change, ed. C.
to the Future Project (IPY-BTF).” AMBIO: A Journal of the B. Field, V. R. Barros, D. J. Dokken, K. J. Mach, M.
Human Environment 40 (6): 705–716. D. Mastrandrea, T. E. Bilir, M. Chatterjee, et al. Cam-
Clifford, Michael J., Neil S. Cobb, and Michaela Buene- bridge University Press. https://doi.org/10.1017/
mann. 2011. “Long-term tree cover dynamics in a pin- CBO9781107415379.
R APID BROAD-SCALE ECOSYSTEM CHANGE S 89
Jump, Alistair S., Csaba Mátyás, and Josep Peñuelas. Settele, J., R. Scholes, R. Betts, S. Bunn, P. Leadley, D.
2009. “The altitude-for-latitude disparity in the range Nepstad, J. T. Overpeck, and M. A. Taboada. 2014.
retractions of woody species.” Trends in Ecology & Evolution “Terrestrial and inland water systems.” In Climate
24 (12): 694–701. Change 2014: Impacts, Adaptation, and Vulnerability. Part A:
Koh, Lian Pin, Robert R. Dunn, Navjot S. Sodhi, Robert Global and Sectoral Aspects. Contribution of Working Group II to
K. Colwell, Heather C. Proctor, and Vincent S. Smith. the Fifth Assessment Report of the Intergovernmental Panel on Cli-
2004. “Species coextinctions and the biodiversity cri- mate Change, ed. C. B. Field, V. R. Barros, D. J. Dokken,
sis.” Science 305 (5690): 1632–1634. K. J. Mach, M. D. Mastrandrea, T. E. Bilir, M. Chat-
Kulakowski, Dominik, Daniel Jarvis, Thomas T. Veblen, terjee, K. L. Ebi, Y. O. Estrada, R. C. Genova, B. Girma,
and Jeremy Smith. 2012. “Stand-replacing fires reduce E. S. Kissel, A. N. Levy, S. MacCracken, P. R. Mastran-
susceptibility of lodgepole pine to mountain pine drea, and L. L. White, 271–359. Cambridge University
beetle outbreaks in Colorado.” Journal of Biogeography 39 Press.
(11): 2052–2060. Silvério, Divino V., Paulo M. Brando, Jennifer K. Balch,
Logan, Jesse A., Jacques Régnière, and James A. Powell. Francis E. Putz, Daniel C. Nepstad, Claudinei Oliveira-
2003. “Assessing the impacts of global warming on Santos, and Mercedes M. C. Bustamante. 2013. “Test-
forest pest dynamics.” Frontiers in Ecology and the Environ- ing the Amazon savannization hypothesis: Fire effects
ment 1 (3): 130–137. on invasion of a neotropical forest by native cerrado
López-Hoffman, Laura, David D. Breshears, Craig D. Al- and exotic pasture grasses.” Philosophical Transactions of
len, and Marc L. Miller. 2013. “Key landscape ecol- the Royal Society of London B: Biological Sciences 368 (1619):
ogy metrics for assessing climate change adaptation 20120427.
options: Rate of change and patchiness of impacts.” Simard, Martin, William H. Romme, Jacob M. Griffin,
Ecosphere 4 (8): 1–18. and Monica G. Turner. 2011. “Do mountain pine bee-
Magrin, G. O., J. A. Marengo, J. P. Boulanger, M. S. Buck- tle outbreaks change the probability of active crown
eridge, E. Castellanos, G. Poveda, F. R. Scarano, and fire in lodgepole pine forests?” Ecological Monographs 81
S. Vicuña. 2014. “Central and South America.” Climate (1): 3–24.
Change: 1499–1566. Stark, Scott C., David D. Breshears, Elizabeth S. Garcia,
Millar, Constance I., and Nathan L. Stephenson. 2015. Darin J. Law, David M. Minor, Scott R. Saleska, Abigail
“Temperate forest health in an era of emerging mega- L. S. Swann, et al. 2016. “Toward accounting for eco-
disturbance.” Science 349 (6250): 823–826. climate teleconnections: Intra- and inter-continental
Minckley, Thomas A., Robert K. Shriver, and B. Shuman. consequences of altered energy balance after vegeta-
2012. “Resilience and regime change in a southern tion change.” Landscape Ecology 31 (1): 181–194.
Rocky Mountain ecosystem during the past 17000 Stephens, Scott L., Neil Burrows, Alexander Buyantuyev,
years.” Ecological Monographs 82 (1): 49–68. Robert W. Gray, Robert E. Keane, Rick Kubian, Shi-
Raffa, Kenneth F., Brian H. Aukema, Barbara J. Bentz, rong Liu, et al. 2014. “Temperate and boreal forest
Allan L. Carroll, Jeffrey A. Hicke, Monica G. Turner, mega-fires: Characteristics and challenges.” Frontiers in
and William H. Romme. 2008. “Cross-scale drivers of Ecology and the Environment 12 (2): 115–122.
natural disturbances prone to anthropogenic amplifi- Swann, A. L. S., M. Laguë, E. S. Garcia, D. D. Breshears,
cation: the dynamics of bark beetle eruptions.” AIBS J. P. Field, D. J. P. Moore, S. R. Saleska, S. C. Stark, J.
Bulletin 58 (6): 501–517. C. Villegas, D. J. Law, and D. M. Minor. 2018. “Con-
Redmond, Miranda D., Neil S. Cobb, Michael J. Clifford, tinental-scale consequences of tree die-offs in North
and Nichole N. Barger. 2015. “Woodland recovery America: Identifying where forest loss matters most.”
following drought-induced tree mortality across an Environmental Research Letters.
environmental stress gradient.” Global Change Biology 21 Trotter, R. T., N. S. Cobb, and T. G. Whitham. 2008. “Ar-
(10): 3685–3695. thropod community diversity and trophic structure: a
Royer, Patrick D., Neil S. Cobb, Michael J. Clifford, Cho- comparison between extremes of plant stress.” Ecologi-
Ying Huang, David D. Breshears, Henry D. Adams, caEntomology 33: 1–11.
and Juan Camilo Villegas. 2011. “Extreme climatic Villegas, J. C., D. J. Law, S. C. Stark, D. M. Minor, D. D.
event-triggered overstorey vegetation loss increases Breshears, S. R. Saleska, A. L. S. Swann, E. S. Garcia,
understorey solar input regionally: Primary and sec- E. M. Bella, J. M. Morton, N. S. Cobb, G. A. Barron-
ondary ecological implications.” Journal of Ecology 99 Gafford, M. E. Litvak, and T. E. Kolb. 2017. “Prototype
(3): 714–723. campaign assessment of disturbance-induced tree-loss
Safranyik, Les, and Allan L. Carroll. 2006. “The biol- effects on surface properties for atmospheric model-
ogy and epidemiology of the mountain pine beetle ing.” Ecosphere 8 (3): e01698.
in lodgepole pine forests.” In The Mountain Pine Beetle: A Villegas, Juan Camilo, Conrado Tobón, and David D.
Synthesis of Biology, Management, and Impacts on Lodgepole Pine, Breshears. 2008. “Fog interception by non-vascular
ed. L. Safranyik and W. R. Wilson, 3–66. Victoria, BC: epiphytes in tropical montane cloud forests: Depen-
Natural Resources Canada, Canadian Forest Service, dencies on gauge type and meteorological condi-
and Pacific Forestry Centre. tions.” Hydrological Processes 22 (14): 2484–2492.
90 W H AT C H A NGE S A R E W E OB SE RV I NG?
Whittaker, Robert Harding. 1960. “Vegetation of the Zemp, Delphine Clara, Carl-Friedrich Schleussner, Hen-
Siskiyou mountains, Oregon and California.” Ecological rique M. J. Barbosa, Marina Hirota, Vincente Mon-
Monographs 30 (3): 279–338. tade, Gilvan Sampaio, Arie Staal, Lan Wang-Erlands-
Zarnetske, Phoebe L., David K. Skelly, and Mark C. Ur- son, and Anja Rammig. 2017. “Self-amplified Amazon
ban. 2012. “Biotic multipliers of climate change.” Sci- forest loss due to vegetation-atmosphere feedbacks.”
ence 336 (6088): 1516–1518. Nature Communications 8: 14681.
CASE STUDY 3
Adélie penguins (Pygoscelis adeliae) thrive in some of the most dramatically variable habitats
and weather on Earth. They spend much of their life at sea, more than 40 m underwater,
often under sea ice, where they find their food. They also spend roughly 3 months per
year mostly on land, living in dense colonies for breeding. They regularly contend with the
transition between open and frozen ocean, and with terrain alternately blanketed in snow
and ice, then swept clear by high winds. Adélie penguins are one of only a handful of spe-
cies to be able to survive extended periods in subzero air temperatures, out of the relatively
warm water that harbors most of the biodiversity to be found at the highest latitudes of the
Southern Ocean. They require both ice-free terrain to nest and nearby open water to forage.
The combination of these two things is rare in Antarctica, but where they are found, so, too,
are Adélies.
Over the past 12 million–15 million the Antarctic Peninsula region. This change
years, Adélie penguins have contended with in persistence coincides with dramatic re-
a wide range of climates and consequent ductions in the overall extent of sea ice in
impacts to their habitat. Ice sheets have re- both those regions. The opposite is true in
peatedly expanded and retreated hundreds East Antarctica and in the Ross Sea region,
of kilometers, destroying or creating nesting where the sea-ice season has extended by
habitat, and Adélie populations have grown two months over the same period. These
and shrunk corresponding to interglacial and changes are among the largest phenological
glacial periods, respectively (Li et al. 2014). shifts so far associated with anthropogenic
The comings and goings of Adélies through climate change, and along with warming
geologic time, as determined by dating sub- temperatures and increasing precipitation,
fossil bones, have been used to validate the they have profound implications for sea-ice
dates of ice-sheet advances and retreats. A ecosystems (Sailley et al. 2013). In response,
warm period 2,000–4,000 years BP is even breeding populations of Adélie penguins
known as the “penguin optimum” for the are retreating southward from much of the
widespread extent of ancient penguin colo- Antarctic Peninsula—a place continuously
nies found from this time period—includ- occupied by Adélies for 500–800 years
ing places where they have not yet reoccu- (Emslie 2001)—but they are slowly increas-
pied (Baroni and Orombelli 1997). ing as sea ice loosens farther south in that
As a result of recent and ongoing region (Lynch et al. 2012). Concomitantly,
changes in climate and sea ice, Adélies are Adélie populations in the southern Ross Sea
now encountering changes of magnitudes are expanding rapidly (Lyver et al. 2014; see
previously inferred from the geologic and Plate 5), in some cases exploiting nesting
genetic record, but at a faster pace. Be- habitat recently exposed by retreating ice
tween 1979 and 2010, the period of time sheets (LaRue et al. 2013).
during which the sea-ice field in Antarc-
tica expands, known as the sea-ice season
(Stammerjohn et al. 2012; see Plate 5), has Copyright © 2019 by Yale University. All rights reserved.
declined by 3 months in the Arctic and in ISBN 978-0-300-20611-1.
91
92 W H AT C H A NGE S A R E W E OB SE RV I NG?
The mechanisms driving these changes have sufficient time to raise young, given
also relate to the specific adaptations of the short breeding season of high-latitude
Adélie penguins to the sea-ice environment. ecosystems. In this way, Adélies are caught
During summer months, the presence of 6 between astronomically imposed limits and
percent–15 percent sea-ice cover is optimal; anthropogenic climate change.
more or less ice and foraging trips grow lon- The diverging Adélie penguin popula-
ger, with less food delivered to chicks (Bal- tion trends described here are projected to
lard, Dugger, et al. 2010). The sea ice itself continue during the next few decades, af-
is a substrate for diatom growth and subse- ter which sea ice, as warming reaches 2°C
quent grazing by krill and copepods, which above preindustrial levels, is predicted to
in turn are food for small fish (especially decrease everywhere in the Southern Ocean
sea-ice-dwelling silverfish—Pleuragramma (Ainley et al. 2010). With the retreat of sea
antarcticum) and penguins. Silverfish are also ice, Adélie penguins and the other three
major prey for Adélie penguins, particularly truly ice-obligate seabird species (Antarctic
in the Ross Sea (Ainley et al. 2003). It is pos- petrel—Thalassoica antarctica; snow petrel—
sibly the loss of silverfish that explains the Pagodrama nivea; and emperor penguin—Ap-
decrease in Adélie penguins in the northern tenodytes forsteri) will eventually disappear
Antarctic Peninsula. Without sea ice, the (Ainley et al. 2010; Jenouvrier et al. 2014),
food web in the Southern Ocean is far less to be replaced by ice-tolerant and ice-avoid-
complex and less suited to higher trophic ing species.
level predators like penguins, as it is domi-
nated by algae (Phaeocystis antarctica) with lim-
ited grazing by pteropods, which penguins ACK NOWLE DGME NTS
and fish do not appear to consume. Sea ice
is also a platform for Adélies to rest upon, We are grateful for support from the Na-
molt, and seek shelter from predators, and tional Science Foundation Grants OPP-
can provide a source of freshwater when 0944411 and 1543498, and to Ian Gaffney
snow accumulates on its surface. for production of Plate 5. Point Blue Conser-
Sea-ice variability has also substantially vation Science Contribution #2113.
impacted Adélie penguins’ migratory pat-
terns through time, with today’s popula-
tions ranging from essentially nonmigratory R E FE R E NCE S
at the more northerly parts of their range
to long-distance migratory at the southern- Ainley, David, Joellen Russell, Stephanie Jenouvrier,
Eric Woehler, Philip O’B. Lyver, William R. Fraser,
most portions—with annual journeys of up
and Gerald L. Kooyman. 2010. “Antarctic penguin
to ~18,000 km (round trip)—an astound- response to habitat change as Earth’s troposphere
ing feat for a flightless animal (Ballard, To- reaches 2ºC above preindustrial levels.” Ecological Mono-
niolo, et al. 2010). Their apparent need for graphs 80 (1): 49–66.
some amount of daylight during all phases Ballard, Grant, Katie M. Dugger, Nadav Nur, and David
G. Ainley. 2010. “Foraging strategies of Adélie pen-
of their annual cycle appears to limit the po-
guins: Adjusting body condition to cope with envi-
tential range of wintering and migration to ronmental variability.” Marine Ecology Progress Series 405:
north of 72.7°S—the latitude of zero mid- 287–302.
winter twilight. It is likely that, as a result Ballard, Grant, Viola Toniolo, David G. Ainley, Claire L.
of climate change, more suitable breeding Parkinson, Kevin R. Arrigo, and Phil N. Trathan. 2010.
“Responding to climate change: Adélie penguins con-
habitat will become available farther south
front astronomical and ocean boundaries.” Ecology 91
than it currently exists (as glaciers retreat (7): 2056–2069.
and free up more coastline for nesting), but Baroni, Carlo, and Giuseppe Orombelli. 1994. “Aban-
it may not be possible for Adélies to migrate doned penguin rookeries as Holocene paleoclimatic
much farther than they already do and still indicators in Antarctica.” Geology 22 (1): 23–26.
R A PI DLY DI V E RGI NG P OPU L AT ION T R E N D S OF A DÉ L I E PE NGU I N S 93
Emslie, Steven D. 2001. “Radiocarbon dates from aban- lations on the Antarctic Peninsula.” Ecology 93 (6):
doned penguin colonies in the Antarctic Peninsula 1367–1377.
region.” Antarctic Science 13 (3): 289–295. Lyver, Phil O’B., Mandy Barron, Kerry J. Barton, David
Jenouvrier, Stéphanie, Marika Holland, Julienne Stroeve, G. Ainley, Annie Pollard, Shulamit Gordon, Stephen
Mark Serreze, Christophe Barbraud, Henri Weimer- McNeill, Grant Ballard, and Peter R. Wilson. 2014.
skirch, and Hal Caswell. 2014. “Projected continent- “Trends in the breeding population of Adélie pen-
wide declines of the emperor penguin under climate guins in the Ross Sea, 1981–2012: A coincidence of
change.” Nature Climate Change 4 (8): 715–718. https:// climate and resource extraction effects.” PLOS One 9
doi.org/10.1038/NCLIMATE2280. (3): e91188.
LaRue, Michelle A., David G. Ainley, Matt Swanson, Ka- Sailley, Sévrine F., Hugh W. Ducklow, Holly V. Moeller,
tie M. Dugger, O. Phil, B. Lyver, Kerry Barton, and William R. Fraser, Oscar M. E. Schofield, Deborah K.
Grant Ballard. 2013. “Climate change winners: Reced- Steinberg, Lori M. Garzio, and Scott C. Doney. 2013.
ing ice fields facilitate colony expansion and altered “Carbon fluxes and pelagic ecosystem dynamics near
dynamics in an Adélie penguin metapopulation.” PLOS two western Antarctic Peninsula Adélie penguin
One 8 (4): e60568. https://doi.org/10.1371/journal colonies: An inverse model approach.” Marine Ecology
.pone.0060568. Progress Series 492: 253–272. https://doi.org/10.3354/
Li, Cai, Yong Zhang, Jianwen Li, Lesheng Kong, Haofu meps10534.
Hu, Hailin Pan, Luohao Xu, et al. 2014. “Two Antarc- Stammerjohn, Sharon, Robert Massom, David Rind,
tic penguin genomes reveal insights into their evolu- and Douglas Martinson. 2012. “Regions of rapid
tionary history and molecular changes related to the sea ice change: An inter-hemispheric seasonal com-
Antarctic environment.” GigaScience 3 (1): 27. parison.” Geophysical Research Letters 39 (6). https://doi
Lynch, Heather J., Ron Naveen, Philip N. Trathan, and .org/10.1029/2012GL050874.
William F. Fagan. 2012. “Spatially integrated assess-
ment reveals widespread changes in penguin popu-
This page intentionally left blank
PA RT I I I
97
98 W H AT D OE S T H E PA ST T E L L US?
T H E E A RT H SYST E M A N D
ITS C A R BON C YC L E
widespread glaciation following LIP erup- mass extinctions required a severe shock of
tion (Isaacson et al. 2008; Finnegan et al. a different kind. A hopeful interpretation,
2011; Retallack 2015; Jones et al. 2017). At at least as far as a potential Anthropocene
least one major kill factor for marine eco- mass extinction is concerned, is that the
systems, at the end of the Givetian epoch gradual diversification of life into new eco-
(387.7 Ma–382.7 Ma) of the Devonian pe- logical niches has made Earth’s biosphere
riod, has been associated with rapid sea- more resilient to exogenous carbon-cycle
level rise, possibly an indirect effect of disruption. Several of the biodiversity crises
greenhouse warming, which drowned of the Paleozoic, for instance, are notable
reefs and opened marine passageways for for the extinction of whole classes of reef-
invasive species (McGhee et al. 2013). building organisms (McGhee et al. 2013).
The possible exception to this rule is the This suggests that reefs, which formed large
well-known end-Cretaceous mass extinc- test beds for evolutionary innovation and
tion. The Deccan Traps LIP eruption in In- also were among Earth’s principal sinks for
dia was roughly coeval with the Chicxulub carbonate, lacked resiliency to carbon-cycle
impact and has fueled controversy over the shocks in the Paleozoic and early Meso-
meteor hypothesis for decades (e.g., Chenet zoic. Either the evolution of corals provided
et al. 2007). Henehan et al. (2016) compares better resiliency, or the evolution of other
the lethal impacts of the Deccan Traps erup- carbonate-secreting organisms created an
tion with those of the meteor strike, using effective “crumple zone” for reef ecosys-
both sedimentary markers and a carbon- tems, similar to how a well-engineered au-
cycle model. Improvements in radiometric tomobile can absorb the impact of a traffic
dating now place the largest Deccan erup- collision.
tion roughly 250 kyr prior to the Chicxulub Large-scale environmental disruptions
bolide impact and end-Cretaceous mass ex- and sudden biotic turnover did not cease
tinction (Schoene et al. 2015; but see Renne in the later Mesozoic and Cenozoic eras.
et al. 2015). During the preimpact interval, Many marine environments in the Meso-
carbonate microfossils accumulated at the zoic experienced episodic anoxia, often
seafloor in poor condition, suggesting deg- paced by Earth’s orbital cycles (Lanci et al.
radation by added acidity from CO2 emit- 2010; Eldrett et al. 2015). These cyclic varia-
ted from the Deccan Traps. However, only tions have scarce impact on overall extinc-
2°C–3°C greenhouse warming is estimated tion rates, although preserved microfossil
from isotopic data, along with disruption in abundances varied. These regular cycles are
species habitat ranges, but not widespread punctuated at intervals of 10 Myr to 60 Myr
preimpact extinction (Thibault and Gardin with clusters of orbital layers that contain
2010). Henehan et al. (2016) hypothesize greatly elevated levels of organic carbon
that widespread carbonate dissolution on (Jenkyns 2010; Laurin et al. 2016). When
the seafloor, sourced from carbonate plank- the sedimentary evidence can be correlated
ton in the photic zone, acted to buffer the globally, geologists term these clusters of
CO2-induced acidity of the ocean. orbital cycles the ocean anoxic events, or
OAEs (Erba 2004). Jenkyns (2010) reviews
global OAEs in detail at 183 Ma (Toarcian
H AS BIODI V E R SIT Y E VOLV E D stage of the Jurassic period), 120 Ma (Selli
R E SISTA NC E TO C L IM AT E event), 111 Ma (Paquier event), and 93 Ma
DISRU P TION? (Bonarelli event), and suggests extending
the OAE classification to the Early Ceno-
The end-Cretaceous extinction began as zoic Paleocene-Eocene Thermal Maximum
a volcanism-related carbon-cycle disrup- (PETM) at 56.0 Ma. In some cases the OAEs
tion, but its status as the last of Sepkoski’s are correlated with large igneous-province
A PA L E OE COLO GIC A L PE R SPE C T I V E ON SU DDE N C L I M AT E C H A NGE 105
(LIP) eruptions, such as the Karoo-Farrar curred. A dramatic greenhouse “hyperther-
LIP for the Toarcian OAE (Caruthers et al. mal” that marks the stratigraphic boundary
2013) and the undersea eruption of the On- between the Paleocene and Eocene epochs
tong Java Plateau with the Selli event (Bot- has been proposed as an analogue for the
tini et al. 2012). The associated ecosystem Anthropocene by Zeebe and Zachos (2013).
disruption of the Toarcian OAE (183 Ma) The Paleocene-Eocene Thermal Maximum
has been described as a “mass extinction” (PETM) elevated Earth’s surface tempera-
by some (Gomez and Goy 2011; Caruthers tures by 5°C–8°C for 170 kyr–200 kyr via
et al. 2013; Krencker et al. 2014), but its im- an injection of either CH4 or CO2 at rates
pact was significantly smaller than the bio- that are sudden (e.g., 3,000 petagrams of
diversity crises reviewed by McGhee et al. carbon in 5 kyr–10 kyr) relative to baseline
(2013). geologic processes, but far slower than at-
Jenkyns (2010) notes that all OAEs of this mospheric change since the Industrial Rev-
time period share evidence for a sudden olution (perhaps 3,500–5,000 petagrams of
injection of CO2 or CH4 into Earth’s atmo- carbon in 500 years). Several hyperthermal
sphere. Oxygen isotopes within carbonate events have been identified in the isotope
fossils typically suggest increases to ocean fluctuations of benthic foraminifera of the
temperatures of 5°C–7°C (see also Gomez early Cenozoic, with repeat timing that cor-
and Goy 2011; Krencker et al. 2014) but related with 100-kyr orbital eccentricity
also evidence from strontium or osmium cycles (Nicolo et al. 2007; Sexton et al. 2011;
isotopes that continental weathering ac- Laurin et al. 2016).
celerated to limit the CO2 at least partially. An abrupt cooling at 34 Ma (Liu et al.
The clustering of orbital cycles within OAEs 2009; Hren et al. 2013) accompanied the
indicates that enhanced-greenhouse condi- extinction of many planktonic marine fora-
tions persisted for 50 kyr–400 kyr, during minifera species at the Eocene-Oligocene
which marine biologic activity is strong and epoch transition (Zachos et al. 1999; Ivany
the relative absence of carbonate microfos- et al. 2000; Pearson et al. 2008), after which
sils suggest an acidic deep ocean. Positive the Antarctic ice sheet becomes an impor-
excursions in the carbon-isotope ratio į13C tant factor in climate. After this transition
of marine carbonates are consistent with Earth’s orbital cycles modulated the growth
enhanced burial of organic carbon. Nega- of ice sheets in an icehouse climate, rather
tive shifts in į13C also occur within the than governed the release of buried carbon
duration of most OAEs, which suggests the in a greenhouse climate. Waxing and wan-
injection of isotopically-light carbon from ing of the ice sheets before the mid-Pleis-
volcanism, methane emissions, or other tocene (0.8 Ma) appear to have responded
mechanisms for volatizing organic-carbon in a simple linear manner to high-latitude
deposits. The detailed stratigraphy of geo- insolation (Imbrie et al. 1992), but the most
chemical markers within OAEs suggests recent ice-age cycles have the time scale of
complex feedbacks within the Earth system Earth’s eccentricity cycle (100 kyr), the in-
over 100-kyr intervals. OAEs typically con- solation signal of which is too small to gov-
clude with a return to the previous climate ern ice volume directly (Imbrie et al. 1993).
conditions. Extinctions impact biodiversity Ambient air preserved in ice cores indicates
during OAEs, but ecosystem functions ap- that during the glacial periods atmospheric
pear to recover quickly, at least from the CO2 was at least 100 ppm lower than dur-
proxy studies thus far. ing warm interglacial times (Jouzel 2013).
During the Cenozoic, Earth’s carbon cy- The CO2 variations validate dramatically the
cle has been tested multiple times by sudden greenhouse gas control of Earth’s climate
inputs or extractions of atmospheric CO2 change but pose a problem that is not yet
or CH4, but no biodiversity crises have oc- solved. What caused 100-kyr cycles in at-
106 W H AT D OE S T H E PA ST T E L L US?
mospheric CO2 during the late Pleistocene (Allen et al. 2015; Howe et al. 2016; Yu et al.
epoch? 2016; Jaccard et al. 2016) focus on changes
A popular Earth-system paradigm for in the relative volumes of NADW and Ant-
the 100-kyr ice-age cycle is the oceanic arctic bottom water (AABW) in the abyssal
“conveyor belt for salt,” which is facilitated ocean, with CO2 storage in glacial periods
by the production of deep-ocean water in associated with the Antarctic water mass.
the high-latitude North Atlantic (Broecker There is observational evidence that dis-
1990). North Atlantic deep water (NADW) solved CO2 trapped in the deep Southern
balances the transport of salt against the Ocean was vented to the atmosphere during
transport of water, via the atmosphere, the last transition out of glacial conditions,
from the net-evaporative Atlantic to the at roughly 12 ka (Skinner et al. 2010; Basak
net-precipitative Pacific. In the conveyor- et al. 2018).
belt model, the NADW supplies the deep An emphasis on the physical transport
ocean with greenhouse gasses absorbed of CO2 by water masses, however, ignores
from Earth’s atmosphere at the sea surface key chemical and biological carbon-cycle
(Thornalley et al. 2011). Fluctuations in į13C interactions in the ocean. A stockpile of dis-
within deep-sea sediment cores indicate solved CO2 in the deep ocean would lead
that NADW formation is unstable, strong to acidification. Although ocean acidifica-
during interglacial intervals similar to the tion impedes carbonate secretion by ma-
present day and weak during glacial inter- rine organisms (Doney et al. 2009, 2014),
vals. Maasch and Saltzmann (1990) pro- the long coexistence of high CO2 levels
posed a simple nonlinear ice-age climate and carbonate deposition in Earth history
model in which NADW formation stock- tells us that the biosphere has strategies to
piles atmospheric CO2 in the deep ocean, compensate, if given enough time to adjust.
decreasing greenhouse gas concentrations Broecker and Peng (1987) hypothesized the
in the shallow ocean and atmosphere, and carbonate compensation mechanism to ex-
leading to glacial conditions worldwide. plain glacial-to-interglacial transitions. It
Given the estimated time scales of chemical buffers the deep ocean, after a time delay
transport in the ocean (Broecker and Peng of a few millennia, via a reorganization of
1987), this nonlinear model predicted an dissolved-nutrient transport and ecosystem
alternation between glacial and interglacial function. Geologic proxies for carbonate-
states, which is consistent with paleocli- anion (CO3–2) in the world ocean suggest
mate data. that the glacial and interglacial states in the
Subsequent research has revised the deep Pacific and Indian Oceans had similar
original conveyor-belt model significantly. values (Honisch et al. 2008; Yu et al. 2014),
Keeling and Stephens (2001) proposed that in contrast to strong differences in the sur-
deepwater sources in the Antarctic, not in face-water proxies. These studies indicate
the North Atlantic, govern the supply of that global patterns of biological carbon
cold dense deep water to the world ocean. consumption, either through photosynthe-
Motivated by strong late-Pleistocene corre- sis or CaCO3 secretion, differed greatly dur-
lations between atmospheric CO2 and Ant- ing glacial times, relative to the Holocene.
arctic temperature proxies, Stephens and Gottschalk et al. (2016) isolated transient
Keeling (2000) argued that fluctuations in upward fluctuations in atmospheric CO2
the sea ice bordering Antarctica, not the during the most-recent interval of glacial
glacial meltwater from the North American climate, which occurred at irregular in-
ice sheet, govern the principal drawdown tervals of 5–10 kyr between 25 ka and 65
mechanism of CO2 from Earth’s atmosphere ka, and correlated these with indicators of
into the deep ocean. Recent oceanographic decreased primary biological production in
models for glacial-interglacial variability the Southern Ocean. These fluctuations are
A PA L E OE COLO GIC A L PE R SPE C T I V E ON SU DDE N C L I M AT E C H A NGE 107
strong evidence of a biological carbon pump 3,000 Pg C from the climate system in a far-
during glacial periods, which helps the deep shorter time. Their modeling experiments
ocean store dissolved CO2 without excessive suggest that a persistent input of carbon
acidification. Despite its importance, the to the atmosphere maintained the PETM
precise biological actors for this buffering warmth. Zeebe et al. (2009) model this
are not yet certain (Marchitto et al. 2005; Yu steady input as exogenous, but it could also
et al. 2010, 2013, 2014). The glacial and in- arise from a new balance of carbon utiliza-
terglacial extremes of the late Pleistocene ice tion within the existing range of biodiver-
ages are maintained by biochemical steady sity. The same species assemblage (for the
states, as well as by dynamic ocean-circula- most part), but with different abundances,
tion and ice-sheet balances (e.g., Brovkin et might recycle atmospheric CO2 to maintain
al. 2007; Buchanan et al. 2016). elevated concentrations. As the Anthropo-
Extinction rates in the marine realm cene proceeds, a global ecological reorga-
have been low overall during the “ice- nization could create a new, warmer “nor-
house” Earth climate interval since the Eo- mal” that will persist longer than simple
cene-Oligocene boundary (Thomas 2007; nonbiotic models predict.
Moritz and Agudo 2013). Interestingly, sub-
dued extinction rates characterized the late-
Paleozoic icehouse-climate interval as well. L E S S T H A N C ATAST ROPH E :
After the Serpukhovian biodiversity crisis, BIODI V E R SIT Y I N A C H A NGE A BL E
caused by sudden global cooling (McGhee C L I M AT E
et al. 2014), both extinction and speciation
occurred at subdued rates through the later Extinction is only one of the potential im-
Carboniferous period (Stanley and Powell pacts to biodiversity by ice-age cycles or an
2003). Milankovitch cycles have been iden- enhanced-greenhouse excursion. Human
tified in organic-rich cyclothems (Fielding civilization has developed within a stable
et al. 2008; Davydov et al. 2010) deposited Holocene climate, tempting us to interpret
by swampy Carboniferous forests, suggest- modern-day ecosystems as stable and ma-
ing that excessive carbon burial periodically ture. Both DNA evidence and comparisons
depleted Earth’s greenhouse blanket and led between theoretical and real ecosystems
to glaciation. The geographical pattern of have revealed that many modern ecosys-
glaciation changed over the 50-My duration tems had not reached a stable state before
of the late-Paleozoic ice ages (Montanez the impacts of modern civilization could
and Poulsen 2013), suggesting, as with be felt. Terrestrial ecosystems disappear
the Pleistocene, multiple steady-state bal- or transform as ice sheets wax and wane,
ances in Earth’s past carbon cycle between but biodiversity at the species and genus
ocean circulation, surface weathering, plant level is preserved in climate refugia, iso-
growth, and marine calcification. lated locations where warm-adapted or-
The concept of multiple steady climate ganisms can survive a glacial interval, or
states, mediated by separate biological in- where cold-adapted species can survive a
teractions with Earth’s carbon cycle, might interglacial interval (Svenning et al. 2008;
apply to enhanced-greenhouse episodes as Woodruff 2010). Using DNA to separate
well. Zeebe et al. (2009) argue from mod- species lineages, Hewitt (2004) argues
eling experiments that the initial burst of that many modern high-latitude Northern
greenhouse gases that started the Paleo- Hemisphere ecosystems were tethered to
cene-Eocene Thermal Maximum (PETM) specific refugia during the past glacial in-
were insufficient to explain the 170–200 tervals and have remained distinct during
kyr duration of the event. Natural sinks for the interglacials. Sandel et al. (2011) defines
atmospheric CO2 would have scrubbed even climate-change velocity as the rate at which
108 W H AT D OE S T H E PA ST T E L L US?
a climate-specialized organism must mi- In some cases evidence for widespread gla-
grate to follow the environmental condi- ciation is found instead, consistent with CO2
tions to which it has adapted. Ordonez and consumption via chemical weathering of the
Svenning (2015) argue from distributions exposed LIP eruption, which in some cases
of plant species in Europe that many spe- can overcompensate for the released CO2.
cies cannot migrate quickly enough for Over geologic time the extinction events
some ecosystems to stabilize during a typi- associated with LIP volcanism and other
cal 10–20 kyr interglacial interval. Ordonez methods of rapid carbon release have weak-
and Svenning (2016) argue that cyclic or in- ened, although impacts often were substan-
termittent climate transitions impair “func- tial in particular ecosystems. The ocean an-
tional diversity,” a metric of how efficiently oxic events (OAEs) of the Mesozoic era and
an ecosystem utilizes the trophic opportu- the hyperthermal events of the Eocene ep-
nities of its ambient environment. och elevated temperatures and CO2 levels for
The lesson offered by cyclic ice-age bio- 20–200 kyr, longer than one might expect
diversity disruptions is that we should not a single CO2 emission to persist in the at-
expect natural ecosystems to adjust quickly mosphere. We hypothesize that the climate
and completely to the Anthropocene green- system found steady-state balances between
house excursion, at least not within centu- carbon fluxes, ocean circulations, and radia-
ries or millennia. Ordonez et al. (2016) use tion balances during these enhanced green-
species information and climate projections house intervals, possibly containing biotic
to identify locations that will experience feedbacks that buffered the severity of the
novel climate conditions, regions where climate change on overall biodiversity. The
environmental gradients will migrate most glacial and interglacial climate extremes of
rapidly, and locations where temperature the recent late-Pleistocene ice ages have been
and precipitation trends will diverge, so that modeled to represent two separate such cli-
an existing species has no migration path mate-biosphere steady-state balances.
that could follow its optimal environment. In the Anthropocene, the environmental
The bumpy path that ecosystems must fol- impacts of anthropogenic greenhouse gas
low to their new locations is reflected in release are exacerbated by other ecosystem
the sedimentary record of fires during the disruptions from humans, such as chemi-
most recent glacial-to-interglacial transition cal pollution and habitat reduction and
(Marlon et al. 2009, 2013). Intervals of rapid fragmentation. We cannot feel assured that
climate change, such as near the onset and the apparent resilience of biodiversity dur-
conclusion of the Younger Dryas cooling ing the Cenozoic to sudden climate changes
interval (12.9–11.7 ka), are marked by large will prevail in the context of these addi-
increases of fire activity. tional impacts. Anthropogenic greenhouse
Nearly all the large biodiversity crises in gases are playing the role of the Large Igne-
last 541 million years of Phanerozoic Earth ous Provinces in past crises, but we humans
history have involved a severe transient dis- may also be playing the role of the end-Cre-
ruption to the carbon cycle and to the atmo- taceous meteor. There is much we still do
spheric concentration of carbon dioxide, the not understand about the detailed balances
most important terrestrial greenhouse gas. within Earth’s carbon cycle. We had better
Most large biodiversity crises correlate in learn them quickly.
time with surface eruptions of large igneous
provinces (LIPs), which release large volumes
of CO2 that were dissolved in their magma. NOTE S
Isotopic evidence for large sudden increases
in ocean temperature is common during cri- 1. Ages of geologic events and materials are denoted by ka,
Ma, and Ga. These correspond to thousands, millions, and bil-
ses, consistent with an enhanced greenhouse.
A PA L E OE COLO GIC A L PE R SPE C T I V E ON SU DDE N C L I M AT E C H A NGE 109
lions of years ago, respectively. Intervals of time are denoted by spheric CO2 change.” Global Biogeochemical Cycles 1: 15–
kyr, Myr, and Gyr (e.g., if two meteorites struck the Earth at 10 Ma 29. https://doi.org/10.1029/GB001i001p00015.
and 13 Ma, their impacts would be 3 Myr apart). The commonly Broecker, W. S. 1990. The Great Ocean Conveyor: Discovering
accepted value of Earth’s age (4.54 Ga) is bracketed by the age of
the Trigger for Abrupt Climate Change. Princeton University
meteorites in the Solar System (Dalrymple 2001), and the oldest
Press.
surviving terrestrial material (Wilde et al. 2001).
2. If we choose 0.25 ka for the onset of the Anthropocene, Brovkin, V., A. Ganopolski, D. Archer, and S. Rahmstorf.
it coincides roughly with the 1769 patent for James Watt’s steam 2007. “Lowering of glacial atmospheric CO2 in re-
engine. sponse to changes in oceanic circulation and marine
biogeochemistry.” Paleoceanography 22: PA4202. https://
doi.org/10.1029/2006PA001380.
Buchanan, P. J., R. J. Matear, A. Lenton, S. J. Phipps, Z.
R E FE R E NCE S Chase, and D. M. Etheridge. 2016. “The simulated cli-
mate of the Last Glacial Maximum and insights into
Allen, K. A., E. L. Sikes, B. Hönisch, A. C. Elmore, T. P. the global marine carbon cycle.” Climate of the Past 12:
Guilderson, Y. Rosenthal, and R. F. Anderson. 2015. 2271–2295. https://doi.org/10.5194/cp-12-2271-2016.
“Southwest Pacific deep water carbonate chemistry Burgess, S. D., S. Bowring, and S. Z. Shen. 2014. “High-
linked to high southern latitude climate and atmo- precision timeline for Earth’s most severe extinction.”
spheric CO2 during the Last Glacial Termination.” Proceedings of the National Academy of Sciences 111: 3316–3321.
Quaternary Science Review 122: 180–191. https://doi https://doi.org/10.1073/pnas.1317692111.
.org/10.1016/j.quascirev.2015.05.007. Burgess, S. D., J. D. Muirhead, and S. Bowring. 2017. “Ini-
Alley, R. 2012. The Climate Control Knob, National Climate tial pulse of Siberian Traps sills as the trigger of the
Seminar Series. Island Press. end-Permian mass extinction.” Nature Communications
Barnosky, A. D., N. Matzke, S. Tomiy, G. O. U. Wogan, 8: 164. https://doi.org/10.1038/s41467-017-00083-9.
B. Swartz, T. B. Quental, C. Marshall, J. L. McGuire, E. Caruthers, A. H., P. L. Smith, and D. R. Gröcke. 2013.
L. Lindsey, K. C. Maguire, B. Mersey, and E. A. Ferrer. “The Pliensbachian–Toarcian (Early Jurassic) extinc-
2011. “Has the Earth’s sixth mass extinction already tion, a global multi-phased event.” Palaeogeography, Pa-
arrived?” Nature 471: 51–57. https://doi.org/10.1038/ laeoclimatology, Palaeoecology 386: 104–118. https://doi
nature09678. .org/10.1016/j.palaeo.2013.05.010.
Basak, C., H. Fröllje, F. Lamy, R. Gersonde, V. Benz, R. F. Chenet, A. L., X. Quidelleur, F. Fluteau, V. Courtillot, and
Anderson, M. Molina-Kescher, and K. Pahnke. 2018. S. Bajpai. 2007. “40K– 40Ar dating of the Main Deccan
“Breakup of last glacial deep stratification in the South large igneous province: Further evidence of KTB age
Pacific.” Science 359: 900–904. https://doi.org/10.1126/ and short duration.” Earth and Planetary Science Letters 263:
science.aao2473. 1–15. https://doi.org/10.1016/j.epsl.2007.07.011.
Beaulieu, E., Y. Goddéris, D. Labat, C. Roelandt, P. Oliva, Cui, Y., and L. R. Kump. 2015. “Global warming and the
and B. Guerrero. 2010. “Impact of atmospheric CO2 end-Permian extinction event: Proxy and modeling
levels on continental silicate weathering.” Geochem- perspectives.” Earth-Science Reviews 149: 5–22. https://
istry, Geophysics, Geosystems 11: Q07007. https://doi doi.org/10.1016/j.earscirev.2014.04.007.
.org/10.1029/2010GC003078. Dalrymple, G. B. 2001. “The age of the Earth in the
Berner, R. A., A. C. Lasaga, and R. M. Garrels. 1983. “The twentieth century: A problem (mostly) solved.” In The
carbonate-silicate geochemical cycle and its effect on Age of the Earth: From 4004BC to AD 2002, ed. C. L. E. Lewis
atmospheric carbon-dioxide over the past 100 million and S. J. Knell, 190, 205–221. Geological Society, Lon-
years.” American Journal of Science 283: 641–683. https:// don. https://doi.org/10.1144/GSL.SP.2001.190.01.14.
doi.org/10.2475/ajs.283.7.641. Davydov, V. I., J. L. Crowley, M. D. Schmitz, and V. I.
Bond, D. P., and P. B. Wignall. 2014. “Large igneous prov- Poletaev. 2010. “High-precision U-Pb zircon age cali-
inces and mass extinctions: An update.” In Volcanism, bration of the global Carboniferous time scale and Mi-
Impacts, and Mass Extinctions: Causes and Effects, ed. G. Keller lankovitch band cyclicity in the Donets Basin, eastern
and A. C. Kerr, Geological Society of America Special Papers, Ukraine.” Geochemistry, Geophysics, Geosystems 11: Q0AA04.
505: 29–55. https://doi.org/10.1130/2014.2505(02). https://doi.org/10.1029/2009GC002736.
Bond, D. P., and S. E. Grasby. 2017. “On the causes D’Hondt, S. 2005. “Consequences of the Cretaceous/
of mass extinctions.” Palaeogeography, Palaeoclimatol- Paleogene mass extinction for marine ecosys-
ogy, Palaeoecology 478: 3–29. https://doi.org/10.1016/j tems.” Annual Review of Ecology, Evolution, and Systemat-
.palaeo.2016.11.005. ics, 36: 295–317, https://doi.org/10.1146/annurev
Bottini, C., A. S. Cohen, E. Erba, H. C. Jenkyns, and A. .ecolsys.35.021103.105715.
L. Coe. 2012. “Osmium-isotope evidence for vol- Dickens, G. R. 2011. “Down the rabbit hole: Toward
canism, weathering, and ocean mixing during the appropriate discussion of methane release from gas
early Aptian OAE 1a.” Geology 40: 583–586. https://doi hydrate systems during the Paleocene-Eocene ther-
.org/10.1130/G33140.1. mal maximum and other past hyperthermal events.”
Broecker, W. S., and T. H. Peng. 1987. “The role of CaCO3 Climate of the Past 7: 831–846. https://doi.org/10.5194/
compensation in the glacial to interglacial atmo- cp-7-831-2011.
110 W H AT D OE S T H E PA ST T E L L US?
Dirzo, R., H. S. Young, M. Galetti, G. Ceballos, N. J. Isaac, tween climate and weathering.” Earth and Planetary
and B. Collen. 2014. “Defaunation in the Anthropo- Science Letters 277: 213–222. https://doi.org/10.1016/j
cene.” Science 345: 401–406. https://doi.org/10.1126/ .epsl.2008.10.018.
science.1251817. Gómez, J. J., and A. Goy. 2011. “Warming-driven mass
Doney, S. C., V. J. Fabry, R. A. Feely, and J. A. Kleypas. extinction in the Early Toarcian (Early Jurassic) of
2009. “Ocean acidification: The other CO2 problem.” northern and central Spain: Correlation with other
Annual Review of Marine Science 1: 169–192. https://doi time-equivalent European sections.” Palaeogeography,
.org/10.1146/annurev.marine.010908.163834. Palaeoclimatology, Palaeoecology 306: 176–195. https://doi
Doney, S. C., L. Bopp, and M. C. Long. 2014. “Histori- .org/10.1016/j.palaeo.2011.04.018.
cal and future trends in ocean climate and biogeo- Gottschalk, J., L. C. Skinner, J. Lippold, H. Vogel, N.
chemistry.” Oceanography 27: 108–119. https://doi.org/ Frank, S. L. Jaccard, and C. Waelbroeck. 2016. “Bio-
10.5670/oceanog.2014.14. logical and physical controls in the Southern Ocean on
Eldrett, J. S., C. Ma, S. C. Bergman, A. Ozkan, D. Mini- past millennial-scale atmospheric CO2 changes.” Na-
sini, B. Lutz, S. J. Jackett, C. Macaulay, and A. E. Kelly. ture Communications 7: 11539. https://doi.org/10.1038/
2015. “Origin of limestone–marlstone cycles: As- ncomms11539.
tronomic forcing of organic-rich sedimentary rocks Gradstein, F. M., and J. G. Ogg. 2012. “The chrono-
from the Cenomanian to early Coniacian of the Creta- graphic scale.” In The Geologic Time Scale 2012, ed.
ceous Western Interior Seaway, USA.” Earth and Planetary F. M. Gradstein, J. G. Ogg, M. Schmitz, and G.
Science Letters 423: 98–113. https://doi.org/10.1016/j Ogg, 31–42. Elsevier. https://doi.org/10.1016/
.epsl.2015.04.026. B978–0-444-59425-9.00002-0.
Erba, E. 2004. “Calcareous nannofossils and Mesozoic Henehan, M. J., P. M. Hull, D. E. Penman, J. W. Rae, and
oceanic anoxic events.” Marine Micropaleontology 52: 85– D. N. Schmidt. 2016. “Biogeochemical significance
106. https://doi.org/10.1016/j.marmicro.2004.04.007. of pelagic ecosystem function: An end-Cretaceous
Ernst, R. E., and N. Youbi. 2017. “How Large Igneous case study.” Philosophical Transactions of the Royal Society
Provinces affect global climate, sometimes cause (Series B) 371: 20150510. https://doi.org/10.1098/
mass extinctions, and represent natural markers in rstb.2015.0510.
the geological record.” Palaeogeography, Palaeoclimatol- Hewitt, G. M. 2004. “Genetic consequences of climatic
ogy, Palaeoecology 478: 30–52. https://doi.org/10.1016/j oscillations in the Quaternary.” Philosophical Transactions
.palaeo.2017.03.014. of the Royal Society (Series B) 359: 183–195. https://doi
Erwin, D. H. 2015. “Early metazoan life: Divergence, .org/10.1098/rstb.2003.1388.
environment and ecology.” Philosophical Transactions Hoffman, P. F., A. J. Kaufman, G. P. Halverson, and D.
of the Royal Society (Series B) 370: 20150036. https://doi P. Schrag. 1998. “A Neoproterozoic snowball earth.”
.org/10.1098/rstb.2015.0036. Science 281: 1342–1346. https://doi.org/10.1126/
Fielding, C. R., T. D. Frank, and J. L. Isbell. 2008. “The science.281.5381.1342.
late Paleozoic ice age—A review of current under- Hönisch, B., T. Bickert, and N. G. Hemming. 2008.
standing and synthesis of global climate patterns.” In “Modern and Pleistocene boron isotope composi-
Resolving the Late Paleozoic Ice Age in Time and Space, ed. C. tion of the benthic foraminifer Cibicidoides wuellerstorfi.”
R. Fielding, T. D. Frank, and J. L. Isbell, Geological So- Earth and Planetary Science Letters 272: 309–318. https://doi
ciety of America Special Papers, 441: 343–354. https://doi .org/10.1016/j.epsl.2008.04.047.
.org/10.1130/2008.2441(24). Houghton, J. 2015. Global Warming: The Complete Briefing. 5th
Finnegan, S., K. Bergmann, J. M. Eiler, D. S. Jones, D. ed. Cambridge University Press.
A. Fike, I. Eisenman, N. C. Hughes, A. K. Tripati, and Howe, J. N. W., A. M. Piotrowski, T. L. Noble, S. Mu-
W. W. Fischer. 2011. “The magnitude and duration of litza, C. M. Chiessi, and G. Bayon. 2016. “North At-
Late Ordovician–Early Silurian glaciation.” Science 331: lantic deep water production during the last glacial
903–906. https://doi.org/10.1126/science.1200803. maximum.” Nature Communications 7: 11765. https://doi
Franks, P. J., D. L. Royer, D. J. Beerling, P. K. Van de Wa- .org/10.1038/ncomms11765.
ter, D. J. Cantrill, M. M. Barbour, and J. A. Berry. 2014. Hren, M. T., N. D. Sheldon, S. T. Grimes, M. E. Col-
“New constraints on atmospheric CO2 concentra- linson, J. J. Hooker, M. Bugler, and K. C. Lohmann.
tion for the Phanerozoic.” Geophysical Research Letters 41: 2013. “Terrestrial cooling in Northern Europe dur-
4685–4694. https://doi.org/10.1002/2014GL060457. ing the Eocene–Oligocene transition.” Proceedings of the
Friedrich, O., R. D. Norris, and J. Erbacher. 2012. “Evolu- National Academy of Sciences 110: 7562–7567. https://doi
tion of middle to Late Cretaceous oceans: A 55 m.y. re- .org/10.1073/pnas.1210930110.
cord of Earth’s temperature and carbon cycle.” Geology Hull, P. M., R. D. Norris, T. J. Bralower, and J. D. Schueth.
40: 107–110. https://doi.org/10.1130/G32701.1. 2011. “A role for chance in marine recovery from the
Gislason, S. R., E. H. Oelkers, E. S. Eiriksdottir, M. I. end-Cretaceous extinction.” Nature Geoscience 4: 856–
Kardjilov, G. Gisladottir, B. Sigfusson, A. Snorrason, 860. https://doi.org/10.1038/NGEO1302.
S. Elefsen, J. Hardardottir, P. Torssander, and N. Os- Imbrie, J., E. A. Boyle, S. C. Clemens, A. Duffy, W. R.
karsson. 2009. “Direct evidence of the feedback be- Howard, G. Kukla, J. Kutzbach, D. G. Martinson, A.
A PA L E OE COLO GIC A L PE R SPE C T I V E ON SU DDE N C L I M AT E C H A NGE 111
McIntyre, A. C. Mix, B. Molfino, J. J. Morley, L. C. Knoll, A. H., R. K. Bambach, J. L. Payne, S. Pruss, and W.
Peterson, N. G. Pisias, W. L. Prell, M. E. Raymo, N. J. W. Fischer. 2007. “Paleophysiology and end-Permian
Shackleton, and J. R. Toggweiler. 1992. “On the struc- mass extinction.” Earth and Planetary Science Letters 256:
ture and origin of major glaciation cycles 1: Linear 295–313. https://doi.org/10.1016/j.epsl.2007.02.018.
responses to Milankovitch forcing.” Paleoceanography 7: Krencker, F. N., S. Bodin, R. Hoffmann, G. Suan, E. Mat-
701–738. https://doi.org/10.1029/92PA02253. tioli, L. Kabiri, K. B. Föllmi, and A. Immenhauser.
Imbrie, J., A. Berger, E. A. Boyle, S. C. Clemens, A. Duffy, 2014. “The middle Toarcian cold snap: Trigger of mass
W. R. Howard, G. Kukla, J. Kutzbach, D. G. Martin- extinction and carbonate factory demise.” Global and
son, A. Mcintyre, A. C. Mix, B. Molfino, J. J. Morley, Planetary Change 117: 64–78. https://doi.org/10.1016/j
L. C. Peterson, N. G. Pisias, W. L. Prell, M. E. Raymo, .gloplacha.2014.03.008.
N. J. Shackleton, and J. R. Toggweiler. 1993. “On the Lacis, A. A., G. A. Schmidt, D. Rind, and R. A. Ruedy.
structure and origin of major glaciation cycles 2: 2010. “Atmospheric CO2: Principal control knob gov-
The 100,000-year cycle.” Paleoceanography 8: 699–735. erning Earth’s temperature.” Science 330: 356–359.
https://doi.org/10.1029/93PA02751. https://doi.org/10.1126/science.1190653.
Irvine, P. J., A. Ridgwell, and D. J. Lunt. 2010. “Assess- Lanci, L., G. Muttoni, and E. Erba. 2010. “Astronomical
ing the regional disparities in geoengineering im- tuning of the Cenomanian Scaglia Bianca Formation
pacts.” Geophysical Research Letters 37: L18702. https://doi at Furlo, Italy.” Earth and Planetary Science Letters 292: 231–
.org/10.1029/2010GL044447. 237. https://doi.org/10.1016/j.epsl.2010.01.041.
Isaacson, P. E., E. Diaz-Martinez, G. W. Grader, J. Kalvoda, Laurin, J., S. R. Meyers, S. Galeotti, and L. Lanci. 2016.
O. Babek, and F. X. Devuyst. 2008. “Late Devonian– “Frequency modulation reveals the phasing of orbital
earliest Mississippian glaciation in Gondwanaland eccentricity during Cretaceous Oceanic Anoxic Event
and its biogeographic consequences.” Palaeogeography, II and the Eocene hyperthermals.” Earth and Planetary
Palaeoclimatology, Palaeoecology 268: 126–142. https://doi Science Letters 442: 143–156. https://doi.org/10.1016/j
.org/10.1016/j.palaeo.2008.03.047. .epsl.2016.02.047.
Ivany, L. C., W. P. Patterson, and K. C. Lohmann. 2000. Levin, S. A. 1998. “Ecosystems and the biosphere as
“Cooler winters as a possible cause of mass extinc- complex adaptive systems.” Ecosystems 1: 431–436.
tions at the Eocene/Oligocene boundary.” Nature 407: https://doi.org/10.1007/s100219900037.
887–890. https://doi.org/10.1038/35038044. Liu, Z., M. Pagani, D. Zinniker, R. DeConto, M. Huber,
Jaccard, S. L., E. D. Galbraith, A. Martínez-García, and H. Brinkhuis, S. R. Shah, R. M. Leckie, and A. Pearson.
R. F. Anderson. 2016. “Covariation of deep southern- 2009. “Global cooling during the Eocene-Oligocene
ocean oxygenation and atmospheric CO2 through climate transition.” Science 323: 1187–1190. https://doi
the last ice age.” Nature 530: 207–210. https://doi .org/10.1126/science.1166368.
.org/10.1038/naturel6514. Maasch, K. A., and B. Saltzman. 1990. “A low-order dy-
Jenkyns, H. C. 2010. “Geochemistry of oceanic anoxic namical model of global climatic variability over the
events.” Geochemistry, Geophysics, Geosystems 11: Q03004, full Pleistocene.” Journal of Geophysical Research 95: 1955–
https://doi.org/10.1029/2009GC002788. 1963. https://doi.org/10.1029/JD095iD02p01955.
Joachimski, M. M., X. Lai, S. Shen, H. Jiang, G. Luo, B. Marchitto, T. M., J. Lynch-Stieglitz, and S. R. Hemming.
Chen, J. Chen, and Y. Sun. 2012. “Climate warm- 2005. “Deep Pacific CaCO3 compensation and gla-
ing in the latest Permian and the Permian–Triassic cial–interglacial atmospheric CO2.” Earth and Planetary
mass extinction.” Geology 40: 195–198. https://doi Science Letters 231: 317–336. https://doi.org/10.1016/j
.org/10.1130/G32707.1. .epsl.2004.12.024.
Jones, D. S., A. M. Martini, D. A. Fike, and K. Kaiho. 2017. Marlon, J. R., P. J. Bartlein, M. K. Walsh, S. P. Harrison,
“A volcanic trigger for the Late Ordovician mass extinc- K. J. Brown, M. E. Edwards, P. E. Higuera, M. J. Power,
tion? Mercury data from south China and Laurentia.” R. S. Anderson, C. Briles, A. Brunelle, C. Carcaillet, M.
Geology 45: 631–634. https://doi.org/10.1130/G38940.1. Daniels, F. S. Hu, M. Lavoie, C. Long, T. Minckley, P.
Jouzel, J. 2013. “A brief history of ice core science over J. H. Richard, A. C. Scott, D. S. Shafer, W. Tinner, C.
the last 50 yr.” Climate of the Past 9: 2525–2547. https:// E. Umbanhowar Jr., and C. Whitlock. 2009. “Wildfire
doi.org/10.5194/cp-9-2525-2013. responses to abrupt climate change in North Amer-
Keeling, R. F., and B. B. Stephens. 2001. “Antarctic ica.” Proceedings of the National Academy of Sciences 106: 2519–
sea ice and the control of Pleistocene climate in- 2524. https://doi.org/10.1073/pnas.0808212106.
stability.” Paleoceanography 16: 112–131. https://doi Marlon, J. R., P. J. Bartlein, A. L. Daniau, S. P. Harri-
.org/10.1029/2000PA000529. son, S. Y. Maezumi, M. J. Power, W. Tinner, and B.
Kidder, D. L., and T. R. Worsley. 2010. “Phanerozoic Vanniére. 2013. “Global biomass burning: A synthesis
large igneous provinces (LIPs), HEATT (haline eux- and review of Holocene paleofire records and their
inic acidic thermal transgression) episodes, and controls.” Quaternary Science Reviews 65: 5–25. https://doi
mass extinctions.” Palaeogeography, Palaeoclimatology, Pa- .org/10.1016/j.quascirev.2012.11.029.
laeoecology 295: 162–191. https://doi.org/10.1016/j McGhee, G. R., M. E. Clapham, P. M. Sheehan, D. J. Bot-
.palaeo.2010.05.036. tjer, and M. L. Droser. 2013. “A new ecological-severity
112 W H AT D OE S T H E PA ST T E L L US?
114
C L I M AT E C H A N G E , C O N S E RVAT I O N , A N D T H E M E TA P H O R O F D E E P T I M E 115
environments illustrate the heuristic value which multicellular life as we know it has
of the time stamp. Climate change is also dominated the global biota. The Phanero-
shifting the time stamp of benthic com- zoic succeeded the Proterozoic eon (the
munities in Antarctica. Because Antarctica “Precambrian”), which ran from 2,500–541
is one of the last places on Earth to remain Ma. This brief summary of the Phanerozoic
in near-pristine condition, the time-shifting history of marine life draws from reviews
impacts of climate and the policy options by Sepkoski (1991) and Sheehan (2001). The
available are instructive and cautionary. focus is on nearshore, shallow-marine, soft-
bottom habitats, which have a fossil record
that is abundant, detailed, and well studied.
M A R I N E COM M U N ITIE S I N DE E P TIM E Sepkoski partitioned marine diversity
into three broad groups, or “evolutionary
The Phanerozoic eon, which extends from faunas,” that were globally dominant at
about 541 Ma (millions of years ago) to successive times during the Phanerozoic:
the present (Box 9.1), is the interval during a Cambrian fauna, a Paleozoic fauna, and
Geologic time is divided hierarchically into eons, eras, periods, epochs, and ages. The
abridged table below lists the names and time spans of the divisions relevant to this
chapter. Ma denotes millions of years ago, with dates based on the US Geological Sur-
vey’s 2012 stratigraphic chart. The Anthropocene, a developing concept for an epoch of
geologically discernible human influence, began in the nineteenth century or earlier.
Modern fauna. The Cambrian fauna diversi- Evolutionary novelties, notably includ-
fied during the Cambrian period (541–485 ing the modern, durophagous predators,
Ma) and was dominated by trilobites, which evolved primarily in nearshore, coastal hab-
consumed organic matter at the sedimentary itats and then expanded offshore to outer-
surface. Other important elements included shelf and bathyal (upper-deep-sea) environ-
low-lying suspension feeders, such as inar- ments (Jablonski et al. 1983). The Modern
ticulate brachiopods (lamp shells) (suspen- fauna progressively displaced many taxa of
sion feeders are animals that feed on organic the Paleozoic fauna, especially stalked cri-
particles, including plankton, bacteria, and noids and other epifaunal suspension feed-
detritus, in the water column). By the end of ers, to offshore, deepwater environments.
the Paleozoic, 252 Ma, the Cambrian fauna Today, the deep sea is the refuge of Paleo-
was reduced to a minor component of ma- zoic-type communities and “living fossils”
rine diversity, and trilobites were extinct. such as the coelacanth Latimeria, the nauti-
The Paleozoic fauna initially diversified lus, and stalked crinoids. Dense populations
in the Ordovician Period (485–444 Ma) of epifaunal ophiuroids occur in outer-shelf
and was dominated by epifaunal suspen- and bathyal environments (Metaxas and
sion-feeders that fed higher above the sea- Griffin 2004; Aronson 2017). “Brittlestar
floor. Particularly abundant were articulate beds” also persist in a few coastal locations
brachiopods (lamp shells designated as characterized by low levels of durophagous
rhynchonelliforms), stalked crinoids (sea predation, either because predators are
lilies), stromatoporoids (an archaic group naturally absent or as a likely consequence
of sponges), rugose and tabulate corals (the of centuries of overfishing (Aronson 1989).
extinct horn and table corals), bryozoans The most prevalent predators within those
(moss animals), and ophiuroids (brittle- ophiuroid populations are seastars and
stars). The primary predators were asteroids other Paleozoic-grade invertebrates.
(sea stars), polychaetes (marine worms), The increase in predation that began in
shelled cephalopods (similar to today’s nau- the Mesozoic modernized benthic com-
tilus), placoderms (ancient armored fish), munities in nearshore environments. New,
and primitive cartilaginous fish (relatives of durophagous predators were the primary
modern sharks and rays). Burrowing organ- drivers of several important trends: (1) the
isms, or bioturbators, expanded downward evolution of increased defensive architec-
into the sediments. ture in mollusks and other skeletonized
The Modern fauna diversified during the invertebrates; (2) the loss of dense popula-
Mesozoic (252–66 Ma) and Cenozoic (66 tions of epifaunal suspension feeders from
Ma–present). Gastropods (snails), infaunal nearshore, shallow, soft-bottom environ-
bivalves (burrowing clams), and echinoids ments; and (3) the transition from epifau-
(sea urchins and sand dollars) are promi- nal to infaunal life modes in bivalves. This
nent constituents. Skeleton-crushing preda- “Mesozoic marine revolution” (Vermeij
tors diversified and became the trophic 1977) transcended the mass extinction at
dominants. These “durophagous” predators the end of the Cretaceous and accelerated
include teleosts (modern bony fish), neose- during the Cenozoic. Although it is a top-
lachians (modern sharks and rays), reptant down construct, its macroevolutionary dy-
decapods (large, walking crustaceans such namics were fueled from the bottom up by
as crabs and lobsters), coleoids (shell-less an increased supply of energy from plank-
cephalopods, including octopus and squid), ton in the water column (Knoll and Follows
marine mammals, and now-extinct marine 2016). These trends are expressed to vary-
reptiles such as mosasaurs. Bioturbators ing degrees and at different times in dif-
burrowed deeper and more actively than ferent geographic regions (Monarrez et al.
their Paleozoic counterparts. 2017; Whittle et al. 2018).
C L I M AT E C H A N G E , C O N S E RVAT I O N , A N D T H E M E TA P H O R O F D E E P T I M E 117
C L IM AT E A N D BIODI V E R SIT Y different times reflect the temporal scales
I N DE E P TI M E over which the temperature changes oc-
curred (along with the associated changes
Direct impacts of climate change on marine in sea level, acidification, and other factors),
biodiversity have been difficult to discern the departures of the thermal shifts from
over deep time. Some episodes of elevated background conditions, the temporal reso-
extinction have been causally linked to lution of the fossil record in different time
trends of heating or cooling. Examples, in frames, the prevailing composition of the
temporal order, include cooling in the Late biosphere, and other events at the time. The
Ordovician mass extinction, circa 450 Ma; variability of climatic impacts in the past
the late Paleozoic ice age, which began in notwithstanding, contemporary climate
the Mississippian, circa 330 Ma; high-tem- change is altering living communities in
perature conditions 252–250 Ma associated ways the fossil record can illuminate.
with the end-Permian mass extinction;
cooling at the Eocene–Oligocene transition,
34 Ma; and cooling at the Pliocene–Pleisto- T H E M E TA PHOR IC A L TIM E STA M P
cene transition, 2.6 Ma. At other times, bio-
logical patterns have been driven by other The idea of time stamping marine commu-
physical changes or have been complicated nities is founded on three broad generaliza-
by the ecology of the taxa on which climate tions. The first is the primacy of top-down
is acting. Ocean acidification, for example, control of marine food webs. Strong, cas-
has played a critical role in the history of cading, predatory impacts are readily ap-
marine biodiversity (Norris et al. 2013). parent in many or most nearshore marine
Glaciations that occurred 850–635 Ma, ecosystems (Estes et al. 2011). The second is
near the end of the Proterozoic, were the that top predators, being the largest, most
most intense in Earth’s history. These Va- energetic, and least abundant members of
rangian glaciations, which were followed marine communities, are more vulner-
by climatic warming, may have delayed— able to direct exploitation by humans than
or perhaps accelerated!—the diversification are lower-level consumers and producers
of animal life in the Cambrian by causing (Pauly et al. 1998; Estes et al. 2011). The
mass extinctions of the existing faunas, al- third is that trophic levels have been added
though rising oxygen levels, burial of or- sequentially to marine-benthic communi-
ganic carbon, and tectonic shifts probably ties through Phanerozoic time, as we have
played a more central role in the timing of already seen. Combining these generaliza-
the “Cambrian explosion.” In contrast, the tions, we can view the cascading impacts
Pleistocene glaciations did not drive mass of predation through the lens of paleobiol-
extinctions in the ocean, but they altered ogy. Artificially reducing the abundance of
the geographic ranges of marine species top predators through exploitation, climate
and, therefore, changed the composition change, and other impacts can metaphori-
of marine communities (Roy and Pandolfi cally force a community back in time to an
2005). Likewise, although benthic forami- anachronistic state with a retrograde struc-
niferans (bottom-dwelling, shelled amoe- ture. Alternatively, perturbation can create
bas) experienced high rates of extinction new, no-analogue communities (Williams
during the warming trend of the Paleo- and Jackson 2007; Jackson 2008), which,
cene–Eocene Thermal Maximum (PETM; maintaining the metaphor, could be con-
ca. 56 Ma), other groups of small marine sidered in some sense futuristic as opposed
organisms responded primarily by shift- to anachronistic. The metaphor can be
ing their geographic ranges (Norris et al. extended to changes in the abundance of
2013). The different responses of biotas at strongly interacting consumers at lower tro-
118 W H AT D OE S T H E PA ST T E L L US?
phic levels, as well as to nontrophic classes which were the basis of the earliest living
of strong interactors such as foundation reefs until they were outmoded by the rapid
species, or ecosystem engineers. diversification of mobile consumers in the
Stamping a living community with Cambrian. Modern, stromatolite-dominated
its time equivalent from the fossil record seascapes are likewise characterized by low
should not be taken to mean that the liv- levels of grazing. Moreover, anachronistic
ing community fully replicates a particu- communities dominated by microbially
lar paleocommunity from the distant past. built structures reappeared at various times
That would be impossible after millions to in the Phanerozoic following the clearance
hundreds of millions of years of change in of ecospace by extinction events (Sheehan
the ocean. Rather, the time stamp denotes and Harris 2004). Whether coral reefs are
some commonality of pattern and process, degrading to a Proterozoic state and on what
yielding clues to anthropogenic causation schedule remain contentious issues. The
and suggesting remedial actions. The intent complementary prediction for pelagic sys-
is to draw a revealing caricature, not one tems is that gelatinous zooplankton will in-
that is overly exaggerated. As in politics and herit the water column after all the fish are
life generally, absurd caricatures are not so gone. The hypothesis of retrogradation to a
useful for identifying the features that need Proterozoic-style “ocean of jelly” has found
improvement. only partial support (Condon et al. 2012).
The time stamp is implicit in interpret-
ing the Pleistocene extinctions of terres-
trial megafauna. Neotropical fruiting plants OV E R F ISH I NG A N D POST-MODE R N
with large seeds, bereft of the megafaunal COM M U N ITIE S
dispersers with which they coevolved, now
live in a no-analogue world but find some The increased abundance of epifaunal
temporal redress in the form of large, do- ophiuroids in overfished coastal habitats il-
mesticated animals such as cattle and horses lustrates how removing top predators can
(Janzen and Martin 1982; Guimarães et al. drive benthic communities in a Paleozoic
2008). Levin et al. (2002) likewise argued direction. Human exploitation of predators
that grazing by feral horses at least partially such as fish and sea otters can also lead to
re-creates the impacts of Pleistocene mega- explosive increases in sea urchins. Sea ur-
herbivores in North American salt marshes. chins, which belong to Sepkoski’s Modern
Rewilding (Donlan et al. 2006) is an explicit fauna, are potent herbivores. When released
management proposal to restore North from predation pressure they can defoliate
American ecosystems to some semblance of expansive beds of seagrasses and macroal-
their prehuman states by importing living gae in shallow-water environments. Aron-
megafauna from other continents. son (1990) distinguished modern, temper-
The degradation of coral reefs provides ate-zone communities, which are naturally
an example from the marine realm. Cli- dominated by predators and macroalgae,
mate change, overharvesting, and nutri- from overfished, “post-modern” communi-
ent loading all kill corals and promote the ties dominated by humans and sea urchins.
growth of macroalgae (seaweeds) on reefs. On tropical reefs, in contrast, cultivating ar-
Some ecologists predict that reef commu- tificially dense, post-modern populations of
nities will ultimately degrade to a “seabed sea urchins might be a useful management
of slime,” dominated by expansive mats of tool to combat the displacement of corals by
cyanobacteria (i.e., blue-green algae) remi- macroalgae and cyanobacterial mats (Precht
niscent of the Proterozoic seas (Pandolfi et and Aronson 2006).
al. 2003; Jackson 2008). Cyanobacterial mats Aronson (1990) broadened the onshore-
formed hardened, layered stromatolites, offshore model to incorporate human ex-
C L I M AT E C H A N G E , C O N S E RVAT I O N , A N D T H E M E TA P H O R O F D E E P T I M E 119
ploitation. Fishing activity by Homo sapiens, climate change (Aronson et al. 2011; Chown
an evolutionary novelty that originated in et al. 2015). Apart from endemism per se,
nearshore environments, sequentially alters marine-benthic communities on the con-
community structure along the onshore-to- tinental shelves surrounding Antarctica
offshore gradient. Food webs progressively are unique in their structure and function.
become depleted from their top predators Declining temperatures over the last 40
downward (Pauly et al. 1998) and from million years have placed a Paleozoic time
coastal habitats to the edge of the conti- stamp on the living benthic communities,
nental shelf and into the bathyal zone. The but now rapidly warming seas are poised to
onshore-offshore trend is reflected in pro- reverse that dynamic.
jections that bottom fisheries will continue The living bottom-fauna in Antarctica
to expand to the slow-growing, highly vul- lacks the durophagous teleosts, neoselachi-
nerable stocks that live in deep-sea environ- ans, and reptant decapods that generally ex-
ments, with obvious implications for ma- ert top-down control at temperate, tropical,
rine policy and conservation planning. and Arctic latitudes. As a result, shallow-
The post-modern time stamp also ap- benthic communities in Antarctica resem-
plies to overfished demersal and pelagic ble benthic communities of the Paleozoic
assemblages in which coleoid cephalopods marine world, as well as contemporary off-
are increasing. Like sea urchins, coleoids are shore or deep-sea communities (Aronson et
components of the Modern fauna, and they al. 2007; Aronson 2017). Living macrofau-
are remarkably convergent on fish in their nal assemblages on soft bottoms in Antarc-
structure, function, and ecology (Packard tica are dominated by ophiuroids, crinoids,
1972). Catches of octopus, squid, and cut- anthozoans (sea anemones and their rela-
tlefish have increased recently, in large part tives), sponges, and other epifaunal suspen-
because the teleosts that are their predators sion feeders. Brachiopods are common on
and competitors have been overfished by rocky outcrops in some localities. The pri-
humans (Doubleday et al. 2016). mary predators are slow moving, Paleozoic-
Jackson (2008) expressed the same dark grade invertebrates, including asteroids, gi-
sentiment that marine communities have ant nemerteans (ribbon worms), and giant
entered post-modern times with his dys- pycnogonids (sea spiders). The thin, unde-
topian reference to the brave new ocean. fended shells of Antarctic gastropods are
Whether we view perturbed marine com- more vulnerable to shell-crushing predators
munities as retrograde fragments or tempo- than their counterparts in the temperate
ral chimeras, we can best understand the zones and the tropics, a latitudinal analogue
bleak future of life in the sea as a warped of the Mesozoic marine revolution (Vermeij
and shattered reflection of the distant past. 1978; Watson et al. 2017). Other retrograde,
Antarctica provides a vivid case study of Paleozoic features include a phylogeneti-
the value of the metaphorical time stamp cally archaic bivalve fauna and, related, an
for managing a highly vulnerable marine impoverished infauna (Crame 2014).
ecosystem that is being altered by climate Climatic cooling began in Antarctica in
change. the late Eocene with the opening of the
Drake Passage and establishment of the
Antarctic Circumpolar Current (ACC). The
E O C E N E C L I M AT E C H A NGE A N D cold and isolation created by the ACC drove
A N TA RCTIC M A R I N E COM M U N ITIE S the trend toward anachronistic community
structure in the Antarctic shelf fauna. Geo-
Antarctic ecosystems are extraordinarily chemical analysis of bivalve shells from the
sensitive to human impacts, and a high de- Eocene La Meseta Formation at Seymour
gree of endemism aggravates the threat of Island, in the Weddell Sea off the eastern
120 W H AT D OE S T H E PA ST T E L L US?
coast of the Antarctic Peninsula, suggest a remodernizing the Antarctic benthos by re-
temperature drop around 41 Ma (Douglas admitting durophagous predators.
et al. 2014). That event was the first pulse
of the long-term trend that led to glaciation
of the continent and its polar climate as we T H E F U T U R E OF A N TA RCTIC
know it today. MARINE LIFE
The La Meseta Formation, which spans
~55 Ma to 33.5 Ma—most of the Eocene— Much has been written about the ecologi-
is a highly fossiliferous, uplifted channel- cal impacts of ocean acidification at polar
fill that preserves faunas from nearshore, latitudes, but warming seas pose a more
shallow-marine environments. The ben- immediate threat to the continued survival
thic communities were dominated by gas- of benthic species in Antarctica (Gutt et
tropods and bivalves, typical of Cenozoic al. 2015). Because water temperatures are
nearshore communities worldwide. Far virtually aseasonal, marine ectotherms in
from typical, however, was the abrupt ap- Antarctica have evolved physiologies that
pearance of dense populations of epifaunal are cold stenothermal: they tolerate a nar-
ophiuroids and crinoids following the cool- row range of low temperatures. Warming
ing pulse 41 Ma. Other changes across the directly threatens Antarctic ectotherms
cooling event included a decline of sharks, with extinction through physiological
a decline of defensive architecture in gas- stress (Pörtner 2006; see also Griffiths et
tropods, and an increase of epifaunal, sus- al. 2017). Rising temperatures also act in-
pension-feeding bivalves (Long 1992; Stil- directly, accelerating biological invasions
well and Zinsmeister 1992). Durophagous and range expansions from warmer envi-
teleosts, sharks, and decapods went extinct ronments (e.g., Walther et al. 2002). As a
in Antarctica after the Eocene-Oligocene consequence, durophagous crustaceans ap-
boundary. The notothenioid fish, which pear to be on the verge of returning to shelf
survive in Antarctica by producing anti- communities off the western Antarctic Pen-
freeze glycoproteins, radiated as early as the insula (WAP).
late Eocene in response to cooling (Near et The continental shelf of the WAP is the
al. 2012). They are the only remaining te- marine region of Antarctica most suscep-
leosts in shelf environments in Antarctica, tible to biological invasion for several rea-
and they never evolved to feed on hard- sons. First, the WAP is close to source popu-
shelled prey. This functional idiosyncrasy is lations of potential invaders from southern
an accident of phylogeny not shared with South America. A vagrant crab from South
Arctic teleosts, many of which produce America, likely transported in the ballast
antifreeze compounds and are also shell of a tourist ship, was recently discovered at
crushers (Aronson et al. 2007). Deception Island, and crab larvae are drift-
The faunal response to cooling 41 Ma— ing southward into the Southern Ocean in
retrogradation to a quasi-Paleozoic state— eddies coming off the ACC (Aronson, Fred-
was an initial exemplar of later, widespread erich, et al. 2015). Second, sea surface tem-
ecological changes to the Antarctic benthos. peratures over the shelf adjacent to the WAP
Fossil evidence suggests that subsequent have risen approximately 1.5°C during the
warming and cooling were accompanied by past 50 years, which is considerably more
fluctuations in the time stamp of commu- rapid than the global average (Schmidtko
nity structure. At some point, however, sea et al. 2014). Third, the near-bottom waters
temperatures were uniformly cold enough over the continental slope of the WAP are
that retrograde communities became the slightly warmer than the shallow waters
norm. Rapidly warming seas now appear over the inner shelf. This oceanographic
to be erasing the Paleozoic time stamp and peculiarity arises from downwelling within
C L I M AT E C H A N G E , C O N S E RVAT I O N , A N D T H E M E TA P H O R O F D E E P T I M E 121
Lithodids prey on invertebrates, includ-
ing echinoderms and mollusks. They and
their prey exhibit complementary depth
distributions in Palmer Deep and on the
continental slope where they occur at high
abundance (C. Smith et al. 2012; K. Smith
et al. 2017). Nearshore waters are still too
cold for lithodids to survive, but the outer
shelf, at 400–600 m depth, is already warm
enough. At current rates of warming, no
physiological barriers will prevent lithodids
from expanding shoreward across the shelf
over the next few decades. The abundant,
Figure 9.1. Annual temperature range within shallow-
marine habitats as a function of latitude. The lack of
naïve, undefended echinoderms, mollusks,
thermal seasonality has driven the evolution of cold and other invertebrates will be easy pick-
stenothermy in Antarctic ectotherms. Ectotherms living ings, compromising the Paleozoic character
near the equator tend to be warm stenothermal, again of the Antarctic shelf faunas.
because of reduced seasonality. Temperate-marine ecto-
therms are eurythermal in comparison. Extinction risks
are elevated in Antarctica from the physiological stresses
of climate change. (Modified from Aronson et al. 2011.) SU M M A RY A N D CONC LUSIONS
piece et al. 2011), but climatically induced ACK NOWLE DGME NTS
reductions in krill are having a significant
impact on the abundance of penguins. I thank J. Eastman, K. Heck, L. Ivany, J. Mc-
There is only so far south along the WAP Clintock, R. Moody, W. Precht, K. Smith,
that penguins and other marine species can and S. Thatje for helpful advice and dis-
shift their ranges before running into the cussion. Comments from the editors and
continent and facing extinction. several anonymous reviewers greatly im-
The direct climatic threat to marine bio- proved the manuscript. Contribution 140
diversity on a regional level within Antarc- from the Institute for Research on Global
tica is magnified by the prospect of inter- Climate Change at Florida Tech, this work
continental homogenization of community was supported by the National Science
structure: the loss of diversity among bio- Foundation through Grants ANT-1141877
geographic provinces (Aronson et al. 2007; and OCE-1535007.
Aronson, Frederich, et al. 2015; Aronson,
Smith, et al. 2015). Warming seas and in-
creasing ship traffic are placing the Antarc- R E FE R E NCE S
tic fauna at risk from climatically mediated
biological invasions and range expansions Aronson, R. B. 1990. “Onshore-offshore patterns of hu-
of marine taxa that include durophagous man fishing activity.” Palaios 5: 88–93.
Aronson, R. B. 2017. “Metaphor, inference, and predic-
predators. Crabs will be the first shell- tion in paleoecology: Climate change and the Antarc-
crushing predators to arrive in benthic- tic bottom fauna.” In Conservation Paleobiology: Science and
shelf communities, but durophagous fish Practice, ed. G. P. Dietl and K. W. Flessa, 183–200. Uni-
will follow (Aronson et al. 2007). Remod- versity of Chicago Press.
ernization of the shelf benthos would de- Aronson, R. B., M. Frederich, R. Price, and S. Thatje.
2015. “Prospects for the return of shell-crushing crabs
stroy the endemic, retrograde character of to Antarctica.” Journal of Biogeography 42: 1–7.
Antarctic marine life. Aronson, R. B., K. E. Smith, S. C. Vos, J. B. McClintock,
Apart from limiting direct environmen- M. O. Amsler, P.-O. Moksnes, D. S. Ellis, et al. 2015.
tal damage from pollution and fishing, two “No barrier to emergence of bathyal king crabs on
clear prescriptions are to control carbon the Antarctic shelf.” Proceedings of the National Academy of
Sciences 112: 12997–13002.
emissions globally and regulate ship traf- Aronson, R. B., S. Thatje, A. Clarke, L. S. Peck, D. B.
fic in the Southern Ocean (Aronson et al. Blake, C. D. Wilga, and B. A. Siebel. 2007. “Climate
2011). The Ross Sea Marine Protected Area, change and invasibility of the Antarctic benthos.” An-
which was created in 2016, will protect nual Review of Ecology, Evolution, and Systematics 38: 129–154.
Antarctic fish stocks and other resources. Aronson, R. B., S. Thatje, J. B. McClintock, and K. A.
Hughes. 2011. “Anthropogenic impacts on marine
Like many other marine protected areas, ecosystems in Antarctica: The Yearbook in Ecology
however, this MPA—the world’s largest at and Conservation Biology.” Annals of the New York Academy
1.57 million km2—will rapidly be rendered of Sciences 1223: 82–107.
obsolete if climate change is not halted and Chown, S. L., A. Clarke, C. I. Fraser, S. C. Cary, K. L.
reversed. As with the broader environ- Moon, and M. A. McGeoch. 2015. “The changing
form of Antarctic biodiversity.” Nature 522: 431–438.
mental crisis, the question is whether the Condon, R. H., W. M. Graham, C. M. Duarte, K. A. Pitt,
international community has the political K. H. Lucas, S. H. D. Haddock, K. R. Sutherland, et al.
will to rise to the occasion. The conceptual 2012. “Questioning the rise of gelatinous zooplankton
framework provided by the metaphor of the in the world’s oceans.” BioScience 62: 160–169.
time stamp, including the allied concept of Crame, J. A. 2014. “Evolutionary setting.” In Biogeographic
Atlas of the Southern Ocean, ed. C. De Broye and P. Koubbi,
no-analogue communities, hopefully will 32–35. Scientific Committee on Antarctic Research,
help align practitioners and academics as Cambridge.
they work together on desperate problems Donlan, C. J., J. Berger, C. E. Bock, J. H. Bock, D. A. Bur-
of monumental proportion. ney, J. A. Estes, D. Forman, et al. 2006. “Pleistocene
C L I M AT E C H A N G E , C O N S E RVAT I O N , A N D T H E M E TA P H O R O F D E E P T I M E 123
rewilding: An optimistic agenda for twenty-first cen- Kennett and D. A. Warnke, 131–139. American Geo-
tury conservation.” American Naturalist 168: 660–681. physical Union.
Doubleday, Z. A., T. A. A. Prowse, A. Arkhipkin, G. J. Metaxas, A., and B. Griffin. 2004. “Dense beds of the ophi-
Pierce, J. Semmens, M. Steer, S. C. Leporati, S. Lou- uroid Ophiacantha abyssicola on the continental slope off
renço, A. Quetglas, W. Sauer, and B. M. Gillanders. Nova Scotia, Canada.” Deep-Sea Research I 51: 1307–1317.
2016. “Global proliferation of cephalopods.” Current Molinos, J.-G., B. S. Halpern, D. S. Schoeman, C. J. Brown,
Biology 26: R406–R407. W. Kiessling, P. J. Moore, J. M. Pandolfi, E. S. Poloczan-
Douglas, P. M. J., H. P. Affek, L. C. Ivany, A. J. P. Houben, ska, A. J. Richardson, and M. T. Burrows. 2015. “Cli-
W. P. Sijp, A. Sluijs, S. Schouten, and M. Pagani. 2014. mate velocity and the future global redistribution of
“Pronounced zonal heterogeneity in Eocene southern marine biodiversity.” Nature Climate Change 6: 83–88.
high-latitude sea surface temperatures.” Proceedings of the Monarrez, P. M., M. Aberhan, and S. M. Holland. 2017.
National Academy of Sciences 111: 6582–6587. “Regional and environmental variation in escalatory
Estes, J. A., J. Terborgh, J. S. Bashares, M. E. Power, ecological trends: A western Tethys hotspot for escala-
J. Berger, W. J. Bond, S. R. Carpenter, et al. 2011. tion?” Paleobiology 43: 569–586.
“Trophic downgrading of Planet Earth.” Science 333: Near, T. J., A. Dornburg, K. L. Kuhn, J. T. Eastman, J. N.
301–306. Pennington, T. Patarnello, L. Zane, D. A. Fernández,
Finnegan, S., S. C. Anderson, P. G. Harnik, C. Simpson, and C. D. Jones. 2012. “Ancient climate change, an-
D. P. Tittensor, J. E. Byrnes, Z. V. Finkel, et al. 2015. tifreeze, and the evolutionary diversification of Ant-
“Paleontological baselines for evaluating extinction arctic fishes.” Proceedings of the National Academy of Sciences
risk in the modern oceans.” Science 348: 567–570. 109: 3434–3439.
Griffiths, H. J., A. J. S. Meijers, and T. J. Bracegirdle. Norris, R. D., S. K. Turner, P. M. Hull, and A. Ridgwell.
2017. “More losers than winners in a century of fu- 2013. “Marine ecosystem responses to Cenozoic cli-
ture Southern Ocean seafloor warming.” Nature Climate mate change.” Science 341: 492–498.
Change 7: 749–754. Packard, A. 1972. “Cephalopods and fish: The limits of
Griffiths, H. J., R. J. Whittle, S. J. Roberts, M. Belchier, convergence.” Biological Reviews of the Cambridge Philosophical
and K. Linse. 2013. “Antarctic crabs: Invasion or en- Society 47: 241–307.
durance?” PLOS One 8: e66981. Pandolfi, J. M., J. B. C. Jackson, N. Baron, R. H. Bradbury,
Guimarães, P. R., M. Galetti, and P. Jordano. 2008. “Seed H. M. Guzman, T. P. Hughes, C. V. Kappel, F. Micheli,
dispersal anachronisms: Rethinking the fruits the J. C. Ogden, H. P. Possingham, and E. Sala. 2005. “Are
gomphotheres ate.” PLOS One 3: e1745. U.S. coral reefs on the slippery slope to slime?” Science
Gutt, J., N. Bertler, T. J. Bracegirdle, A. Buschmann, J. 307: 1725–1726.
Comiso, G. Hosie, E. Isla, et al. 2015. “The South- Pauly, D., V. Christiansen, J. Dalsgaard, R. Froese, and F.
ern Ocean ecosystem under multiple climate change C. Torres Jr. 1998. “Fishing down marine food webs.”
stresses—An integrated circumpolar assessment.” Science 279: 860–863.
Global Change Biology 21: 1434–1453. Pörtner, H. O. 2006. “Climate-dependent evolution of
Hall, S., and S. Thatje. 2011. “Temperature-driven bioge- Antarctic ectotherms: An integrative analysis.” Deep-
ography of the deep-sea family Lithodidae (Crustacea: Sea Research II 53: 1071–1104.
Decapoda: Anomura) in the Southern Ocean.” Polar Precht, W. F., and R. B. Aronson. 2006. “Death and res-
Biology 34: 363–370. urrection of Caribbean coral reefs: A paleoecological
Jablonski, D., J. J. Sepkoski Jr., D. J. Bottjer, and P. M. perspective.” In Coral Reef Conservation, ed. I. Côté and J.
Sheehan. 1983. “Onshore-offshore patterns in the Reynolds, 40–77. Cambridge University Press.
evolution of Phanerozoic shelf communities.” Science Roy, K., and J. M. Pandolfi. 2005. “Responses of species
222: 1123–1125. and ecosystems to past climate change.” In Climate
Jackson, J. B. C. 2008. “Ecological extinction and evolu- Change and Biodiversity, ed. T. E. Lovejoy and L. Hannah,
tion in the brave new ocean.” Proceedings of the National 160–175. Yale University Press.
Academy of Sciences 105: 11458–11465. Schmidtko, S., K. J. Heywood, A. F. Thompson, and S.
Janzen, D. H., and P. S. Martin. 1982. “Neotropical Aoki. 2014. “Multidecadal warming of Antarctic wa-
anachronisms: The fruits the gomphotheres ate.” Sci- ters.” Science 346: 1227–1230.
ence 215: 19–27. Sepkoski, J. J., Jr. 1991. “Diversity in the Phanerozoic
Knoll, A. H., and M. J. Follows. 2016. “A bottom-up per- oceans: A partisan view.” In The Unity of Evolutionary Biol-
spective on ecosystem change in Mesozoic oceans.” ogy: Proceedings of the Fourth International Congress of System-
Proceedings of the Royal Society, Series B 283: 20161755. atic and Evolutionary Biology (Volume 1), ed. E. C. Dudley,
Levin, P. S., J. Ellis, R. Petrik, and M. E. Hay. 2002. “Indi- 210–236. Dioscorides Press.
rect effects of feral horses on estuarine communities.” Sheehan, P. M. 2001. “History of marine diversity.” Geo-
Conservation Biology 16: 1364–1371. logical Journal 36: 231–249.
Long, D. J. 1992. “Paleoecology of Eocene Antarctic Sheehan, P. M., and M. T. Harris. 2004. “Microbialite re-
sharks.” In The Antarctic Paleoenvironment: A Perspective on surgence after the Late Ordovician extinction.” Nature
Global Change, Part 1. Antarctic Research Series 56, ed. J. P. 430: 75–78.
124 W H AT D OE S T H E PA ST T E L L US?
Smith, C. R., L. J. Grange, D. L. Honig, L. Naudts, B. Vermeij, G. J. 1977. “The Mesozoic marine revolution:
Huber, L. Guidi, and E. Domack. 2012. “A large Evidence from snails, predators and grazers.” Paleobiol-
population of king crabs in Palmer Deep on the west ogy 3: 245–258.
Antarctic Peninsula shelf and potential invasive im- Vermeij, G. J. 1978. Biogeography and Adaptation: Patterns of Ma-
pacts.” Proceedings of the Royal Society, Series B 279: 1017– rine Life. Harvard University Press.
1026. Walther, G.-R., E. Post, P. Convey, A. Menzel, C. Parme-
Smith, K. E., R. B. Aronson, B. V. Steffel, M. O. Amsler, san, T. J. C. Beebee, J.-M. Fromentin, O. Hoegh-Guld-
S. Thatje, H. Singh, J. Anderson, et al. 2017. “Climate berg, and F. Bairlein. 2002. “Ecological responses to
change and the threat of novel marine predators in recent climate change.” Nature 416: 389–395.
Antarctica.” Ecosphere 8: e02017. Watson, S.-A., S. A. Morley, and L. S. Peck. 2017. “Latitu-
Stilwell, J. D., and W. J. Zinsmeister. 1992. “Molluscan dinal trends in shell production cost from the tropics
systematics and biostratigraphy: Lower Tertiary La to the poles.” Science Advances 3: e1701362.
Meseta Formation, Seymour Island, Antarctic Penin- Whittle, R. J., A. W. Hunter, D. J. Cantrill, and K. J.
sula.” Antarctic Research Series 55. American Geophysical McNamara. 2018. “Globally discordant Isocrinida
Union. (Crinoidea) migration confirms asynchronous Ma-
Trivelpiece, W. Z., J. T. Hinke, A. K. Miller, C. S. Reiss, S. rine Mesozoic Revolution.” Communications Biology 1: 46.
G. Trivelpiece, and G. M. Watters. 2011. “Variability in https://doi.org/10.1038/s42003-018-0048-0.
krill biomass links harvesting and climate warming to Williams, J. W., and S. T. Jackson. 2007. “Novel climates,
penguin population changes in Antarctica.” Proceedings no-analog communities, and ecological surprises.”
of the National Academy of Sciences 108: 7625–7628. Frontiers in Ecology and the Environment 5: 475–482.
CASE STUDY 4
Based on “Impact of sea level rise on the 10 insular biodiversity hotspots,” Bellard et al. (2014). All the
sources cited here, if not listed at the end of the case study, can be found in this key reference.
Sea-level rise is one of the main components of climate change. In agreement with climate
models, projections showed that sea level will likely continue to rise in more than 95 percent
of the ocean area during the coming decades. The current estimates give an increase of 0.26
to 0.98 m for the period 2081–2100 (Stocker et al. 2013). Small islands and other low-lying
coastal regions are expected to be the most affected by sea-level rise. There are about 180,000
islands worldwide that are often characterized by a high proportion of endemic species, par-
ticularly vulnerable to threats. Several studies have investigated how sea-level rise may affect
biodiversity. For instance, sea-level rise may affect marine ecosystems by inducing changes
of parameters such as available light, salinity, and temperature. The long-term persistence of
freshwater-dependent ecosystems can also be affected by the increase of sea level (Traill et
al. 2011). In addition, species in terrestrial ecosystems that might be permanently inundated
could be locally extinct as a result of sea-level rise. Besides, the economic cost of sea-level rise
is also important. For example, the average global flood losses are estimated to be approxi-
mately 5 percent of global gross domestic product in 2005 across all cities (Hanson et al.
2010). As a result, sea-level change will cause habitat change and loss of coastal areas, both of
which will have important consequences for society and biodiversity. Because sea-level rise is
predicted to continue to increase in the next decades, a number of studies have investigated
its future effects on insular biodiversity. For example, Bellard et al. (2014) investigated the
potential effects of sea-level rise on the 10 insular biodiversity hotspots following an increase
in sea level by 1 m to 6 m. These hotspots contain more than 50,000 endemic plant species
and about 670 endemic mammals. Because insular hotspots are densely populated islands
(31.8 percent of all humanity) the implications of sea-level rise could be considerable for
human societies. In this study, they estimated the number of islands partially and entirely
submerged (Figure CS4.1) following sea-level rise. They showed that 6 percent–19 percent of
the 4,447 islands studied could be submerged under an increase of sea-level rise of 1 m–6 m.
125
126 W H AT D OE S T H E PA ST T E L L US?
[5;23]
[63;356]
[48;113]
[15;84]
[39;67]
[23;33]
[61;122]
[1;5]
[4;13]
[8;10]
Proportion of islands
submerged by a sea
level rise of:
1 Meter Meters Area stable
[Number of islands entirely submerged by a sea level rise of (1m:6m)]
Figure CS4.1. Percentage of islands included in the 10 in- annual losses of 0.3 percent–9.3 percent to
sular biodiversity hotspots predicted to be submerged global gross domestic product.
from 1 m to 6 m by sea-level rise scenarios.
It should be noted that the majority of
the studies that investigated the effect of
be completely inundated under scenarios of sea-level rise at a regional or global scale
sea-level rise by 1 m to 6 m), highlighting used simplified scenarios. For instance, they
the extinction risk for 54 mammals of the simulated a uniform increase in sea-level
Indo-Malaysian islands. Using the hotspot of rise worldwide. In addition, most of them
the forest of East Australia as a case study, did not consider the potential effects of
Bellard et al. (2016) also illustrate that nearly lateral erosion, intrusion of saline waters,
860 km2 could be entirely submerged with decadal and centennial tides, and eventu-
an increase of sea level by 1 m, with water ally secondary impacts like displacement of
intrusion reaching up to 35 km inland. This human populations and agricultural lands.
inundated area mainly includes natural areas Besides, they do not include the dynamic
such as national parks that harbor high spe- response of islands. Yet Webb and Kench
cies richness (about 150 species per pixel) (2010) highlighted the dynamic nature of
and low protected areas. They also demon- sea-level rise response of islets in four atolls
strated that inundated areas might lead to the of the Central Pacific. Over periods of 19
displacement of 20,600 people, especially to 61 years, 43 percent of islands showed
around Graton, Port Macquarie, and Taree an increase (and stability), whereas only 14
cities in Australia. Consequently, the future percent showed a decrease.
cost of sea-level rise for society is expected Finally, there are a variety of models,
to be important. Hallegatte et al. (2013) cal- from a simple drowning model to a pro-
culated that average global flood losses could cess-based model, that have been recently
reach US$52 billion by 2050 in the 136 developed with their own strengths and
largest coastal cities. Without adaptation to weaknesses to investigate the effect of sea-
mitigate the effect of rising sea levels, Hin- level rise. These techniques are suited for
kel et al. (2014) estimate that 0.2 percent–4.6 different management objectives regarding
percent of global population is expected to sea-level rise (Runting, Wilson, and Rhodes
be flooded annually in 2100, with expected 2013). Therefore, there is much room to
THE EFFECTS OF SE A-L EV EL R ISE ON H A BITATS A ND SPECIE S 127
improve the current techniques and knowl- Forests of East Australia.” Environmental Conservation 43:
edge of the likely effects of sea-level rise. 79–89. https://doi.org/10.1017/S037689291500020X.
Hallegatte, S., C. Green, R. J. Nicholls, and J. Corfee-
To conclude, although the magnitude Morlot. 2013. “Future flood losses in major coastal
of sea-level rise and its potential effects are cities.” Nature Climate Change 3: 802–806. https://doi.
difficult to predict with certainty, it is es- org/10.1038/nclimate1979.
tablished that sea-level rise will occur and Hinkel, J., D. Lincke, A. T. Vafeidis, M. Perrette, R. J.
threaten a large and unique part of biodi- Nicholls, R. S. J. Tol, et al. 2014. “Coastal flood dam-
age and adaptation costs under 21st century sea-level
versity. Therefore, vulnerability assessments rise.” Proceedings of the National Academy of Sciences 111:
and current conservation programs should 3293–3297. https://doi.org/10.1073/pnas.1222469111.
include this threat among others criteria of Runting, R. K., K. A. Wilson, and J. R. Rhodes. 2013.
prioritization schemes. “Does more mean less? The value of information for
conservation planning under sea level rise.” Global
Change Biology 19: 352–363. https://doi.org/10.1111/
gcb.12064.
R E FE R E NCE S Stocker, T. F., D. Qin, G.-K. Plattner, L. V. Alexander, S. K.
Allen, N. L. Bindoff, F.-M. Bréon, et al. 2013. “Techni-
Bellard, C., C. Leclerc, and F. Courchamp. 2014. “Im- cal summary.” In Climate Change 2013: The Physical Science
pact of sea level rise on the 10 insular biodiversity Basis. Contribution of Working Group I to the Fifth Assessment
hotspots.” Global Ecology and Biogeography 23: 203–212. Report of the Intergovernmental Panel on Climate Change, ed. T.
https://doi.org/10.1111/geb.12093. F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S. K. Al-
Bellard, C., C. Leclerc, B. D. Hoffmann, and F. Cour- len, J. Boschung, A. Nauels, Y. Xia, V. Bex, and P. M.
champ. 2016. “Vulnerability to climate change and Midgley, 33–115. Cambridge University Press. https://
sea-level rise of the 35th biodiversity hotspot, the doi.org/10.1017/CBO9781107415324.005.
CHAPTER TEN I N T RODUCTION
128
PA ST A BRU P T C H A NGE S I N C L I M AT E A N D T E R R E ST R I A L E CO S Y ST E M S 129
turies to millennia (Williams et al. 2004), climates were characterized by en-
as well as ecological responses to large but hanced hydrological variability, with
infrequent climatic extreme events. Thus, widespread ecological and societal
the resilience of ecological systems to rapid impacts.
climate change and rates of ecological re-
sponse can be studied. Species have four op- “Abrupt” has several definitions and
tions in a changing climate—move, adapt, application across many time scales (Wil-
persist, or die—and all four have happened liams et al. 2011, National Research Council
in the past. 2013). We focus on climatic and ecological
No geological event is a perfect analog changes large and fast enough to challenge
for the climatic and ecological changes climate adaptation (i.e., over years to de-
expected for this century, but many offer cades) or where an ecological tipping point
instructive model systems (Williams et al. is indicated by an ecological response much
2013). Here we review several such events, faster than the climatic forcing.
each illuminating a different aspect of bio-
logical dynamics during periods of rapid
change. We focus on terrestrial vegetation, PA L E O C E N E -E O C E N E T H E R M A L
because plants are well documented in the M A X IM U M (PE T M)
fossil record, are fundamental to terrestrial
ecosystem function, and are at risk of both During the Paleocene-Eocene Thermal
abrupt threshold-type responses and adap- Maximum, 56.1 million years ago (McIn-
tive lags (Aitken et al. 2008; Williams et al. erney and Wing 2011), vast amounts of or-
2011; Svenning and Sandel 2013; Scheffer ganic carbon from geobiological reservoirs
et al. 2012): were released rapidly into the atmosphere
and ocean. This release is unequivocally
1. The Paleocene-Eocene Thermal signaled by a negative excursion in carbon
Maximum (Chapter 8), 56.1 million isotopes (~4.6‰) in multiple marine and
years ago, is a model system to study terrestrial records (McInerney and Wing
the transient and equilibrium re- 2011): a carbon pulse (ca. 5,000 to 10,000
sponses of terrestrial ecosystems to gigatons carbon [GtC]) equivalent to or
a massive release of organic carbon greater than all contemporary reserves of
into the atmosphere and ocean and fossil fuels. Ocean acidification and carbon-
consequent changes in temperature ate undersaturation are indicated by de-
and ocean biogeochemistry. creased preservation of calcium carbonate
in marine sediments (Zachos et al. 2005).
2. The last deglaciation, 19,000 to Carbon release was on the order of thou-
8,000 years ago, is a model sys- sands of years and less than 20,000 years
tem to study the ability of species (McInerney and Wing 2011). The pertur-
to track temperature changes that bation and reequilibration of the carbon
were, at least regionally, comparable cycle and climate system lasted 150,000 to
to those expected for this century. 200,000 years (McInerney and Wing 2011).
The favored explanation for the PETM
3. The middle to late Holocene, 8,000 until recently was degasing of methane
years ago to ~1800 AD, is a model from shallow-marine clathrates. However,
system for the effects of hydrologi- the amount of carbon stored in Paleocene
cal variability and extreme events clathrates (~1000 GtC, Buffett and Archer
in a warmer world. Sometimes 2004) is insufficient to explain the mag-
misperceived as stable, Holocene nitude of carbon isotopic excursions and
Table 10.1. Abrupt climatic and ecological events in the geological record
Event What happened? When? Rapidity of event Ecological effects Rapidity of response References
Paleocene-Eocene Thermal Release of 5–10 Gt organic PETM: 56.1 Ma ETM2: 53.7 <20,000 years Intercontinental species im- Rapid response, lags McInerney and Wing
Maximum (PETM), carbon; ocean acidifica- Ma ETM3: 52.4 Ma migration; community unclear 2011; DeConto et
ETM2, ETM3 tion; 4°C–5°C warming turnover; dwarfing al. 2012
of tropics
Glacial Terminations Global temperature rise of 9 terminations over the 5,000–10,000 Range shifts over 102 to Rapid response, lags Lüthi et al. 2008
3°C–5°C, GHG rise, ice past 800,000 years years 103 km, community unclear
sheets melt turnover
Dansgaard-Oeschger Events 5°C warming in Greenland 20 events between 10,000 <30 years Europe: afforestation from Rapid—lags likely Bond et al. 1993; Al-
and 80,000 years ago grasslands to wooded but not yet len et al. 1999
steppe measured
Bølling Warming 9°C–14°C warming in 14,700 years ago 1–3 years Rapid afforestation of tun- Initial response: Ammann et al. 2013
Greenland dra, expansion of species 8–16 years; Pinus
from glacial refugia expansion delayed
800 years
Younger Dryas Initiation 5°C–9°C cooling in Green- 12,700 years ago 1 year? Rapid plant community <40 years, i.e. Brauer et al. 2008;
land, 2°C cooling in turnover, declines of within sampling Buizert et al.
Western Europe thermophilous species resolution 2014; Rach et al.
2014
Younger Dryas Termination, 8°C–12°C warming in 11,700 years ago <60 years Similar to Bølling Warming Synchronous with Birks 2015
Start of Holocene Greenland, 3°C in warming; tree
Northern Europe Betula arrival de-
layed 650 years
8.2 ka Event 5°C–7°C cooling in 8,400 years ago 5 years Similar to YD Initiation 0 to 40 years Tinner and Lotter
Greenland 2001; Alley and
Ágústsdóttir 2005
Holocene Aridification Waning of Northern Hemi- 11,000 to 5,000 years ago, Thousands of Collapse of North African Locally abrupt Kuper and Kröpelin
sphere monsoons and timing varies regionally years grasslands, retreat of for- (101–102 years) 2008
declines in subtropical est ecotones, activation shifts embedded
precipitation of dunes, C3/C4 shifts, within slower
altered fire regime aridification
Holocene Megadroughts Well documented by den- 1–10 years Slowed tree growth rates, Tree population Cook et al. 2010a;
droclimatic records for mortality of mesic tree collapses as fast as Cook et al. 2010c;
last 2000 years; height- species, abandonment of <10 years Williams et al.
ened variability 5,500 to early agricultural sites 2011; Shuman
4,000 years ago 2012
PA ST A BRU P T C H A NGE S I N C L I M AT E A N D T E R R E ST R I A L E CO S Y ST E M S 131
carbonate undersaturation (Higgins and by 4°C to 5°C (Higgins and Schrag 2006).
Schrag 2006). Consequently, other mecha- Biological responses varied strongly among
nisms for the PETM carbon release have taxonomic groups. The PETM event was
been proposed (Higgins and Schrag 2006; most severe for benthic foraminifera, 30
McInerney and Wing 2011). percent to 50 percent of which went ex-
Of these, the recent Antarctic perma- tinct. It also profoundly affected the ecol-
frost hypothesis (DeConto et al. 2012) el- ogy and evolution of terrestrial species
egantly unifies several lines of evidence. (McInerney and Wing 2011). Mammal
In this hypothesis, organic carbon was re- body sizes rapidly decreased, perhaps as a
leased by thawing of Antarctic reservoirs result of higher temperatures or reduced
of permafrost and peat. Antarctica was forage quality (Gingerich 2006; McInerney
unglaciated during the Paleocene, creating and Wing 2011). The duration of dwarfed
a large polar land area (~12 million km2) body size is limited to the PETM inter-
available for carbon sequestration in peat- val, suggesting highly plastic responses of
land and permafrost. The PETM was fol- mammalian body size to environmental
lowed by thermal maxima and releases of change. Connectivity and intercontinental
organic carbon into the atmosphere-ocean dispersal increased among North America,
system (ETM2 and ETM3) at 53.7 mil- Europe, and Asia, with long-lasting impacts
lion years ago and 52.4 million years ago. on faunal distributions and evolution (Gin-
These carbon releases align with past con- gerich 2006).
figurations of the earth’s slowly wobbling Although there was no increase in global
orbit that favor higher incoming sunlight terrestrial extinctions during the PETM,
at the poles (DeConto et al. 2012), suggest- turnover in plant communities was rapid
ing that permafrost thawing was triggered (<10,000 years) and implies the local extir-
by orbitally driven variations in incoming pation of some taxa and range expansions
sunlight over the South Pole. The Antarctic of others across distances up to 1,000 km
peatland hypothesis can explain both the (Wing et al. 2005; McInerney and Wing
magnitude and rate of carbon release and 2011). Interestingly, plants experienced few
the multiple thermal maxima. It also ex- long-lasting impacts: pre-PETM and post-
plains why thermal events are absent after PETM floral assemblages are closely similar.
Antarctica glaciated approximately 35 mil- Herbivory-damage analyses of PETM leaves
lion years ago and why the amount of car- suggest intensified and more specialized
bon released declined between subsequent types of insect feeding under conditions
events: each thermal maximum depleted of higher temperatures and atmospheric
the organic carbon reserves in Antarc- CO2 (Currano et al. 2008). Several orders of
tic permafrost and peatlands, leaving less modern mammals, including primates, first
available for release in subsequent thermal appear after the PETM (Gingerich 2006).
events (DeConto et al. 2012). The Antarctic The magnitude of carbon cycle disrup-
peatland hypothesis underscores the sensi- tion during the PETM is similar to the high-
tivity of terrestrial carbon reserves to thaw- end emission scenarios for the coming cen-
ing and the potential of high-latitude soil turies, but the PETM onset is slower than
carbon and vegetation feedbacks to acceler- Anthropocene onset by one to two orders
ate warming. of magnitude (i.e., a PETM carbon pulse and
The climatic, biogeochemical, and bio- temperature rise over 103 to 104 years vs.
logical effects of the PETM were profound. an Anthropocene rise over 102 to 103 years).
Surface air temperatures rose by 4°C to 5°C Hence, the ecological and evolutionary ef-
in the tropics and 6°C to 8°C in the high fects of the PETM should be conservative
latitudes, while the deep ocean warmed relative to those of the coming century.
132 W H AT D OE S T H E PA ST T E L L US?
Figure 10.1. Fast and slow vegetation responses to the tundra is summarized by Pinus, Betula, Artemisia, Cypera-
abrupt climate changes during the last deglaciation, ceae, and Poaceae pollen abundances. (C) Lake Kråkenes
based on four high-resolution multiproxy records from (Norway), where abrupt warming at start of Holocene
Western Europe. (A) Lake Gerzensee (Switzerland), (highlighted by gray band) is inferred from chirono-
where rapid warming at Bølling onset (14,700 years mid assemblages (Birks 2015). This warming triggered
ago, dashed line) is captured by δ18O. Rapid woodland immediate losses of some Arctic and alpine species and
expansion and delayed Pinus migration are summarized multiple stages of vegetation turnover, culminating 700
by relative pollen abundances for Artemisia, Cyperaceae, years later with the immigration of tree birch (Betula pu-
Juniperus, Betula, and Pinus and stomatal counts for Pinus bescens). (D) Soppensee Lake (Switzerland), where forest
and Juniperus (Ammann et al. 2013). (B) Meerfelder Maar community composition responded within 0–40 years
(Germany), where abrupt cooling and increased wind to the abrupt cooling at 8.2 ka (dashed line), in turn trig-
strength at the start of the Younger Dryas are indicated gered by Laurentide Ice Sheet collapse (Tinner and Lotter
by hydrogen isotopes from terrestrially sourced or- 2001). Gains or losses were transitory for some tree taxa
ganic compounds, varve thickness, and varve mineral- (Pinus, Betula, Tilia) and long lasting for others (e.g., Corylus
ogy (Rach et al. 2014). Rapid conversion from forest to avellana, Fagus sylvatica).
134 W H AT D OE S T H E PA ST T E L L US?
pollen is ubiquitous and a good proxy for away or migration occurred in a series of
plant abundance, although weighted to- rapid jumps) or slower (if cryptic refugia
ward wind-dispersed pollen types. Plant were <500 km or if migration began prior
macrofossils are less common but definitive to Bølling onset).
indicators of the presence of a source plant. Meerfelder Maar in western Germany
Here we present four examples (Figure (Figure 10.1B) captures the effect of the
10.1), from different sites and time periods rapid return to cold and dry conditions in
in Western Europe, where the climatic sig- Europe at the onset of the Younger Dryas
nal of abrupt climate change is particularly (YD). At Meerfelder, the YD began with a
strong. These records demonstrate a mix- gradual cooling at 12,850 years ago (based
ture of fast (near instantaneous) and slow on hydrogen isotopes in plant biomarkers),
(century scale) response rates by terrestrial followed by an abrupt shift (1 to 3 years)
plant communities. Forest composition and 170 years later to drier and windier cli-
many tree species abundances are sensi- mates (Rach et al. 2014). The initial cool-
tive to climate change, with rapid climatic ing correlates closely to Greenland isotopic
tracking by some components. Often the lag records but took several centuries to mani-
between climatic forcing and initial vegeta- fest, whereas the second shift was extraor-
tion response is statistically indistinguish- dinarily rapid (perhaps within a single
able from zero. Other taxa have century- year) and likely caused by expansion of sea
scale response times, presumably due to ice over the North Atlantic, rapid atmo-
successional dynamics and migration rates. spheric reorganization, and reduced mois-
At Lake Gerzensee in Switzerland (Fig- ture advection to Western Europe (Brauer
ure 10.1A), initial biotic responses to rapid et al. 2008; Rach et al. 2014). Changes in
warming 14,685 years BP, at the start of the vegetation composition were synchronous
Bølling, occurred within 8 to 16 years (Am- with both events: declines in Pinus and Betula
mann et al. 2013). Terrestrial and aquatic and rises in Artemisia, Cyperaceae, and Poa-
communities responded simultaneously, ceae closely parallel the initial cooling and
suggesting that intertaxonomic differences quickly accelerated with the second shift.
in generation time or climatic sensitivity The pollen sampling resolution at Meer-
had no effect. However, the complex se- felder is 8 to 40 years, indicating short
quence of vegetation changes triggered by (at most, decadal-scale) time lags between
the Bølling-Allerød warming lasted nearly regional cooling and aridification, mortal-
1,000 years. Juniperus and Betula were fast re- ity of Pinus and Betula, and replacement of
sponders, with near-instantaneous expan- woodlands with tundra. Meerfelder offers
sion into shrub tundra. These woodlands a cautionary note against assessing past
persisted for 180 years, until replacement rates of ecological response based on high-
by Betula forests. Pinus arrived at Gerzensee resolution but geographically dispersed re-
by about 13,850 years ago, based on Pinus cords; a comparison of Meerfelder pollen to
stomata and rapidly rising Pinus pollen Greenland isotopic records would suggest
abundances. The closest known glacial re- an approximate 150-year vegetation lag af-
fugia for Pinus are in the southeastern Alps, ter the Younger Dryas onset, when in fact
500 km from Gerzensee (Ammann et al. any lag is at most a few decades.
2013) If we assume that Pinus was initially Kråkenes Lake, in coastal Norway (Fig-
500 km distant and that climates became ure 10.1C), has a rich plant macrofossil
locally suitable for Pinus at the start of the record, which enables precise estimates of
Bølling warming, then we can calculate an local turnover in plant community compo-
average migration rate for Pinus of 0.6 km/ sition and the relative timing of immigra-
year. Actual migration rates could have been tion and extirpation events. July tempera-
faster (if source populations were >500 km tures rose from 8°C to 11°C at Kråkenes
PA ST A BRU P T C H A NGE S I N C L I M AT E A N D T E R R E ST R I A L E CO S Y ST E M S 135
at the start of the Holocene (Birks 2015). tirpations of plant populations at the trail-
This rise began 11,630 years ago and lasted ing edge of species ranges and presumably
until 11,550 years BP, for an average of caused species to move among microcli-
3.75°C/century. Vegetation composition at mates within their established range. De-
Kråkenes abruptly changed between 11,600 spite these large shifts, extinctions of plant
and 11,550 years ago, synchronous with species were rare. One extinction, Picea
the temperature rise. Many Arctic and al- critchfieldii (Jackson and Weng 1999), is well
pine plants were locally extirpated (e.g., Koe- documented. This contrasts with the global
nigia islandica) or reduced to trace presences wave of megafaunal extinctions that began
(e.g., Saxifraga rivularis, S. cespitosa). This rapid 40,000 years ago and continued through the
response was followed by a succession of late Holocene (Koch and Barnosky 2006).
vegetation communities over the next 700 However megafaunal populations on re-
years. High abundances of Salix herbacea per- mote islands (Madagascar, New Zealand,
sisted for 150 years (until 11,400 years BP), Fiji, Mediterranean islands) did not suf-
followed by increases in grassland and fen fer widespread extinctions during the last
taxa, then Empetrum heathlands. Betula pu- deglaciation, despite major environmental
bescens (tree birch) apparently arrived at change and sea-level rise (but see Graham
Kråkenes by 10,940 years BP, roughly 650 et al. 2016), suggesting that late-Holocene
years after the initial event. The initial rise human arrival was the critical driver.
to 11°C should have been sufficient for B.
pubescens survival, suggesting that dispersal
limitation was the primary reason for the H Y DROLO GIC A L VA R I A BI L IT Y A N D
600-year lag. During the Younger Dryas, E COLO GIC A L TIPPI NG POI N TS
the nearest known B. pubescens refugia are DU R I NG T H E HOLO C E N E
in Denmark and southern Sweden (Birks
2015), roughly 600 km away, suggesting a Hydrological variability and the frequency
migration rate of ~1 km/year. In northern of extreme events is expected to increase
Norway, expansion of B. pubescens wood- over this century because a warmer atmo-
lands kept pace with temperature velocities sphere can hold more water vapor and be-
(Birks 2015). cause enhanced evaporation rates will de-
At Soppensee Lake, Switzerland, forest liver more water to the atmosphere while
composition responded quickly to 8.2 ka locally drying out land surfaces. Similarly,
cooling, with no time lags for several tree a defining characteristic of the Holocene is
taxa (Figure 10.1D). The varved Soppensee its heightened hydrological variability and
sediments enable annual-scale analyses, the profound consequences of this variabil-
and the pollen record at Soppensee was ity for terrestrial ecosystems (Figure 10.2)
analyzed at decadal resolution (Tinner and and early agricultural societies (deMenocal
Lotter 2001). Corylus avellana abundances col- 2001).
lapsed, while Pinus, Betula, Fagus sylvatica, and There is a persistent myth that the Ho-
Tilia all expanded. Time lags are short, rang- locene was climatically stable and benign,
ing from 0 to 40 years, suggesting that for- because of the muted temperature variabil-
est composition turned over within a single ity over the last 11,000 years. For example,
tree (and human) generation. Most forest global mean temperatures variations were
responses to the 8.2 ka event were tempo- on the order of <1°C to 2°C from the early to
rary, but the event ended the early Holo- late Holocene (Marcott et al. 2013; Marsicek
cene dominance of Corylus avellana and began et al. 2018), which is small relative to the
the rise of Fagus sylvatica. temperature swings discussed above. This
The large changes in abundances and myth of Holocene stability, however, over-
geographic distributions often caused ex- looks profound variations in hydrological,
136 W H AT D OE S T H E PA ST T E L L US?
Figure 10.2. During the Holocene, slow and progressive ecological, and societal systems during the
regional aridification produced locally abrupt collapses Holocene. Hence, the Pleistocene-Holocene
of forest cover and precipitation. (A) Insolation trends at
6.5°N and a north-south transect of hydrogen isotopic
transition should be viewed instead as a
records from North Africa, showing locally abrupt and fundamental shift in the type and drivers of
regionally time-transgressive collapses in monsoonal climate variability, from a cold glacial world
rainfall, highlighted by dashed line (Shanahan et al. characterized by large temperature swings
2015). (B) Insolation trends at 47°N and an east-west and positive feedbacks strongly governed by
transect of pollen records from the eastern Great Plains,
showing locally abrupt and regionally time-transgressive
ice-sheet dynamics and cryosphere-ocean-
collapses in forest cover (Williams et al. 2009). (C) Map atmosphere feedbacks to a warm intergla-
showing early drying of northern Sahara and later dry- cial world characterized by a wetter and
ing in south (Shanahan et al. 2015). (D) Map showing more energetic atmosphere and governed
earlier drying of sites in interior Great Plains and later by vegetation-atmosphere and ocean-atmo-
drying of sites nearer the prairie-forest ecotone, based
on a compilation of eolian, carbon isotopic, pollen, and
sphere feedbacks.
paleohydrological records (Williams et al. 2010). High hydrological variability during the
Holocene triggered major ecological and
PA ST A BRU P T C H A NGE S I N C L I M AT E A N D T E R R E ST R I A L E CO S Y ST E M S 137
societal effects at regional to subcontinen- ability (Liu et al. 2007), positive surface-
tal scales, often abrupt, particularly in sub- atmosphere feedbacks (Claussen 2009), or
tropical, semiarid, and temperate regions. simply internal dynamics with no climatic
Grasslands in North Africa collapsed during trigger (Williams et al. 2011; Booth et al.
the middle Holocene and were replaced by 2012; Seddon et al. 2015). Holocene hydro-
the Sahara Desert (Figure 10.2), causing re- logical variability and ecological response
treat of pastoral societies (Kuper and Kröpe- are spatially and temporally complex; rarely
lin 2008). In the Great Plains, the Nebraska are events globally or even regionally syn-
Sand Hills repeatedly flipped between active chronous. Rather, hydrological and veg-
dunes and stabilized grasslands (Miao et al. etation variability occurred as a temporal
2007). Forests in the eastern Great Plains mosaic (Williams et al. 2011), with epi-
converted to prairie in the early Holocene, sodes of heightened hydrological variability
reverting to forest in the middle to late Ho- or reduced variability but few individual
locene (Williams et al. 2010). Mesophilic events that can be confidently correlated
trees in the temperate deciduous forests of across sites. For example, the mid-Holocene
the eastern United States (e.g., Tsuga canaden- collapse of Tsuga canadensis in eastern North
sis, Fagus grandifolia) experienced repeated America was once thought to have been a
population crashes, likely tied to decadal- unique and synchronous event, caused by
to centennial-scale hydrological variability a single pest outbreak, but newly resolved
(Foster et al. 2006; Booth et al. 2012; Shu- records show multiple collapses, varying
man 2012). The Holocene expansion of Pinus in timing across sites and regions, and the
edulis (piñon pine) and Juniperus osteosperma in collapses have been linked to hydrological
the western United States was controlled by variability (Shuman 2012). Hydrological
a hydrological ratchet (Jackson et al. 2009), variability appears to have been particularly
with recruitment and range expansion dur- high between 5,500 and 4,000 years ago,
ing wet periods and persistence during dry for unclear reasons, with global effects on
periods. The strength of El Niño–Southern societies and ecosystems (deMenocal 2001;
Oscillation has strongly varied during the Shuman 2012).
Holocene (Cobb et al. 2013), with profound Case studies from North Africa and the
effects globally on precipitation patterns, US Great Plains illustrate the abrupt re-
flood risk, rates of tree growth, and fire sponses of terrestrial vegetation to Holo-
regime. Severe megadroughts, lasting de- cene hydrological variability. Progressive
cades, were common in the western United aridification caused large-scale shifts in bi-
States (Cook et al. 2010b) and eastern and ome extent, decreases in tree density, and
southern Asia (Cook et al. 2010a) over the replacement of grasslands with savanna in
past several millennia, the latter tied to semiarid regions (Figure 10.2). In North
shifts in monsoonal precipitation (Zhang Africa, moisture availability peaked dur-
et al. 2008). Early agricultural societies in ing the early Holocene during the African
India, the Fertile Crescent, Egypt, Central Humid Period, caused by orbitally driven
America, and western and central United shifts in summer solar radiation and en-
States were destabilized by drought and hanced monsoonal strength in the North-
floods (deMenocal 2001; Munoz et al. 2015). ern Hemisphere subtropics (Figure 10.2A).
A major research question is to disentan- North African aridification proceeded
gle when these abrupt vegetation changes as the withdrawal of a monsoonal front,
were caused by similarly abrupt extrinsic with northern locations first losing access
forcings (e.g., megadroughts) (Williams et to monsoonal precipitation and southern
al. 2011; Seddon et al. 2015), intrinsic tip- locations later (Shanahan et al. 2015). The
ping points within ecological systems that southward withdrawal took thousands of
cause abrupt responses to hydrological vari- years, but, at any given location, the decline
138 W H AT D OE S T H E PA ST T E L L US?
in rainfall often was abrupt (decades to cen- Regardless of exact mechanism, the Ho-
turies) (Kuper and Kröpelin 2008; Shana- locene pattern is clear: semiarid systems
han et al. 2015). At Lake Yoa, tropical trees are highly sensitive to moisture availability
disappeared, followed by grasslands, and and can quickly shift to new states when
then deserts established (Kröpelin et al. moisture availability changes. The Holo-
2008). Collapses of individual taxa or taxo- cene history of abrupt vegetation change in
nomic groups at individual sites occurred North Africa and Great Plains is consistent
quickly—on time scales of decades—but with contemporary observations of a forest-
multiple collapses of multiple taxonomic grassland bistability along precipitation gra-
groups played out over centuries for any dients, with savannas a relatively uncom-
given site (Kröpelin et al. 2008), and over a mon intermediate state. It is also consistent
3,000-year window across a transect of sites with current concerns about globally en-
in North Africa (Shanahan et al. 2015). hanced rates of tree mortality due to warm-
The Great Plains also show gradual re- ing and drought (Allen et al. 2010).
gional aridification combined with locally
abrupt declines in tree densities (Figure
10.2B). The timing differs from North SU M M A RY
Africa, which was wet in the early Holo-
cene, whereas the Great Plains were dry The geological record is replete with ex-
(Figure 10.2), likely because of regional amples with abrupt climatic and ecological
atmospheric subsidence linked to intensi- change (Table 10.1). We can use the past
fied monsoons in the southwestern United to identify tipping points and tipping ele-
States and enhanced evapotranspiration. ments in the climate system and biologi-
Great Plains proxies consistently indicate cal systems, identify governing processes,
aridification in the early Holocene and in- measure rates of ecological processes oper-
creasing moisture availability during the ating at decadal and longer timescales, and
middle to late Holocene. For example, the assess the adaptive capacity of organisms
prairie-forest ecotone shifted eastward dur- when confronted by abrupt environmental
ing the early Holocene (prairie replacing change. Just as some animals and ecosys-
forest) then westward during the middle tems make better model systems for par-
to late Holocene (forest replacing prairie). ticular questions in medical and ecological
At individual sites, rates of tree-grassland research, some geological events offer par-
shifts are asymmetrical, with rapid defor- ticularly useful insights into contemporary
estation (decades to centuries) and gradual questions in global change ecology.
reforestation (centuries to millennia). On Of the three model systems reviewed here,
the basis of close correspondences among the PETM highlights the potential sensitivity
proxies of vegetation, fire regime, and lake of high-latitude peatlands and reservoirs of
water balance, Nelson and Hu (2008) ar- soil carbon to externally forced warming
gued that these abrupt local vegetation and the ability of soil-carbon feedbacks to
changes were direct responses to locally rapidly accelerate global warming. Terrestrial
rapid declines in water availability and veg- ecosystems responded rapidly to the PETM,
etation-climate-fire feedbacks. Williams et with some effects limited to the PETM itself
al. (2010) argued that locally abrupt vegeta- (high rates of plant community turnover,
tion responses with a regionally time-trans- mammalian body size dwarfing), others
gressive pattern spanning multiple ecotones permanent (intercontinental immigration of
(grassland to desert, C4 to C3 grasses, forest species across land corridors made available
to prairie) (Figure 10.2) indicate that site- through climate). Unlike the marine realm,
specific ecological tipping points were the there is no evidence for enhanced rates of
primary cause of locally abrupt responses. terrestrial extinction during the PETM.
PA ST A BRU P T C H A NGE S I N C L I M AT E A N D T E R R E ST R I A L E CO S Y ST E M S 139
The last deglaciation shows that rapid Willy Tinner, and Willem van der Knaap
climate change can have near-immediate for contributing data. This work benefited
effects on tree abundances and forest com- from discussions with University of Wis-
position. Forest response times were con- consin–Madison graduate seminar partici-
sistently less than 20 to 40 years and often pants in the spring of 2015, and was sup-
had no detectable time lag. There is no obvi- ported by the National Science Foundation
ous difference in the initial response time (EF-1241868, DEB-1353896).
between short-lived herbaceous and long-
lived arboreal species, or between aquatic
and terrestrial organisms. Nevertheless, at R E FE R E NCE S
several sites, particularly Arctic and alpine
Aitken, S. N., S. H. Yeaman, J. A. Holliday, T. Wang, and
sites, the initial temperature rise triggered S. Curtis-McLane. 2008. “Adaptation, migration or ex-
a series of vegetation changes that contin- tirpation: Climate change outcomes for tree popula-
ued for as long as 700 to 1,000 years after tions.” Evolutionary Applications 1: 95–111.
the initial warming. Likely causes for these Allen, C. D., A. K. Macalady, H. Chenchouni, D. Bachelet,
N. McDowell, M. Vennetier, T. Kitzberger, et al. 2010.
longer-term responses are dispersal limita-
“A global overview of drought and heat-induced tree
tions for tree species, local succession, and mortality reveals emerging climate change risks for
soil development. Migration rates for two forests.” Forest Ecology and Management 259: 660–684.
tree species are on the order of 0.6 to 1 km/ Alley, R. 2000. “The Younger Dryas cold interval as
year, albeit with substantial uncertainty. viewed from central Greenland.” Quaternary Science Re-
views 19: 213–226.
During the Holocene, abrupt shifts in
Alley, R. B., and A. M. Ágústsdóttir. 2005. “The 8k event:
terrestrial ecosystems and early agricultural Cause and consequences of a major Holocene abrupt
societies were common and often forced by climate change.” Quaternary Science Reviews 24: 1123–1149.
a combination of hydroclimatic variability Ammann, B., U. J. van Raden, J. Schwander, U. Eicher,
and intrinsic feedback loops and critical A. Gilli, S. M. Bernasconi, J. F. N. van Leeuwen, et
al. 2013. “Responses to rapid warming at Termination
thresholds. These manifest in many ways,
1a at Gerzensee (Central Europe): Primary succession,
including shifts in dunes from active to albedo, soils, lake development, and ecological inter-
vegetated states, rapid deforestation at prai- actions.” Palaeogeography, Palaeoclimatology, Palaeoecology 391,
rie, savanna, and forest ecotones, popula- pt. B: 111–131.
tion collapses of mesophilic tree species and Birks, H. H. 2015. “South to north: Contrasting late-
glacial and early-Holocene climate changes and veg-
abandonment of early agricultural sites. The
etation responses between south and north Norway.”
Great Plains and North Africa show how Holocene 25: 37–52.
slow regional aridification (over thousands Booth, R. K., S. T. Jackson, V. A. Sousa, M. E. Sullivan, T.
of years) can produce locally rapid ecologi- A. Minckley, and M. J. Clifford. 2012. “Multi-decadal
cal shifts, driven by locally abrupt precipi- drought and amplified moisture variability drove
rapid forest community change in a humid region.”
tation changes, critical thresholds, and/
Ecology 93: 219–226.
or positive feedbacks in semiarid systems Brauer, A., G. H. Haug, P. Dulski, D. M. Sigman, and J.
between climate, fire regime, and fuel. We Negendank. 2008. “An abrupt wind shift in Western
can expect similarly rapid and complex re- Europe at the onset of the Younger Dryas cold period.”
sponses to twenty-first-century aridification Nature Geoscience 1: 520–523.
Buffett, B., and D. Archer. 2004. “Global inventory of
and shifts in frequency of extreme events.
methane clathrate: Sensitivity to changes in the deep
ocean.” Earth and Planetary Science Letters 227: 185–199.
Buizert, C., V. Gkinis, J. P. Severinghaus, F. He, B. S.
ACK NOWLE DGME NTS Lecavalier, P. Kindler, M. Leuenberger, et al. 2014.
“Greenland temperature response to climate forcing
during the last deglaciation.” Science 345: 1177–1180.
We gratefully thank Matt McGlone for dis-
Clark, P. U., J. D. Shakun, P. A. Baker, P. J. Bartlein, S.
cussion and editing of this chapter and Bri- Brewer, E. J. Brook, A. E. Carlson, et al. 2012. “Global
gitta Ammann, Hilary Birks, Achim Brauer, climate evolution during the last deglaciation.” Proceed-
André Lotter, Oliver Rach, Tim Shanahan, ings of the National Academy of Sciences 109: E1134–E1142.
140 W H AT D OE S T H E PA ST T E L L US?
Claussen, M. 2009. “Late Quaternary vegetation-climate Kröpelin, S., D. Verschuren, A.-M. Lézine, H. Egger-
feedbacks.” Climate of the Past 5: 203–216. mont, C. Cocquyt, P. Francus, J.-P. Cazet, et al. 2008.
Cobb, K. M., N. Westphal, H. R. Sayani, J. T. Watson, “Climate-driven ecosystem succession in the Sahara:
E. Di Lorenzo, H. Cheng, R. L. Edwards, and C. D. The past 6000 years.” Science 320: 765–768.
Charles. 2013. “Highly variable El Niño–Southern Os- Kuper, R., and S. Kröpelin. 2008. “Climate-controlled
cillation throughout the Holocene.” Science 339: 67–70. Holocene occupation in the Sahara: Motor of Africa’s
Cook, E. R., K. J. Anchukaitis, B. M. Buckley, R. D. evolution.” Science 313: 803–807.
D’Arrigo, G. C. Jacoby, and W. E. Wright. 2010a. Liu, Z., Y. Wang, R. Gallimore, F. Gasse, T. Johnson, P.
“Asian monsoon failure and megadrought during the deMenocal, J. Adkins, et al. 2007. “Simulating the
last millennium.” Science 328: 486–489. transient evolution and abrupt change of Northern
Cook, E. R., R. Seager, R. R. Heim, R. S. Vose, C. Her- Africa atmosphere-ocean-terrestrial ecosystem in the
weijer, and C. Woodhouse. 2010b. “Megadroughts in Holocene.” Quaternary Science Reviews 26: 1818–1837.
North America: Placing IPCC projections of hydrocli- Marcott, S. A., J. D. Shakun, P. U. Clark, and A. C.
matic change in a long-term palaeoclimate context.” Mix. 2013. “A reconstruction of regional and global
Journal of Quaternary Science 25: 48–61. temperature for the past 11,300 years.” Science 339:
Currano, E. D., P. Wilf, S. L. Wing, C. C. Labandeira, 1198–1201.
E. C. Lovelock, and D. L. Royer. 2008. “Sharply in- Marsicek, J., B. N. Shuman, P. J. Bartlein, S. L. Shafer,
creased insect herbivory during the Paleocene–Eocene and S. Brewer. 2018. “Reconciling divergent trends
Thermal Maximum.” Proceedings of the National Academy of and millennial variations in Holocene temperatures.”
Sciences 105: 1960–1964. Nature 554: 92.
DeConto, R. M., S. Galeotti, M. Pagani, D. Tracy, K. McInerney, F. A., and S. L. Wing. 2011. “The Paleocene-
Schaefer, T. Zhang, D. Pollard, and D. J. Beerling. Eocene Thermal Maximum: A perturbation of carbon
2012. “Past extreme warming events linked to mas- cycle, climate, and biosphere with implications for
sive carbon release from thawing permafrost.” Nature the future.” Annual Review of Earth and Planetary Sciences 39:
484: 87–92. 489–516.
deMenocal, P. B. 2001. “Cultural responses to cli- Miao, X., J. A. Mason, J. B. Swinehart, D. B. Loope, P.
mate change during the late Holocene.” Science 292: R. Hanson, R. J. Goble, and X. Liu. 2007. “A 10,000
667–673. year record of dune activity, dust storms, and se-
Foster, D. R., W. W. Oswald, E. K. Faison, E. D. Doughty, vere drought in the central Great Plains.” Geology 35:
and B. C. S. Hansen. 2006. “A climatic driver for abrupt 119–122.
mid-Holocene vegetation dynamics and the hemlock Munoz, S. E., K. E. Gruley, A. Massie, D. A. Fike, S. Schro-
decline in New England.” Ecology 87: 2959–2966. eder, and J. W. Williams. 2015. “Cahokia’s emergence
Gingerich, P. D. 2006. “Environment and evolution and decline coincided with shifts of flood frequency
through the Paleocene-Eocene thermal maximum.” on the Mississippi River.” Proceedings of the National Acad-
Trends in Ecology & Evolution 21: 246–253. emy of Sciences 112: 6319–6324.
Goring, S. J., and J. W. Williams. 2017. “Effect of historic National Research Council. 2013. “Abrupt impacts of cli-
land-use and climate change on tree-climate relation- mate change: Anticipating surprises.” National Acad-
ships in the upper Midwestern United States.” Ecology emy of Sciences, Washington, DC.
Letters 20: 461–470. Nelson, D. M., and F. S. Hu. 2008. “Patterns and driv-
Graham, R. W., S. Belmecheri, K. Choy, B. Culleton, ers of Holocene vegetational change near the prairie-
L. J. Davies, D. Froese, P. D. Heintzman, et al. 2016. forest ecotone in Minnesota: Revisiting McAndrews’
“Timing and causes of a middle Holocene mammoth transect.” New Phytologist 179: 449–459.
extinction on St. Paul Island, Alaska.” Proceedings of the Rach, O., A. Brauer, H. Wilkes, and D. Sachse. 2014. “De-
National Academy of Sciences 113: 9310–9314. layed hydrological response to Greenland cooling at
Higgins, J. A., and D. P. Schrag. 2006. “Beyond meth- the onset of the Younger Dryas in Western Europe.”
ane: Towards a theory for the Paleocene-Eocene Ther- Nature Geoscience 7: 109–112.
mal Maximum.” Earth and Planetary Science Letters 245: Scheffer, M., M. Hirota, M. Holmgren, E. H. Van Nes,
523–537. and F. S. Chapin. 2012. “Thresholds for boreal biome
Jackson, S., and C. Weng. 1999. “Late Quaternary extinc- transitions.” Proceedings of the National Academy of Sciences
tion of a tree species in eastern North America.” Pro- 109: 21384–21389.
ceedings of the National Academy of Sciences 96: 13847–13852. Seddon, A. W., M. Macias-Fauria, and K. J. Willis. 2015.
Jackson, S. T., J. L. Betancourt, R. K. Booth, and S. T. “Climate and abrupt vegetation change in Northern
Gray. 2009. “Ecology and the ratchet of events: Cli- Europe since the last deglaciation.” The Holocene 25:
mate variability, niche dimensions, and species distri- 25–36.
butions.” Proceedings of the National Academy of Sciences 106: Shanahan, T. M., N. P. McKay, K. A. Hughen, J. T. Over-
19685–19692. peck, B. Otto-Bliesner, C. W. Heil, J. King, C. A.
Koch, P. L., and A. D. Barnosky. 2006. “Late Quaternary Scholz, and J. Peck. 2015. “The time-transgressive ter-
extinctions: State of the debate.” Annual Review of Ecology, mination of the African Humid Period.” Nature Geosci-
Evolution, and Systematics 37: 215–250. ence 8: 140–144.
PA ST A BRU P T C H A NGE S I N C L I M AT E A N D T E R R E ST R I A L E CO S Y ST E M S 141
Shuman, B. N. 2012. “Patterns, processes, and impacts Williams, J. W., B. Shuman, P. J. Bartlein, N. S. Diffen-
of abrupt climate change in a warm world: The past baugh, and T. Webb III. 2010. “Rapid, time-trans-
11,700 years.” WIREs Climate Change https://doi.org/10 gressive, and variable responses to early-Holocene
.1002/wcc.152. midcontinental drying in North America.” Geology 38:
Steffensen, J. P., K. K. Andersen, M. Bigler, H. B. Clau- 135–138.
sen, D. Dahl-Jensen, H. Fischer, K. Goto-Azuma, et al. Williams, J. W., B. N. Shuman, T. Webb III, P. J. Bartlein,
2008. “High-resolution Greenland ice core data show and P. L. Leduc. 2004. “Late Quaternary vegetation
abrupt climate change happens in few years.” Science dynamics in North America: Scaling from taxa to bi-
321: 680–684. omes.” Ecological Monographs 74: 309–334.
Svenning, J.-C., and B. Sandel. 2013. “Disequilibrium Wing, S. L., G. J. Harrington, F. A. Smith, J. I. Bloch, D.
vegetation dynamics under future climate change.” M. Boyer, and K. H. Freeman. 2005. “Transient floral
American Journal of Botany 100: 1266–1286. change and rapid global warming at the Paleocene-
Tinner, W., and A. F. Lotter. 2001. “Central European Eocene boundary.” Science 310: 993–996.
vegetation response to abrupt climate change at 8.2 Zachos, J. C., U. Röhl, S. A. Schellenberg, A. Sluijs, D. A.
ka.” Geology 29: 551–554. Hodell, D. C. Kelly, E. Thomas, M. Nicolo, I. Raffi, L.
Williams, J. W., J. L. Blois, J. L. Gill, L. M. Gonzales, E. C. J. Lourens, H. McCarren, and D. Kroon. 2005. “Rapid
Grimm, A. Ordonez, B. Shuman, and S. Veloz. 2013. acidification of the ocean during the Paleocene-Eo-
“Model systems for a no-analog future: Species asso- cene Thermal Maximum.” Science 308: 1611–1615.
ciations and climates during the last deglaciation.” An- Zhang, P., H. Cheng, R. L. Edwards, F. Chen, Y. Wang,
nals of the New York Academy of Sciences 1297: 29–43. X. Yang, J. Liu, et al. 2008. “A test of climate, sun, and
Williams, J. W., J. L. Blois, and B. N. Shuman. 2011. culture relationships from an 1810-year Chinese cave
“Extrinsic and intrinsic forcing of abrupt ecological record.” Science 322: 940–942.
change: Case studies from the late Quaternary.” Journal
of Ecology 99: 664–677.
CHAPTER ELEVEN In this chapter we explore the effects of
climate change on Amazonian and Andean
A Neotropical ecosystems. Rather than documenting sim-
ple upslope and downslope range changes,
Perspective on Past we consider the heterogeneity of both
modern and past events. Layers of complex-
Human-Climate ity are added as biotic and abiotic factors are
considered. Last, we add human influence
Interactions and into the mix, emphasizing that fire, which
is almost uniquely associated with human
Biodiversity activity in these forests, will accelerate and
amplify the effects of climate change.
MARK B. BUSH Tropical landscapes are hugely var-
ied, ranging from ice-capped mountains
to lowland tropical rainforest. Even low-
land tropical rainforests are ecologically
highly heterogeneous, often varying as a
result of edaphic, hydrological, historical,
or climatic factors. Heterogeneity of spe-
cies composition and structure across many
spatial scales is the natural state of tropical
systems. At local scales, a single hectare of
mature forest may support more than 350
species of tree, plus a wealth of treelets,
vines, and epiphytes. This diversity has led
to a widely held misconception is that all
species are relatively rare in these forests.
Certainly, the near monodominance seen in
boreal forest trees is absent from neotropi-
cal systems, but tropical tree species are not
all equally rare. Among the approximately
16,000 tree species in Amazonia, about 230
species routinely account for 50 percent of
the stems (ter Steege et al. 2013). The rea-
son for the abundance of these so-called
hyperdominant species may be related to
niche and competition success (Pitman
et al. 2001), neutrality (Hubbell 2001), or
past human activity (Roosevelt 2013). Few
studies of the historical patterns leading to
hyperdominance have been conducted in
Amazonia, but the abundance of the palm
142
A N E OT ROPIC A L PE R SPE C T I V E ON PA ST H U M A N- C L I M AT E I N T E R AC T IONS 143
Mauritia may have been enhanced by human free-draining white-sand soil may be vul-
activity (Rull and Montoya 2014). Contrast- nerable to drought.
ingly, Iriartea, another hyperdominant palm, Tropical climates have varied substan-
showed an overall increase in abundance tially over time. Within the past 2 million
in the past 3,000 years but probably as a years there have been at least 22 ice ages,
result of wetter conditions rather than hu- the last five of which were more extreme
man activity (Bush and McMichael 2016). and lasted longer than previous ones.
These data suggest that membership in the During ice ages, ice caps expanded in the
hyperdominant class of trees may be labile. mountains, with glaciers often extending
Whatever the cause of the success of these about 1500 m–2000 m downslope, leaving
most abundant species, their traits (canopy terminal moraines at about 3200 m–3800
architecture, fruit production, phenology, m elevation (wetter locations supported
leaf palatability, and flammability) play an longer glaciers). Estimates of temperature
important role in shaping current niche change in the high Andes reveal a cooling
availability for other forest organisms. of approximately 7°C–9°C during the last
Many factors contribute to fine-scale ice age (Hooghiemstra 1984; Van ‘t Veer
heterogeneity in forests, including micro- and Hooghiemstra 2000). In the lowlands,
climatic conditions, history, soil moisture the cooling was not as extreme, with many
gradients, flood regimes, time since last locations 2°C–4°C cooler than present for
disturbance, soil type, and frugivore popu- much of the ice age. Brief intervals, lasting
lation size and type. In mountainous areas, a few thousand years, cooled the lowlands
the strongest predictor of species presence is by about 5°C–6°C (Bush, de Oliveira, et al.
temperature, which is negatively correlated 2004; Bush, Silman, et al. 2004. Changes in
with elevation (Richards 1996). Indeed, the precipitation were highly variable in space
most predictable temperature gradients in and time but probably just as influential in
the tropics are associated not with latitude reshaping communities as temperature.
but with elevation. On the flank of the An- Haffer (1969) proposed the refugial hy-
des, temperatures decrease approximately pothesis as a means to account for modern
–5.5°C per 1000 m of vertical ascent. The biogeographic patterns in birds. Haffer’s
steep slopes cause localities just 3 km apart hypothesis that glacial-age aridity caused
to have temperatures differing by >5°C grasslands to replace most Amazonian rain-
(Bush 2002). forest is completely unsupported by paleo-
Regional heterogeneity is also influ- ecological or molecular data (Colinvaux
enced by climatic differences, especially in et al. 2001; Miller et al. 2008; Wang et al.
total amount and seasonality of precipita- 2017). The switching between wet and dry
tion. Within lowland Amazonia there are episodes was faster and not as profound
evergreen forests receiving in excess of as that implied by the refugial hypothesis.
10,000 mm of rain per year, whereas oth- At the risk of oversimplifying the pattern,
ers may receive only 1,500 mm. If the rain- there was a tendency toward alternating
fall is evenly spread through the year less periods of drier and wetter average condi-
is needed to maintain an evergreen forest tions, each lasting approximately 10,000
than in one that experiences a 3-, 4-, or years due to Earth’s precessional variability
5-month-long dry season (Silman 2011). in its orbit around the Sun (Bush et al. 2002;
Soils, too, play a critical role in determining Cruz et al. 2005). This relationship was not
the effects of rainfall. For example, plants invariant, and outside factors, such as the
growing on a slow-draining clay soil will oceanic circulation of the Atlantic Ocean,
be prone to waterlogging, while under the periodically trumped the precessional sig-
same rainfall regime plants growing on a nature (Mosblech et al. 2012; Stríkis et al.
144 W H AT D OE S T H E PA ST T E L L US?
2018). The most extreme drying may have apparent that it is very difficult to predict
occurred during interglacials rather than actual climatic outcomes for given locations
glacials. (Marengo et al. 2008; Marengo et al. 2011).
A potential pitfall in thinking about This uncertainty of outcome increases the
climate is to use the current interglacial greater the distance from the event. A com-
as a representative example of other inter- mon theme, however, is the importance of
glacials. Rather than being warm and wet abrupt climate change in contributing to
like the modern Holocene, analysis of long spatial and temporal climatic heterogene-
sedimentary records has shown that the last ity. It may well be that larger phenomena,
interglacial (Marine Isotope Stage 5e) was such as Dansgaard-Oeschger cycles (Bond
the driest time in the Bolivian Altiplano et al. 1993), were climatically far more vari-
in the past half million years (Baker et al. able within each event than is generally
2001; Fritz et al. 2007). Multimillennial believed.
dry events also occurred within glacial pe- Although the Holocene has been por-
riods, but the relatively low temperatures trayed as a period of climatic stability,
may have reduced evapotranspirative stress some parts of the Andes, Amazonia, and
and thus had a relatively modest impact on southeastern Brazil experienced profound
vegetation in contrast to the refugial model droughts between 9,000 and 5,000 years
(Sato and Cowling 2017). ago (Mayle et al. 2000; Abbott et al. 2003;
During the last ice age, millennial-scale Raczka et al. 2013). Lake Titicaca shrank to
climate events, called Dansgaard-Oeschger half its modern size (Seltzer et al. 1998),
cycles, had the capacity to cause approxi- many shallower lakes dried out completely,
mate 5°C–10°C swings in average tempera- and the proportion of grasses in the Cara-
ture in Greenland. These events often started jas pollen record is higher than at any time
with a very abrupt warming of ~5°C oc- in the past 100,000 years (Absy et al. 1991;
curring in the space of 10–30 years (North Hermanowski et al. 2013). Humans who
Greenland Ice-Core Project [NGRIP] 2004), had only occupied the high Andes about
followed by 3,000–7,000 years of cooling. 12,800 years ago (Rademaker et al. 2014)
These same events affected the tropics but abandoned large areas, and instead of a
generally as changes in precipitation rather signal of exponentially increasing human
than temperature. The effect of these events impact, there is a zone of archaeological
was location specific, with eastern and “silence” (Núñez et al. 2001). More detailed
western Amazonia responding in opposite investigations of this period show—prob-
ways (Cruz et al. 2009). In general, cold ably like many previous dry events—a
events in the Atlantic caused western Ama- highly heterogeneous climate with intense
zonia to become wetter while the eastern wet events punctuating a time of overall
Amazon became drier. More locally, if the lower precipitation (Hillyer et al. 2009).
modern response to anomalously high tem- Indeed, we should be cautious about
peratures in the Atlantic in 2010 is a guide, simplifying complex climatic histories to
climatic responses would have been very infer millennial-scale wet or dry events. Bi-
heterogeneous (Aragão and Shimabukuro ologically, wet interludes within an overall
2010). In 2010, anomalously warm tem- dry event make a huge difference, as wa-
peratures in the northern tropical Atlantic ter tables and root zones can be recharged
caused profound drought in some areas of with water, alleviating some of the effects
western Amazonia while other areas were of the drying. Similarly, as we think about
largely unchanged. As climatic impacts of biological impacts of dry events in the past
phenomena such as the El Niño–Southern or the future, the timing of those events
Oscillation or the Atlantic Multidecadal Os- is critically important. If wet-season rains
cillation become more familiar, it becomes are diminished but the length of the wet
A N E OT ROPIC A L PE R SPE C T I V E ON PA ST H U M A N- C L I M AT E I N T E R AC T ION S 145
season is not affected, there would be a under continuous but different adaptive
minimal biological response in that forest, or evolutionary pressures (Overpeck et al.
as water is not limiting in the wet season. 1985). Ecological disequilibrium was the
Such a change, however, does not preclude norm and played a key role in maintaining
larger ecosystem effects as reduced flood- species richness through time. Responses to
water could alter downstream hydrological climate change would have varied greatly
regimes. If, by contrast, the dry season be- according to landscape and the mobility
comes longer, or the number of consecutive of species; that is, birds can migrate much
days without rain increases, the drought more freely than trees. Temperate system re-
deficit and increased fire probability could sponses to ongoing warming cause upslope
reshape forests (Bush and Silman 2004). or poleward range shifts (Chapter 3). In the
Fire is a great game changer for bio- tropics, the species that live at high eleva-
diversity. In wet forests that do not burn tions have only an upslope route to escape
naturally, a single fire will cause complete warmth as there is little appreciable benefit
turnover in ground nesting bird species, in terms of cooling to migrate from 0° lati-
and repeated fires quickly reduce forest bio- tude to 10° latitude. On mountain flanks,
mass, simplify physical structure, and cause tropical lowland taxa may be able to mi-
almost complete species turnover among grate upslope relatively short migrational
trees (Barlow 2002; Barlow et al. 2003). In distances to maintain a given temperature
much of western lowland Amazonia, fire (Bush 2002). Upslope and downslope re-
was not a natural component of Holocene sponses to past warming and cooling are
ecosystems, but it did accompany human readily documented in the Central Ameri-
occupation. A 4,000-year sediment his- can Isthmus and the flank of the Andes. An
tory from Lake Werth in southeastern Peru asymmetry exists, however, in the capacity
contained no sign of human occupation or of species to migrate in response to warm-
fire in that catchment (Bush et al. 2007). By ing (upslope) versus cooling (downslope).
contrast, just a few kilometers away, Lakes Downslope migration during cold events
Gentry, Parker, and Vargas all showed peri- can be relatively rapid for fluvially dispersed
ods of human occupation and have complex species, with seeds that can withstand im-
fire histories (McMichael et al. 2013). Fires mersion in water for days or weeks. Seeds of
are difficult to ignite in these wet forests, so many species are spread downslope by riv-
it is probable that humans took advantage of ers and their ranges could expand quickly
seasonal droughts to burn and clear forests. through a form of jump dispersal (Pielou
Human impacts on Amazonian and high 1979), but only as conditions cool suffi-
Andean (>3000 m elevation) systems were ciently can cold-adapted species compete in
probably very localized until approximately the lowlands. In contrast, animal-dispersed
10,000 years ago (Dillehay 2009). The areal species expand their ranges incrementally
extent and temporal trajectory of human (Figure 11.1). During cooling events, many
impacts in lowland Amazonia continues low-elevation species see no benefit, as they
to be actively debated, but it is clear that, are not cold requiring, although some bi-
through fire, human action had the capacity otic interactions would be altered, creating
to completely alter occupied landscapes. a different competitive balance of “win-
ners” and “losers.” Consequently, past cool-
ing enabled some montane species (e.g.,
BIOTIC M E C H A N ISMS OF Hedyosmum, Podocarpus, Myrsine) to expand
H E T E RO GE N E IT Y their presence in the lowlands (Colinvaux
et al. 2000), but the real “winners” during
Past climate fluctuations caused re-assort- these times were probably lowland species
ments of species into novel assemblages, favored by the new conditions. In other
146 W H AT D OE S T H E PA ST T E L L US?
Figure 11.1. Selected factors influencing rates of plant mi- Amazon plain is so flat (Bush and Colinvaux
gration following climate change. An elevational profile 1990). As species ranges contracted upslope
of the Andes and Amazon lowlands. Arrows represent
the probable speed of seed transport. Microrefugia can
during warming, pockets of outlying indi-
increase the capacity of species or genotypes to expand viduals were left behind in climatic micro-
rapidly. As temperature change is the strongest driver refugia (e.g., where a cool spring created a
of migration in montane areas, the microrefugia would cool microclimate). Where greater ranges
reflect unusual thermal microclimates. In the lowlands in topography existed, such as in the high
where soil moisture availability is more likely to in-
fluence the success of species, microrefugia would be
Andes, steep slopes may have been a key
primarily edaphically or hydrologically driven. Seeds predictor of microrefugial locations as they
of thermally inhibited (cold tolerant) species are trans- trapped moisture and inhibited the spread
ported by rivers downslope continuously, but establish of fire (Valencia et al. 2016).
only during cool periods. Fluvial upslope transport is not In lowland Amazonia, there is currently
available, and therefore species cannot migrate upslope
or must be moved by animals. Lowland taxa have the
no evidence showing that trees migrated to
same dispersal constraints as the upslope counterparts warmer areas during glacial cooling. Cer-
but lack strong, directional abiotic gradients to facilitate tainly, forests were enriched by the cool-
migration. The net effect is that species with different tolerant taxa, but there is no evidence of
dispersal traits have differential probabilities of disper- species loss due to cooling. Probably the
sal, and downslope migration akin to jump dispersal in
response to cooling is more likely than rapid upslope
temperature changes were within the range
migration in response to warming. of evolutionary experience and hence not
unduly threatening. Responses to moisture
words, the hyperdominants of the glacial changes may have been more pressing. Am-
forest were probably still lowland species, azonia almost certainly remained forested
but not necessarily the same ones as today. during Quaternary ice ages and intergla-
During warm events, upslope migration cials, although the forest was composition-
relies heavily on mammals and birds, and is ally, and perhaps structurally, different from
probably much slower than downslope ex- those of today (Mayle et al. 2004). Species
pansions. For the great majority of trees in stressed by the glacial cooling, or by wet
the lowlands of Amazonia there is no escape or dry events within the past, may simply
from changing temperature. An insufficient have become rarer or more range restricted.
gradient of temperature exists to favor sur- For example, during times of drought,
vival at one edge of a range, and there is no when low CO2 concentrations would have
escape using macrotopography, as the vast increased drought stress, the most sensitive
A N E OT ROPIC A L PE R SPE C T I V E ON PA ST H U M A N- C L I M AT E I N T E R AC T ION S 147
species may have been restricted to micro- Forests are emptied of wildlife by hunting,
refugia in low-lying wet areas or along wa- which reduces the dispersal ability of large-
tercourses (Rull 2009). Such microrefugia seeded plants and ultimately leads to altered
are increasingly being seen as vitally impor- forest composition and structure (Terborgh
tant to maintaining high biodiversity (Mc- 2001; Peres et al. 2016).
Glone and Clark 2005; Collins et al. 2013). A classic example of a synergistic cli-
As a consequence of local migration, mi- mate-human mediated simplification of
crorefugia may have supported population systems is the replacement of the Pleisto-
fragmentation, local selection, and survival cene megafauna. About 57 genera of mega-
of particular genotypes. Population genet- fauna went extinct at the end of the last ice
ics shows that the legacies of microrefugia age in South America (Barnosky and Lind-
persist in genetically identifiable subpopu- sey 2010). Recent research in the Andes in-
lations for thousands of years after the ame- dicates that at least some of the extinctions
lioration of climatic conditions enabled the were initiated by climate change (Rozas-
species to expand its range (McLachlan and Davila et al. 2016), although the final blows
Clark 2005). Estimates of range expansion were by humans. The timing of apparent
for many tree species provide unrealistically extinction differs between sites, ranging
high required rates of migration unless mi- from ~16,000 years ago in the Peruvian
crorefugia are invoked to reduce migratory Andes (Rozas-Davila et al. 2016) to 12,800
distances (Clark et al. 1998; Correa-Metrio years ago in southeastern Brazil (Raczka
et al. 2013). Under more extreme isola- et al. 2018) and about 12,500 years ago in
tion and selection, fragmented populations Patagonia (Metcalf et al. 2016). The role of
could evolve in isolation to become a sepa- seed dispersers, grazers, and browsers—a
rate species. Thus, microrefugia contribute role once performed by those megafauna—
to patchiness of species abundances and to is now filled by people, horses, and cows.
heterogeneity at both long and short tem- North America lost slightly fewer genera
poral scales (Mosblech et al. 2011). of megafauna than South America, but the
Colwell (2008) has suggested that be- species were for the most part quite differ-
cause the temperature range in the past 2 ent. Yet the replacement animals that today
million years was effectively +1°C–2°C and fulfill their roles are again humans, horses,
–5°C relative to modern in Amazonia, the and cows. The same story can be told of
upper temperature threshold of prior evo- Australia, and even of Europe. Plants did not
lutionary experience is approaching. With experience such a cataclysmic extinction at
more warming and limited options for mi- the end of the last ice age, but the homog-
gration, there may well be a temperature enization of floras is evident worldwide as
threshold beyond which there could be a a result of human activity. On the once en-
wave of plant extinctions. The more imme- demic-dominated Galapagos Islands, exotic
diate threat will come from climate desta- plant species now outnumber native ones
bilization as well as increased probability of (Toral-Granda et al. 2017). In the Andes,
drought and forest-changing fire, coupled unique montane forests have been replaced
with human land use. by planted stands of Pinus and Eucalyptus. In
both the uplands and lowlands of South
America, European and African grasses are
T H E PAT H TO HOMO GE N I Z ATION replacing neotropical grasses. At continen-
tal and regional scales, humans continue to
People tend to simplify landscapes, intro- simplify and homogenize ecosystems.
ducing domesticated plants and animals, A recent debate has focused on the role
promoting fire, and altering soils and drain- of pre-Columbian peoples in altering Ama-
age to suit their needs (Kareiva et al. 2007). zonian landscapes. Some archaeologists and
148 W H AT D OE S T H E PA ST T E L L US?
historical ecologists have argued that Ama- if it is incorrect, then the forest would be
zonian ecosystems were so modified that used and lost through misunderstanding. If
they should be regarded as manufactured the more conservative view of disturbance
landscapes (e.g., Heckenberger et al. 2003; being inimical to forest persistence is cor-
Erickson 2008; Roosevelt 2013; Levis et al. rect, then the forest should be conserved.
2017). However, empirical data from >400 Should this view prove incorrect, the worst
randomly distributed soil pits in western that could happen is that the forest be con-
Amazonia did not produce a single example served for later use when an informed deci-
of a modified soil or a human artifact (Mc- sion could be made regarding its loss (Bush
Michael et al. 2012). In light of the archae- and Silman 2007; Bush et al. 2015).
ological and paleoecological data it seems
likely that while major river channels sup-
ported substantial but patchy human popu- L I N E A R A N D NON L I N E A R R E SPONSE S
lations, the inferred disturbance should not TO C L IM AT E C H A NGE
be extrapolated to include areas much more
than a day’s walk from those centers of oc- Cascading ecological change is a natural
cupation (Piperno et al. 2015; McMichael et state following any disturbance, whether
al. 2017). climatic or human-induced (the severity
The debate has generated two very dif- of which often determines the scale of the
ferent conclusions about the resilience response). In general, recovery from a dis-
of Amazonian forests (Bush and Silman turbance is broadly predictable, with spe-
2007). On the one hand, if the forests are cies going through a recognizable succes-
a manufactured landscape and are essen- sion. Similarly, response to climate change
tially a product of human disturbance, then is highly predictable when the response is
they should be expected to be resilient to driven by abiotic factors, such as tempera-
modern disturbance. Past human activi- ture gradients. Where responses become
ties included forest clearance and burning, less predictable is when the primary control
which would have opened the system to on the niche of the species is biotic rather
nutrient loss and extensive hunting. If hu- than abiotic (Schemske et al. 2009). In natu-
mans occupied these forests in the past at ral tropical systems, the importance of biotic
the high densities advocated by some, then factors influencing local diversity often ex-
the forests would have been game poor. The ceeds that of abiotic factors. Consequently,
empty-forest syndrome in which the trees for tropical species, the realized niche of a
remained but the animals were hunted out species may be much smaller than its fun-
(Redford 1992) could easily have prevailed. damental niche. If that is the case, then the
According to this view, the forests will re- potential for counterintuitive responses to
grow to their present diversity even if their climate change would be greatly increased;
food webs are shattered and trees cut down. that is, the realized niche could be reshaped
The more conservative view is that resil- in any direction as a competitor or predator
ience in these systems comes from a closed responds to climate forcing or to an altera-
nutrient system and relatively intact faunas tion in its own biotic controls. The forma-
that are sensitive to, and altered by, fire and tion of no-analogue assemblages in which
human disturbance. novel combinations of species co-occur has
The dichotomy extends to the policy and been shown to have occurred during past
conservation implications of these views. climate change and can be expected in the
In the former, where resilience is assumed, future.
it could lead to a policy advocating forest Positive feedback mechanisms can ac-
clearance in the expectation of regrowth. celerate change, inducing disproportion-
If this view is correct, all would be well; ately large responses to a forcing. An ex-
A N E OT ROPIC A L PE R SPE C T I V E ON PA ST H U M A N- C L I M AT E I N T E R AC T IONS 149
ample may be the Bolivian Altiplano where ment, which promotes more road build-
warming during the last interglacial caused ing and yet more change (Nepstad et al.
abiotic factors to operate in a nonlinear 2001; Barber et al. 2014). People introduce
manner through the positive feedback of fire either accidentally or deliberately, and
lake shrinkage and changing microclimates during times of drought, those fires trans-
(Bush et al. 2010). With the onset of the in- form forest ecology. Thus, the biodiversity
terglacial, warming initially caused a retreat risk to Amazonia comes from both govern-
of ice caps and an upslope expansion of tree ment policies that encourage occupation
species. Instead of trees continuing to move of forest areas and periodic drought rather
upslope as the warming progressed, how- than from a simple protracted reduction in
ever, a tipping point was reached, when precipitation.
evaporative loss progressively lowered the
lake level. As the lake shrank, it no longer
supported evaporation-driven convective CONC LUSIONS
rains and a moist microclimate. The drier
conditions halted the upslope march of For almost half a century a discussion has
trees, and the landscape became saltbush centered on whether temperature, atmo-
dominated. spheric CO2 fluctuations, or precipitation
Another example of positive feedbacks shapes Amazonian forests during climatic
fueled modeled projections of an “Amazo- cycles. More recently, a focus on micro-
nian dieback” in which 80 percent of Ama- refugia has emphasized the individualis-
zonian forests were replaced with savanna tic nature of species’ responses. While the
or bare ground in the coming 50–100 years species driven upslope and downslope by
(Betts et al. 2004). More recent papers have glacial-interglacial temperature changes
recognized a weakness in the precipitation would have thermal microrefugia, low-
estimates of the initial model, and the dire land taxa would probably have been chal-
projections have been retracted (Hunting- lenged to survive changes in soil moisture
ford et al. 2013). While the original model, availability rather than temperature. Not all
sensu stricto, may have exaggerated impacts taxa are equally able to migrate, with water-
of reduced precipitation, the possibility that borne tree taxa able to disperse faster than
a combination of human-mediated land animal-dispersed taxa. The expansion of
clearance and fire activity coupled with a species during periods of cool (downslope)
destabilized climate could induce the pre- migration would be expected to be faster
dicted pattern is very real. The potential of than during warm (upslope) migrations be-
deforestation to degrade the hydrological cause rivers flow downslope.
cycle could induce feedbacks that increase The standard response of migrations
the likelihood of regional fires (Salati and poleward or upslope in response to future
Vose 1984; De Faria et al. 2017). Even if the warming may have limited applicability to
result is not savannization of the landscape, the tropics. Migration is certainly possible
the dispersal of humans ever deeper into in mountainous areas, where maintaining
Amazonian landscapes produces a differ- vertical corridors to facilitate that move-
ent modification to landscape than any pro- ment should be a conservation priority.
duced by pre-Columbian societies. Grazing Microrefugia are probably important to the
animals, African grasses, and weeds from long-term survival of many species in both
all over the world, paired with defaunation mountainous and lowland landscapes, and
of forests, will homogenize systems and re- to the extent that these settings are iden-
duce biodiversity (Solar et al. 2015). Here tifiable, they, too, should be priorities for
the positive feedback mechanism is one conservation. In the great flatlands of the
of new means of access begetting settle- Amazon Basin, microrefugia are probably
150 W H AT D OE S T H E PA ST T E L L US?
more important than migration in species gating effect of protected areas in the Amazon.” Bio-
survival. Here the clash between the prob- logical Conservation 177: 203–209.
Barlow, J., et al. 2002. “Effects of ground fires on under-
able location of microrefugia along wa- storey bird assemblages in Amazonian forests.” Biologi-
tercourses that are also preferred sites for cal Conservation 105: 157–169.
human habitation may result in a direct Barlow, J., C. A. Peres, B. O. Lagan, and T. Haugaasen.
conservation conflict. 2003. “Large tree mortality and the decline of forest
Although Amazonian forests have with- biomass following Amazonian wildfires.” Ecology Let-
ters 6: 6–8.
stood millennia of climate change, the new Barnosky, A. D., and E. L. Lindsey. 2010. “Timing of
threat posed is not from climate change Quaternary megafaunal extinction in South America
alone but from the synergy of future mega- in relation to human arrival and climate change.” Qua-
droughts coupled with human activity. ternary International 217: 10–29.
Unprecedented numbers of people, more Betts, R. A., P. M. Cox, M. Collins, P. P. Harris, C. Hunting-
ford, and C. D. Jones. 2004. “The role of ecosystem-at-
widely distributed than ever before, will in- mosphere interactions in simulated Amazonian precip-
crease forest flammability through land-use itation decrease and forest dieback under global climate
changes and climate volatility, and will thus warming.” Theoretical and Applied Climatology 78: 157–175.
provide new ignition sources. Human activ- Bond, G., W. Broecker, S. Johnsen, J. McManus, L.
ity homogenizes landscapes and ultimately Labeyrie, J. Jouzel, and G. Bonani. 1993. “Correlations
between climate records from the North Atlantic sed-
leads to the loss of biodiversity. While the iments and Greenland Ice.” Nature 365: 143–147.
scale of pre-Columbian human occupation Bush, M. B. 2002. “Distributional change and conserva-
is still debated, all would agree that contin- tion on the Andean flank: A palaeoecological perspec-
ued modern ecosystem replacement with tive.” Global Ecology and Biogeography 11: 463–467.
exotic crops such as soybeans and African Bush, M. B., and P. A. Colinvaux. 1990. “A long record of
climatic and vegetation change in lowland Panama.”
grasses will lead to an unparalleled loss of Journal of Vegetation Science 1: 105–119.
biodiversity and a completely unsustainable Bush, M. B., P. E. De Oliveira, P. A. Colinvaux, M. C.
use of the landscape. Miller, and E. Moreno. 2004. “Amazonian paleoeco-
logical histories: One hill, three watersheds.” Palaeo-
geography, Palaeoclimatology, Palaeoecology 214: 359–393.
Bush, M. B., J. A. Hanselman, and W. D. Gosling. 2010.
R E FE R E NCE S “Non-linear climate change and Andean feedbacks:
An imminent turning point?” Global Change Biology 16:
Abbott, M. B., B. B. Wolfe, A. P. Wolfe, G. O. Seltzer, 3223–3232.
R. Aravena, B. G. Mark, P. J. Polissar, D. T. Rodbell, Bush, M. B., C. N. H. McMichael, D. R. Piperno, M. R.
H. D. Rowe, and M. Vuille. 2003. “Holocene paleohy- Silman, J. Barlow, C. A. Peres, M. Power, and M. W.
drology and glacial history of the central Andes using Palace. 2015. “Anthropogenic influence on Amazo-
multiproxy lake sediment studies.” Palaeogeography, Pal- nian forests in pre-history: An ecological perspec-
aeoclimatology, Palaeoecology 194: 123–138. tive.” Journal of Biogeography 42: 2277–2288.
Absy, M. L., A. Clief, M. Fournier, L. Martin, M. Ser- Bush, M. B., and C. N. H. McMichael. 2016. “Holocene
vant, A. Sifeddine, F. d. Silva, F. Soubiès, K. T. Suguio, variability of an Amazonian hyperdominant.” Journal of
and T. van der Hammen. 1991. “Mise en évidence de Ecology 104: 1370–1378.
quatre phases d’ouverture de la forêt dense dans le Bush, M. B., M. C. Miller, P. E. de Oliveira, and P. A.
sud-est de L’Amazonie au cours des 60,000 dernières Colinvaux. 2002. “Orbital forcing signal in sediments
années: Première comparaison avec d’autres régions of two Amazonian lakes.” Journal of Paleolimnology 27:
tropicales.” Comptes rendus Academie des Sciences Paris, ser. 341–352.
II, 312: 673–678. Bush, M. B., and M. R. Silman. 2004. “Observations
Aragão, L. E. O. C., and Y. E. Shimabukuro. 2010. “The on Late Pleistocene cooling and precipitation in the
incidence of fire in Amazonian forests with implica- lowland neotropics.” Journal of Quaternary Science 19:
tions for REDD.” Science 328: 1275–1278. 677–684.
Baker, P. A., C. A. Rigsby, G. O. Seltzer, S. C. Fritz, T. K. Bush, M. B., and M. R. Silman. 2007. “Amazonian ex-
Lowenstein, N. P. Bacher, and C. Veliz. 2001. “Tropi- ploitation revisited: Ecological asymmetry and the
cal climate changes at millennial and orbital times- policy pendulum.” Frontiers in Ecology and the Environment
cales on the Bolivian Altiplano.” Nature 409: 698–701. 5: 457–465.
Barber, C. P., M. A. Cochrane, C. M. Souza Jr., and W. F. Bush, M. B., M. R. Silman, M. B. de Toledo, C. R. S. Lis-
Laurance. 2014. “Roads, deforestation, and the miti- topad, W. D. Gosling, C. Williams, P. E. de Oliveira,
A N E OT ROPIC A L PE R SPE C T I V E ON PA ST H U M A N- C L I M AT E I N T E R AC T ION S 151
and C. Krisel. 2007. “Holocene fire and occupation in nary glaciation and hydrologic variation in the South
Amazonia: Records from two lake districts.” Philosophi- American tropics as reconstructed from the Lake Titi-
cal Transactions of the Royal Society of London B 362: 209–218. caca drilling project.” Quaternary Research 68: 410–420.
Bush, M. B., M. R. Silman, and D. H. Urrego. 2004. Haffer, J. 1969. “Speciation in Amazonian forest birds.”
“48,000 years of climate and forest change from a Science 165: 131–137.
biodiversity hotspot.” Science 303: 827–829. Heckenberger, M. J., A. Kuikuro, U. T. Kuikuro, J. C.
Clark, J. S., C. Fastie, G. Hurtt, S. T. Jackson, C. Johnson, Russell, M. Schmidt, C. Fausto, and B. Franchetto.
G. A. King, M. Lewis, J. Lynch, S. Pacala, C. Prentice, E. 2003. “Amazonia 1492: Pristine forest or cultural
W. Schupp, T. I. Webb, and P. Wyckoff. 1998. “Reid’s parkland?” Science 301: 1710–1714.
paradox of rapid plant migration.” BioScience 48: 13–24. Hillyer, R., B. G. Valencia, M. B. Bush, M. R. Silman, and
Colinvaux, P., P. De Oliveira, and M. Bush. 2000. “Ama- M. Steinitz-Kannan. 2009. “A 24,700-year paleolim-
zonian and neotropical plant communities on glacial nological history from the Peruvian Andes.” Quaternary
time-scales: The failure of the aridity and refuge hy- Research 71: 71–82.
potheses.” Quaternary Science Reviews 19: 141–169. Hooghiemstra, H. 1984. “Vegetational and climatic his-
Colinvaux, P., G. Irion, M. Räsänen, M. Bush, and J. N. tory of the high plain of Bogota, Colombia.” Dissertacio-
De Mello. 2001. “A paradigm to be discarded: Geo- nes Botanicae 79. J. Cramer.
logical and paleoecological data falsify the Haffer & Hubbell, S. P. 2001. The Unified Neutral Theory of Biodiversity and
Prance refuge hypothesis of Amazonian speciation.” Biogeography. Princeton University Press.
Amazoniana 16: 609–646. Huntingford, C., P. Zelazowski, D. Galbraith, L. M. Mer-
Collins, A. F., M. B. Bush, and J. P. Sachs. 2013. “Mi- cado, S. Sitch, R. Fisher, M. Lomas, et al. 2013. “Simu-
crorefugia and species persistence in the Galápagos lated resilience of tropical rainforests to CO2-induced
highlands: A 26,000-year paleoecological perspec- climate change.” Nature Geoscience 6: 268–273.
tive.” Frontiers in Genetics 4. https://doi.org/10.3389/ Kareiva, P., S. Watts, R. McDonald, and T. Boucher. 2007.
fgene.2013.00269. “Domesticated nature: Shaping landscapes and eco-
Colwell, R. K., G. Brehm, C. L. Cardelus, A. C. Gilman, systems for human welfare.” Science 316: 1866–1869.
and J. T. Longino. 2008. “Global warming, elevational Levis, C., F. R. Costa, F. Bongers, M. Peña-Claros, C. R.
range shifts, and lowland biotic attrition in the wet Clement, A. B. Junqueira, E. G. Neves, E. K. Tamanaha,
tropics.” Science 322: 258–261. F. O. Figueiredo, and R. P. Salomão. 2017. “Persistent
Correa-Metrio, A., M. Bush, S. Lozano-García, and S. effects of pre-Columbian plant domestication on Am-
Sosa-Nájera. 2013. “Millennial-scale temperature azonian forest composition.” Science 355: 925–931.
change velocity in the continental northern neo- Marengo, J. A., C. A. Nobre, J. Tomasella, M. F. Cardoso,
tropics.” PLOS One 8: e81958. and M. D. Oyama. 2008. “Hydro-climatic and ecolog-
Cowling, S. A. 2004. “Tropical forest structure: A miss- ical behaviour of the drought of Amazonia in 2005.”
ing dimension to Pleistocene landscapes.” Journal of Philosophical Transactions of the Royal Society B: Biological Sciences
Quaternary Science 19: 733–743. 363: 1773–1778.
Cruz, F., M. Vuille, S. J. Burns, X. Wang, H. Cheng, M. Marengo, J. A., J. Tomasella, L. M. Alves, W. R. Soares,
Werner, R. L. Edwards, I. Karmann, A. S. Auler, and H. and D. A. Rodriguez. 2011. “The drought of 2010 in
Nguyen. 2009. “Orbitally driven east-west antiphas- the context of historical droughts in the Amazon re-
ing of South American precipitation.” Nature Geoscience gion.” Geophysical Research Letters 38: L12703.
2: 210–214. Mayle, F., R. Burbridge, and T. Killeen. 2000. “Millen-
Cruz Jr., F. W., S. J. Burns, I. Karmann, W. D. Sharp, M. nial-scale dynamics of southern Amazonian rain for-
Vuille, A. O. Cardoso, J. A. Ferrari, P. L. Silva Dias, ests.” Science 290: 2291.
and O. Viana Jr. 2005. “Insolation-driven changes in Mayle, F. E., D. J. Beerling, W. D. Gosling, and M. B.
atmospheric circulation over the past 116,000 years in Bush. 2004. “Responses of Amazonian ecosystems to
subtropical Brazil.” Nature 434: 63–66. climatic and atmospheric CO2 changes since the Last
De Faria, B. L., P. M. Brando, M. N. Macedo, P. K. Panday, Glacial Maximum.” Philosophical Transactions of the Royal
B. S. Soares-Filho, and M. T. Coe. 2017. “Current and Society of London B 359: 499–514.
future patterns of fire-induced forest degradation in McGlone, M. S., and J. S. Clark. 2005. “Microrefugia and
Amazonia.” Environmental Research Letters 12: 095005. Macroecology.” In Climate Change and Biodiversity, ed. T.
Dillehay, T. D. 2009. “Probing deeper into first American Lovejoy and L. Hannah, 157–159. Yale University Press.
studies.” Proceedings of the National Academy of Sciences 106: McLachlan, J. S., and J. S. Clark. 2005. “Molecular indi-
971–978. cators of tree migration capacity under rapid climate
Erickson, C. L. 2008. “Amazonia: The historical ecology change.” Ecology 86: 2088–2098.
of a domesticated landscape.” In The Handbook of South McMichael, C. H., M. B. Bush, M. R. Silman, D. R.
American Archaeology, ed. H. Silverman and W. H. Isbell, Piperno, M. Raczka, L. C. Lobato, M. Zimmerman, S.
157–183. Springer. Hagen, and M. Palace. 2013. “Historical fire and bam-
Fritz, S. C., P. A. Baker, G. O. Seltzer, A. Ballantyne, P. M. boo dynamics in western Amazonia.” Journal of Biogeog-
Tapia, H. Cheng, and R. L. Edwards. 2007. “Quater- raphy 40: 299–309.
152 W H AT D OE S T H E PA ST T E L L US?
McMichael, C. H., D. R. Piperno, and M. B. Bush. 2017. Raczka, M. F., P. D. De Oliveira, and M. B. Bush. 2018.
“Response to Levis et al.: Persistent effects of pre- “Megafaunal extinction in south-eastern Brazil.” Qua-
Columbian plant domestication on Amazonian for- ternary Research 89: 103–118.
est composition.” Science 358. https://doi.org/10.1126/ Redford, K. 1992. “The empty forest.” BioScience 42:
science.aan8347. 412–422.
McMichael, C., D. R. Piperno, M. B. Bush, M. R. Silman, Richards, P. W. 1996. The Tropical Rain Forest. 2nd ed. Cam-
A. R. Zimmerman, M. F. Raczka, and L. C. Lobato. bridge University Press.
2012. “Sparse pre-Columbian human habitation in Roosevelt, A. C. 2013. “The Amazon and the Anthropo-
western Amazonia.” Science 336: 1429–1431. cene: 13,000 years of human influence in a tropical
Metcalf, J. L., C. Turney, R. Barnett, F. Martin, S. C. Bray, rainforest.” Anthropocene 4: 69–87.
J. T. Vilstrup, L. Orlando, R. Salas-Gismondi, D. Lo- Rozas-Davila, A., B. G. Valencia, and M. B. Bush. 2016.
ponte, M. Medina, et al. 2016. “Synergistic roles of cli- “The functional extinction of Andean megafauna.”
mate warming and human occupation in Patagonian Ecology 97: 2533–2539.
megafaunal extinctions during the Last Deglaciation.” Rull, V. 2009. “Microrefugia.” Journal of Biogeography 36:
Science Advances 2: e1501682. 481–484.
Mosblech, N. A., M. B. Bush, W. D. Gosling, D. Hodell, Rull, V., and E. Montoya. 2014. “Mauritia flexuosa palm
L. Thomas, P. van Calsteren, A. Correa-Metrio, B. G. swamp communities: Natural or human-made? A pal-
Valencia, J. Curtis, and R. van Woesik. 2012. “North ynological study of the Gran Sabana region (northern
Atlantic forcing of Amazonian precipitation during South America) within a neotropical context.” Quater-
the last ice age.” Nature Geoscience 5: 817–820. nary Science Reviews 99: 17–33.
Mosblech, N. A., M. B. Bush, and R. Van Woesik. 2011. Salati, E., and P. B. Vose. 1984. “Amazon basin: A system
“On metapopulations and microrefugia: Palaeoeco- in equilibrium.” Science 225: 129–138.
logical insights.” Journal of Biogeography 38: 419–429. Sato, H., and S. A. Cowling. 2017. “Glacial Amazonia at
Nepstad, D., G. Carvalho, A. C. Barros, A. Alencar, J. P. the canopy-scale: Using a biophysical model to under-
Capobianco, J. Bishop, P. Moutinho, P. Lefebvre, U. L. stand forest robustness.” Quaternary Science Reviews 171:
Silva, and E. Prins. 2001. “Road paving, fire regime 38–47.
feedbacks, and the future of Amazon forests.” Forest Schemske, D. W., G. G. Mittelbach, H. V. Cornell, J. M.
Ecology and Management 154: 395–407. Sobel, and K. Roy. 2009. “Is there a latitudinal gradi-
NGRIP. 2004. “High-resolution record of Northern ent in the importance of biotic interactions?” Annual
Hemisphere climate extending into the last intergla- Review of Ecology, Evolution, and Systematics 40: 245–269.
cial period.” Nature 431: 147–151. Seltzer, G. O., S. Cross, P. Baker, R. Dunbar, and S.
Núñez, L., M. Grosjean, and I. Cartajena. 2001. “Human Fritz. 1998. “High-resolution seismic reflection pro-
dimensions of late Pleistocene/Holocene arid events files from Lake Titicaca, Peru/Bolivia: Evidence for
in southern South America.” In Interhemispheric Climate Holocene aridity in the tropical Andes.” Geology 26:
Linkages, ed. M. Vera, 105–117. Academic Press. 167–170.
Overpeck, J. T., T. I. Webb, and I. C. Prentice. 1985. Silman, M. R. 2011. “Plant species diversity in Amazo-
“Quantitative interpretation of fossil pollen spectra: nia.” In Tropical Rainforests and Climate Change, 2nd ed., ed.
Dissimilarity coefficients and the method of modern M. B. Bush, J. R. Flenley, and W. D. Gosling, 269–288.
analogs.” Quaternary Research 23: 87–708. Praxis.
Peres, C. A., T. Emilio, J. Schietti, S. J. Desmoulière, and Solar, R. R. d. C., J. Barlow, J. Ferreira, E. Berenguer, A.
T. Levi. 2016. “Dispersal limitation induces long-term C. Lees, J. R. Thomson, J. Louzada, M. Maués, N. G.
biomass collapse in overhunted Amazonian forests.” Moura, and V. H. Oliveira. 2015. “How pervasive is
Proceedings of the National Academy of Sciences 113: 892–897. biotic homogenization in human-modified tropical
Pielou, E. C. 1979. Biogeography. Wiley. forest landscapes?” Ecology Letters 18: 1108–1118.
Piperno, D. R., C. McMichael, and M. B. Bush. 2015. Stríkis, N. M., F. W. Cruz, E. A. Barreto, F. Naughton, M.
“Amazonia and the Anthropocene: What was the spa- Vuille, H. Cheng, A. H. Voelker, H. Zhang, I. Karmann,
tial extent and intensity of human landscape modifi- and R. L. Edwards. 2018. “South American monsoon
cation in the Amazon Basin at the end of prehistory?” response to iceberg discharge in the North Atlantic.”
Holocene 25: 1588–1597. Proceedings of the National Academy of Sciences: 201717784.
Pitman, N. C. A., J. W. Terborgh, M. R. Silman, P. V. ter Steege, H., N. C. A. Pitman, D. Sabatier, C. Baraloto,
Nunez, D. A. Neill, C. E. Cerón, W. E. Palacios, and M. R. P. Salomão, J. E. Guevara, O. L. Phillips, et al. 2013.
Aulestia. 2001. “Dominance and distribution of tree “Hyperdominance in the Amazonian tree flora.” Sci-
species in upper Amazonian terra firme forests.” Ecol- ence 342: 1243092.
ogy 82: 2101–2117. Terborgh, J., L. Lopez, P. Nuñez, M. Rao, G. Shahabud-
Raczka, M. F., P. E. De Oliveira, M. B. Bush, and C. H. din, G. Orihuela, M. Riveros, R. Ascanio, G. H. Adler,
McMichael. 2013. “Two paleoecological histories T. D. Lambert, and L. Balbas. 2001. “Ecological melt-
spanning the period of human settlement in south- down in predator-free forest fragments.” Science 5548:
eastern Brazil.” Journal of Quaternary Science 28: 144–151. 1923–1925.
A N E OT ROPIC A L PE R SPE C T I V E ON PA ST H U M A N- C L I M AT E I N T E R AC T IONS 153
Toral-Granda, M. V., C. E. Causton, H. Jäger, M. True- Van ‘t Veer, R., and H. Hooghiemstra. 2000. “Montane
man, J. C. Izurieta, E. Araújo, M. Cruz, K. K. Zander, A. forest evolution during the last 650,000 yr in Colom-
Izurieta, and S. T. Garnett. 2017. “Alien species path- bia: A multivariate approach based on pollen record
ways to the Galapagos Islands, Ecuador.” PLOS One 12: Funza-I.” Journal of Quaternary Science 15: 329–346.
e0184379. Wang, X., R. L. Edwards, A. S. Auler, H. Cheng, X. Kong,
Valencia, B. G., F. Matthews-Bird, D. H. Urrego, J. J. Wil- Y. Wang, F. da Cruz, J. A. Dorale, and H.-W. Chiang.
liams, W. D. Gosling, and M. Bush. 2016. “Andean 2017. “Hydroclimate changes across the Amazon low-
microrefugia: Testing the Holocene to predict the An- lands over the past 45,000 years.” Nature 541: 204–207.
thropocene.” New Phytologist 212: 510–522.
This page intentionally left blank
PA RT I V
W H Y MODE L SPE C IE S A N D
V E GE TATION DIST R I BU TIONS?
157
158 W H AT DOE S THE FUTUR E HOL D?
cies responses. Plant functional types— survive and reproduce (the “fundamental
for example, evergreen needleleaf forest, niche”), and few directly address competi-
deciduous shrub land, C3 grassland—are tive interactions or other limitations that
determined by availability of water, spe- create the realized niche of a species (Pear-
cies tolerances, temperature, and other man et al. 2008).
environmental factors. These structural Among the earliest of such models is
and functional attributes can be simulated BioClim (Nix 1986), a bioclimatic enve-
without resolving individual species. Many lope approach that uses simple correlative
ecosystem service changes can be approxi- methods to assess the range of conditions
mated by understanding vegetation and suitable for a species. Other, more recent
plant functional type responses to climate additions to the species modeling quiver
change, thus providing further insights into include tools with more sophisticated statis-
climate-driven change. tical treatment of species-climate relation-
Therefore, species and vegetation mod- ships. These include modeling tools such as
eling together can provide a more complete GARP (genetic algorithm for rule set pro-
view of biotic impacts that can help address duction) and MaxEnt (a machine learning
specific climate change questions. Both model based on maximum entropy theory),
types of models are useful in impact studies and statistical tools such as GLM (general-
to understand the sensitivity of biodiversity ized linear modeling), GDM (generalized
to changes in climate, for example, by pro- dissimilarity modeling), GAM (generalized
viding insights into changes in species lo- additive modeling), and others. MaxEnt is
cation and into changes in ecosystems and the most widely used SDM and has per-
their functioning. Modeling results can be formed relatively well in comparative tests
incorporated into land use and conserva- (Elith et al. 2011).
tion planning tools, such as Marxan, to aid
in local or regional adaptation planning.
E NSE M BL E MODE L I NG TO DE A L W IT H
U NC E RTA I N T Y
SPE C IE S MODE LS
Dealing with uncertainty is a major issue
There are a wide variety of species model- in SDM applications. Tests have shown that
ing tools available to researchers and im- SDM uncertainty may be greater than gen-
pact practitioners. These models are known eral circulation model (GCM) uncertainty
as species distribution models (SDMs) or for future climate scenarios. Sources of
ecological niche models, but they are also uncertainty include incomplete or biased
referred to as bioclimatic envelope models. species observation data and variation in
All of them simulate species distributions number or type of climate variables used in
under current or changed climates (hence model parameterization.
“species distribution models”), and species One technique that has been shown to
distributions are determined by the fun- reduce uncertainty in biodiversity impact
damental climatic tolerances, or niche of studies is ensemble modeling. Ensemble ap-
species (hence “ecological niche models” proaches use a large number of combina-
[ENM]). The term SDM is used in this chap- tions of GCMs and SDM algorithms to as-
ter, but ENM is generally synonymous. sess the impacts of climate change (Araújo
All SDMs are based on the assumption and New 2007; Figure 12.1). To achieve a
that current distributions of species provide sound result, it is important to drive each
information about the climatic conditions SDM with each individual GCM, rather than
under which the species’ populations can averaging the GCM outputs and using that
MODEL ING SPECIE S A ND V EGETATION DIST R IBUTION 159
Data availability and sampling size: A large pro- efforts should include systematic surveys
portion of the world’s biodiversity has along the entire species range distribu-
not been correctly identified, increasing tions to capture the full environmental
uncertainty in model outputs (Cayuela gradient. Absence data are not always
et al. 2009). Limited availability of spe- available, but they are highly desirable
cies occurrence records also affects model for better describing the range of condi-
performance and output applications tions suitable for a species. However, al-
(Cayuela et al. 2009). In some cases data gorithms exist that can fit models by gen-
paucity occurs for rare or endemic species erating pseudoabsences or comparing the
as a result of sampling difficulties; in oth- presence distribution to the background
ers, data are not yet electronically avail- environment (Jarnevich et al. 2015).
able (Graham et al. 2004). In general, a Threshold selection: Many SDM studies
small sample size (<30 occurrence points) provide outputs in binary formats (0 or
may affect model performance due to 1 values) depicting suitable (potentially
poor representation of the environmental present) or unsuitable (potentially ab-
complexity of the species range distribu- sent) areas for a species. Binary outputs
tion and a limited capacity to detect out- are estimated from suitability maps of
lier values (Wisz et al. 2008). There is no continuous values using a threshold value
agreement on minimum sampling size. to define presence and absence (Bean et
Some authors suggest that a sampling al. 2012). Many threshold selection ap-
size of fewer than 70 observations de- proaches exist and have been evaluated
creases model performance (Kadmon et under biased and limited occurrence
al. 2003), whereas others propose using data (Liu et al. 2005; Bean et al. 2012);
no fewer than 30–40 species occurrences however, the selection of an appropriate
(Wisz et al. 2008). threshold is finally case specific and de-
Data quality (time and space bias): Many stud- pends on the objective of the study (Liu
ies rely on herbaria collections and natu- et al. 2005).
ral history data sets to obtain observation Finally, some SDM methods require
records; nonetheless, in many cases spe- considerable data preparation, model pa-
cies occurrence points are time and space rameterization, and evaluation to obtain
biased with reference to the real species accurate results. Tools have been devel-
range distribution (Bean et al. 2012). For oped to simplify their implementation;
example, occurrence points cover a par- for example, the SDM toolbox (Brown
ticular area of the total range or species, 2014) and ensemble modeling packages
or sampling covers particular periods of (Warren et al. 2010; Ranjitkar et al. 2014)
their life cycle (i.e., reproductive or dis- facilitate data processing and intermodel
persal) with smaller range distributions comparisons (Thuiller et al. 2014).
(Cayuela et al. 2009). Ideally, sampling
to drive the SDMs. This is because change then subsequently the SDM outputs are
in climate parameters such as temperature combined to create an ensemble.
and precipitation are not independent in a Ensemble approaches are most suitable
specific GCM simulation. The combined ef- when there is no evidence that any one
fects of multiple climate variables must be model performs better than others. How-
maintained when driving the SDMs, and ever, model intercomparison has shown
160 W H AT DOE S THE FUTUR E HOL D?
Figure 12.1. Different options to analyze the distribution discriminate sets of input variables would
range of a species under future climate as a result of an be used.
ensemble modeling approach (based on artificial data
for Africa and hypothetical models). (A) results from
each bioclimatic model output are represented by each
line, (B) mapping areas suitable for at least one model W H AT H AV E W E L E A R N E D
(larger polygon), all models (smaller polygon) and areas F ROM SDMS I N T H E CON T E XT
where at least half of the models indicate suitable areas OF GLOBA L C H A NGE ?
(medium polygon), (C) map indicating the number of
models simulating suitable areas for a species, and (D)
the likelihood of species presence in the future based on Species distribution models have provided
a probability density function. (From Araújo and New a wide range of projections of future range
2007.) shifts under global change. Maps of poten-
tial species distributions have been pro-
that MaxEnt and GDM perform better than duced for plants, vertebrates, invertebrates,
other models in standard species modeling and freshwater and marine species. SDM
tasks (Elith et al. 2006). Where data suggest projections in the terrestrial realm have
that some models have stronger perfor- generally emphasized poleward and up-
mance, it is preferable to limit the ensemble ward range shifts—the predicted finger-
to those stronger-performing models. This print of climate change on species distri-
is particularly true if using all available butions in a warming world. Insights from
models means that default settings and in- SDMs have also highlighted counterintuitive
MODEL ING SPECIE S A ND V EGETATION DIST R IBUTION 161
range dynamics. In Australia, an analysis of repositories, especially the Global Biodiver-
464 bird species showed equatorward and sity Information Facility (GBIF). As the data
multidirectional shifts in addition to pole- on species, global circulation models, and
ward shifts (VanDerWal et al. 2013). Wisz new algorithms for SDMs are growing, so
et al. (2015) described a potential intercon- are the multiple potential projections that
nection of marine fish species through the can be derived.
Arctic. Their simulation indicated that 41 SDM climate change projections face
and 44 new species might enter the Pacific limitations. Dispersal capacity is not ac-
and the Atlantic, respectively, predomi- counted for, yet varies widely among spe-
nantly east-west range expansions due to cies and is critical in determining species’
the breakdown of a northern (cold) thermal ability to occupy newly suitable range. To
barrier. Thus, SDMs have helped identify account for dispersal effects, studies using
potential novel species assemblages. SDM projections have often used no-dis-
Extinction risk from climate change has persal and full-dispersal scenarios. That is,
been estimated using suites of SDM simula- scenarios assuming either that the species is
tions representing multiple taxa and mul- not able to disperse to new suitable climate
tiple geographic regions (Thomas et al. (no dispersal) or that the species is able to
2004; see Midgley and Hannah in this vol- colonize any new suitable climate regard-
ume). These estimates show major extinc- less of the distance to current suitable habi-
tion risk vulnerability associated with cli- tat (full dispersal). These two contrasting
mate change and have been influential in scenarios can provide a bounding box to
national and international climate change gauge the importance of species dispersal
policy discussions. Initial estimates that on range shifts. More sophisticated meth-
18 percent–34 percent of all species (mid- ods have been proposed in which SDM suit-
range) might be at risk of extinction due ability maps are produced at different time
to climate change have been criticized on steps, and are combined with a dispersal
methodological grounds. However, subse- algorithms (Engler and Guisan 2009; Midg-
quent analyses incorporating the improved ley et al. 2010). Approaches incorporating
methods have largely confirmed these ball- dispersal have shown the importance of
park figures, both using SDMs (Malcolm et assembling climatic requirements to other
al. 2006) and non-SDM modeling (Sinervo species features to derive realistic projec-
et al. 2010). tions of species distributions under global
Conservation outcomes have been ex- change. Recent approaches also explicitly
plored using SDMs, including the effective- consider the effects of land-use change on
ness of protected areas and protected areas the output of SDMs (Beltrán et al. 2014).
networks. Climate-driven range shifts have New SDM approaches derive metrics re-
been found to cause some species to move lated to species exposure to climate change.
out of reserves, effectively reducing species Several metrics can be used to characterize
representation in protected areas (Araújo different features of the effects of climate
2011). Additions of new protected areas change to species distribution and conser-
have been found to be useful in counteract- vation (Figure 12.2): area metrics, popu-
ing this effect, with area increases of as lit- lation metrics, and velocity metrics (see
tle as 5 percent–10 percent in national pro- Serra-Diaz et al. 2014).
tected areas networks restoring the losses
due to climate change (Hannah et al. 2007).
Area Metrics
Large-scale analyses of biodiversity have
multiplied in response to improving species Area metrics are calculated using the rela-
occurrence data in museum and global data tionship between current and future areas
162 W H AT DOE S THE FUTUR E HOL D?
Figure 12.2. The community weighted mean (CWM) of projected suitable climate. Net area
trait values, of 127 old-growth forest plots in northern change (also called species range change) is
Costa Rica, are ordered across two dimensions resulting
from a principal component analysis (PCA). PCA dimen-
by far the most commonly used, in which
sions reflect climatic features related to precipitation: (i) the amount of future suitable habitat is
higher/lower values in PCA Axis 1 indicate higher pre- compared to the present suitable habitat.
cipitation seasonality/higher annual precipitation, and This metric reflects potential shrinkage,
(ii) higher or lower values in PCA Axis 2 reflect higher maintenance, or expansion of a given spe-
seasonal precipitation and precipitation during the wet-
test month/higher annual temperature and temperature
cies under climate change. Less commonly
seasonality. CWM trait values were calculated for leaf applied but useful metrics from landscape
area (LA), specific leaf area (SLA), leaf dry-matter con- ecology can be used to understand poten-
tent (LDMC), leaf nitrogen (N) and phosphorus (P) con- tial shifts in the fragmentation of suitable
tent, and wood basic specific gravity (WSG). No-filled areas.
and gray symbols represent forest plots under current
and future climate, respectively. Black and gray contour
lines represent the probability density function of for- Population-Level Metrics
est plots across the PCA dimensions under current and
future climate respectively. Arrows in (A) show the re- Population-level metrics assess the expo-
lationship between each trait across each PCA dimen- sure to climate change of currently known
sion. Arrows in (B, C, D) indicate shifts in CWM values
induced by climate change for each plot by forest type:
populations of a species. Indexes include
foothill forests (B) and two different lowland forests (C, the number of populations with increasing
D). (From Chain-Guadarrama et al. 2017.)
MODEL ING SPECIE S A ND V EGETATION DIST R IBUTION 163
or decreasing suitability or the number of has attracted critiques. The relative ease of
populations for which suitability is lost. use of the algorithms and the increasing
data availability on species occurrence have
boosted its use, often without full consid-
Velocity Metrics
eration of the array of initial necessary
Velocity metrics are a set of metrics that quality control in data or with poor un-
aim at quantifying the pace of suitability derstanding of model assumptions (Araújo
changes with reference to species sensitiv- and Peterson 2012). One of the most dis-
ity. These metrics are reported in units of cussed assumptions of this modeling ap-
space per units of time (e.g., km/decade, m/ proach is that the current distribution of
year). Bioclimatic velocity is a local index the species is in equilibrium with climate.
that informs, for each grid cell, the velocity We know, however, that species are still
needed to encounter the same probability responding to climatic rebound from the
of species occurrence within a local neigh- last glacial period and that the long-term
borhood distance (Serra-Diaz et al. 2014; effects of land-use change may not yet be
Figure 12.3). This metric is similar to ve- fully manifested in species’ current distri-
locity of climate change, except that prob-
ability of occurrence is substituted for cli-
Figure 12.3. Maps show the averaged bioclimatic veloc-
matic similarity. Thus, bioclimatic velocity
ity of climate change for endemic tree species in Cali-
integrates both climate aspects and species- fornia. The bioclimatic velocity indicates a measure of
specific sensitivities to climate. This metric the changes in climatic suitability (toward increased or
has been shown to improve the estimates of decreased suitability) for a species within its potential
range shift (Comte and Grenouillet 2015). suitable range and was estimated as the ratio of temporal
to spatial gradients (magnitude of change over time and
The biotic velocity (Ordoñez and Williams
space, respectively) under climate change. The velocity
2013) is an analogue-climate based met- was averaged for two GCMs. The inset figure shows that
ric that measures the distance needed by a different velocities can occur within short distances.
population to reach its nearest suitable site. PIBA = Pinus balfouriana, PICO = Pinus coulteri, PIMU = Pinus
The use of SDM to provide estimates of muricata, PISA = Pinus sabiniana, QUDO = Quercus douglasii,
QUEN = Quercus engelmannii, QULO = Quercus lobata, QUWI =
species range change under climate change Quercus wislizeni. (From Serra-Diaz et al. 2014.)
164 W H AT DOE S THE FUTUR E HOL D?
Trait-based biogeography, or the func- pacts of climate change on the future of biodiversity.”
tional component of biodiversity, is expected Ecology Letters 15 (4): 365–377. https://doi.org/10.1111/
j.1461-0248.2011.01736.x.
to bridge the gap between changes in spe- Beltrán, Bray J., Janet Franklin, Alexandra D. Syphard,
cies distribution and biogeochemical cycles Helen M. Regan, Lorraine E. Flint, and Alan L. Flint.
simulated by DGVMs that drive ecosystem 2014. “Effects of climate change and urban develop-
functions and services (Violle et al. 2014). ment on the distribution and conservation of vegeta-
tion in a Mediterranean type ecosystem.” International
Journal of Geographical Information Science 28 (8): 1561–1589.
https://doi.org/10.1080/13658816.2013.846472.
CONC LUSIONS Buckley, Lauren B., and Joel G. Kingsolver. 2012. “Func-
tional and phylogenetic approaches to forecasting spe-
SDM tools have been extensively used to cies’ responses to climate change.” Annual Review of Ecol-
assess potential impacts of land use, cli- ogy, Evolution, and Systematics 43 (1): 205–226. https://doi
.org/10.1146/annurev-ecolsys-110411-160516.
mate change, and other disturbances on Buckley, Lauren B., Stephanie A. Waaser, Heidi J. Ma-
biodiversity to support conservation plan- cLean, and Richard Fox. 2011. “Does including physi-
ning for adaptation to global change. Many ology improve species distribution model predictions
efforts have been based on correlative ap- of responses to recent climate change?” Ecology 92 (12):
proaches modeling the equilibrium distri- 2214–2221.
Chain-Guadarrama, A., P. Imbach, S. Vilchez-Mendoza,
bution of individual species under climate. L. A. Vierling, and B. Finegan. 2017. “Potential tra-
Ongoing efforts for improved modeling jectories of old-growth neotropical forest functional
aim at developing mechanistic approaches composition under climate change.” Ecography 41.
under dynamic environmental conditions https://doi.org/10.1111/ecog.02637.
that account for populations, multiple dis- Comte, Lise, and Gaël Grenouillet. 2015. “Distribution
shifts of freshwater fish under a variable climate:
turbances, and species functional-trait re- Comparing climatic, bioclimatic and biotic veloci-
sponses to environmental gradients. Even- ties.” Diversity and Distributions 21 (9): 1014–1026. https://
tually, methods and tools will evolve into doi.org/10.1111/ddi.12346.
a continuum of modeling approaches from Elith, Jane, Steven J. Phillips, Trevor Hastie, Miroslav
species to macroscale vegetation patterns Dudík, Yung En Chee, and Colin J. Yates. 2011. “A sta-
tistical explanation of MaxEnt for ecologists.” Diversity
that improve our understanding of global and Distributions 17 (1): 43–57. https://doi.org/10.1111/j
change threats to biodiversity and potential .1472-4642.2010.00725.x.
impacts to society in order to support plan- Elith, J., C. H. Graham, R. P. Anderson, M. Dudík, S. Fer-
ning and implementation of coordinated rier, A. Guisan, et al. 2006. “Novel methods improve
conservation efforts across scales. prediction of species’ distributions from occurrence
data.” Ecography: 129–151.
Engler, Robin, and Antoine Guisan. 2009. “MigClim:
Predicting plant distribution and dispersal in a chang-
R E FE R E NCE S ing climate.” Diversity and Distributions 15 (4): 590–601.
https://doi.org/10.1111/j.1472-4642.2009.00566.x.
Araújo, Miguel B., and Mark New. 2007. “Ensemble fore- Higgins, Paul A. T., and John Harte. 2006. “Biophysi-
casting of species distributions.” Trends in Ecology and Evo- cal and biogeochemical responses to climate change
lution 22 (1): 42–47. https://doi.org/10.1016/j.tree.2006 depend on dispersal and migration.” BioScience 56 (5):
.09.010. 407–417. https://doi.org/10.1641/0006-3568(2006
Araújo, Miguel B., Diogo Alagador, Mar Cabeza, David )056[0407:BABRTC]2.0.CO;2.
Nogués-Bravo, and Wilfried Thuiller. 2011. “Climate Kearney, Michael, Richard Shine, and Warren P. Porter.
change threatens european conservation areas.” Ecol- 2009. “The potential for behavioral thermoregula-
ogy Letters 14 (5): 484–492. https://doi.org/10.1111/j tion to buffer ‘cold-blooded’ animals against climate
.1461-0248.2011.01610.x. warming.” Proceedings of the National Academy of Sciences 106
Araújo, Miguel B., and A. Townsend Peterson. 2012. (10): 3835–3840.
“Uses and misuses of bioclimatic envelope model- Keenan, T., J. Maria Serra, F. Lloret, M. Ninyerola, and
ing.” Ecology 93 (7): 1527–1539. https://doi.org/10 S. Sabate. 2011. “Predicting the future of forests in the
.1890/11-1930.1. Mediterranean under climate change, with niche- and
Bellard, Céline, Cleo Bertelsmeier, Paul Leadley, Wil- process-based models: CO2 matters!” Global Change Bi-
fried Thuiller, and Franck Courchamp. 2012. “Im- ology 17 (1): 565–579. https://doi.org/10.1111/j.1365
-2486.2010.02254.x.
MODEL ING SPECIE S A ND V EGETATION DIST R IBUTION 167
Keith, D. A, H. R. Akcakaya, W. Thuiller, G. F. Midgley, Gustafson, and David J. Mladenoff. 2007. “Design,
R. G. Pearson, S. J. Phillips, H. M. Regan, M. B. Araújo, development, and application of LANDIS-II, a spatial
and T. G. Rebelo. 2008. “Predicting extinction risks landscape simulation model with flexible temporal
under climate change: Coupling stochastic population and spatial resolution.” Ecological Modelling 201 (3–4):
models with dynamic bioclimatic habitat models.” 409–419. https://doi.org/10.1016/j.ecolmodel.2006
Biology Letters 4 (5): 560–563. https://doi.org/10.1098/ .10.009.
rsbl.2008.0049. Serra-Diaz, Josep M., Janet Franklin, Miquel Ninyerola,
Levis, Samuel. 2010. “Modeling vegetation and land use Frank W. Davis, Alexandra D. Syphard, Helen M. Re-
in models of the Earth system.” Wiley Interdisciplinary Re- gan, and Makihiko Ikegami. 2014. “Bioclimatic veloc-
views: Climate Change 1 (6): 840–856. https://doi.org/10 ity: The pace of species exposure to climate change.”
.1002/wcc.83. Diversity and Distributions 20 (2): 169–180. https://doi
Merow, Cory, Johan P. Dahlgren, C. Jessica E. Metcalf, .org/10.1111/ddi.12131.
Dylan Z. Childs, Margaret E. K. Evans, Eelke Jongejans, Shugart, H. H. 2002. “Forest gap models.” In Encyclopedia
Sydne Record, Mark Rees, Roberto Salguero-Gómez, of Global Environmental Change, ed. Harold A. Mooney and
and Sean M. McMahon. 2014. “Advancing population Josep G. Canadell, 2: 316–323. John Wiley & Sons.
ecology with integral projection models: A practi- Svenning, Jens-Christian, and Brody Sandel. 2013. “Dis-
cal guide.” Methods in Ecology and Evolution 5 (2): 99–110. equilibrium vegetation dynamics under future climate
https://doi.org/10.1111/2041-210X.12146. change.” American Journal of Botany 100 (7): 1266–1286.
Midgley, G. F., I. D. Davies, C. H. Albert, R. Altwegg, L. Swab, Rebecca M., Helen M. Regan, David A. Keith,
Hannah, G. O. Hughes, and W. Thuiller. 2010. “Bio- Tracey J. Regan, and Mark K. J. Ooi. 2012. “Niche
Move: An integrated platform simulating the dynamic models tell half the story: Spatial context and life-
response of species to environmental change.” Ecogra- history traits influence species responses to global
phy 33 (3): 612–616. change.” Journal of Biogeography 39: 1266–1277. https://
Mladenoff, David J. 2004. “LANDIS and forest land- doi.org/10.1111/j.1365-2699.2012.02690.x.
scape models.” Ecological Modelling 180 (January): 7–19. VanDerWal, Jeremy, Helen T. Murphy, Alex S. Kutt, Gen-
https://doi.org/10.1016/j.ecolmodel.2004.03.016. evieve C. Perkins, Brooke L. Bateman, Justin J. Perry,
Nix, H. A. 1986. “A biogeographic analysis of Australian and April E. Reside. 2013. “Focus on poleward shifts
elapid snakes.” In Atlas of Elapid Snakes of Australia, ed. R. in species’ distribution underestimates the fingerprint
Longmore, 4–15. Australian Government Publishing of climate change.” Nature Climate Change 3 (3): 239–243.
Service. Violle, Cyrille, Peter B. Reich, Stephen W. Pacala, Brian
Ordoñez, Alejandro, and John W. Williams. 2013. “Cli- J. Enquist, and Jens Kattge. 2014. “The emergence
matic and biotic velocities for woody taxa distribu- and promise of functional biogeography.” Proceedings
tions over the last 16,000 years in eastern North of the National Academy of Sciences 111 (38): 13690–13696.
America.” Ecology Letters 16 (6): 773–781. https://doi https://doi.org/10.1073/pnas.1415442111.
.org/10.1111/ele.12110. Wisz, M. S., O. Broennimann, P. Gronkjaer, P. R. Moller,
Parmesan, Camille. 1996. “Climate and species’ range.” S. M. Olsen, D. Swingedouw, R. B. Hedeholm, E. E.
Nature. https://doi.org/10.1038/382765a0. Nielsen, A. Guisan, and L. Pellissier. 2015. “Arctic
Pearman, Peter B., Antoine Guisan, Olivier Broenni- warming will promote Atlantic-Pacific fish inter-
mann, and Christophe F. Randin. 2008. “Niche dy- change.” Nature Climate Change 5 (3): 261–265.
namics in space and time.” Trends in Ecology and Evolution Wullschleger, Stan D., Howard E. Epstein, Elgene O. Box,
23 (3): 149–158. https://doi.org/10.1016/j.tree.2007 Eugénie S. Euskirchen, Santonu Goswami, Colleen M.
.11.005. Iversen, Jens Kattge, Richard J. Norby, Peter M. Van
Peng, Changhui. 2000. “From static biogeographical Bodegom, and Xiaofeng Xu. 2014. “Plant functional
model to dynamic global vegetation model: A global types in earth system models: Past experiences and fu-
perspective on modelling vegetation dynamics.” Eco- ture directions for application of dynamic vegetation
logical Modelling 135 (1): 33–54. models in high-latitude ecosystems.” Annals of Botany
Scheiter, Simon, and Steven Higgins. 2009. “Impacts of 114 (1): 1–16.
climate change on the vegetation of Africa: An adap- Yang, Yanzheng, Qiuan Zhu, Changhui Peng, Han Wang,
tive dynamic vegetation modelling approach.” Global and Huai Chen. 2015. “From plant functional types
Change Biology 15 (9): 2224–2246. https://doi.org/10 to plant functional traits: A new paradigm in mod-
.1111/j.1365-2486.2008.01838.x. elling global vegetation dynamics.” Progress in Physical
Scheller, Robert M., James B. Domingo, Brian R. Stur- Geography 39 (4): 514–535. https://doi.org/10.1177/
tevant, Jeremy S. Williams, Arnold Rudy, Eric J. 0309133315582018.
CHAPTER THIRTEEN I N T RODUCTION
168
CLIM ATE CH A NGE A ND M A R INE BIODIV ER SIT Y 169
proximately 30 percent since preindustrial within their natural environment (Pörtner
levels (Doney et al. 2009), while oxygen 2001). Organism physiology performs op-
concentration in the open ocean decreased timally at a certain temperature range; per-
by 3–5 μmol kg-1 per decade (IPCC 2014). formance decreases when environmental
Global oxygen content has decreased by temperature increases or decreases from
more than 2 percent since 1960 (Schmidtko that range with upper and lower tempera-
et al. 2017) and is projected to continue to ture limits at which body functions that are
decrease in the twenty-first century (Bopp necessary for survival are halted. Such a re-
et al. 2013) as a result of increase in stratifi- lationship between temperature and physi-
cation and decrease in mix-layer depth. ological performance of organisms applies
Marine biodiversity responds to chang- to a wide range of organismal groups, from
ing temperature and other ocean conditions plants to vertebrates. For marine water-
manifested through physiology, population breathing ectotherms, theories suggest that
dynamics, distributions, and phenology the characteristic window of optimal physi-
(Poloczanska et al. 2013; Pörtner et al. 2014). ological performance is largely determined
These responses to ocean and atmospheric by the relationship between oxygen avail-
changes have been projected to lead to al- ability for metabolism and growth, and the
tered patterns of species richness (Cheung level of available oxygen for such purposes
et al. 2009), changes in community struc- is referred to as aerobic scope (Pauly 2010a;
ture (MacNeil et al. 2010), ecosystem func- Pörtner et al. 2017).
tions (Petchey et al. 1999), and consequen- An organism’s temperature tolerance
tial changes in marine goods and services limits are dependent on its organization
(Cheung and Sumaila 2008; Sumaila et al. complexity, the temperature stability that
2011; Madin et al. 2012). the species is adapted to, its life stage, and
This chapter aims to provide an overview the existence of other physiological stress-
of marine biodiversity under climate change ors (Figure 13.1). For example, some polar
and ocean acidification. First, we describe species have adapted to stable cold polar
overarching theories and hypotheses of how environments, leading to fewer red blood
changing ocean conditions affect marine cells, oxygen-binding proteins (Nikinmaa
organisms. Second, we discuss observed 2002), and enzymes that are especially
and projected effects of climate change and sensitive to temperature. Such physiologi-
ocean acidification on major marine taxa. cal adaptations render these polar species
We then examine the potential interactions particularly sensitive to ocean warming.
of climate impacts on marine biodiversity Larval and spawning fishes also have higher
with other non-CO2-related human stress- temperature sensitivity. Moreover, aerobic
ors. Finally, we discuss the key challenges scope (and temperature tolerance) decreases
and opportunities for marine biodiversity with limitation in food supply, increase in
conservation under climate change. Topics acidity, or decrease in oxygen level (Pörtner
related to ocean acidification and coral reefs et al. 2017). Evaluating the effect of rapidly
under climate change are covered in sepa- changing climate on species’ survival and
rate chapters (Chapters 5 and 14). distributions therefore requires linking the
geographic pattern of climatic change to
the physiological sensitivity of study organ-
M E C H A N ISMS OF C L IM AT E C H A NGE isms (Deutsch et al. 2008).
E F F E CTS ON M A R I N E BIODI V E R SIT Y Organisms can sustain physiological
processes in low-oxygen areas only un-
Marine organisms are adapted to and de- der certain environmental and biological
pendent on the maintenance of a character- conditions. Low temperatures reduce en-
istic window of environmental conditions ergetic demands and the need for oxygen.
170 W H AT DOE S THE FUTUR E HOL D?
Figure 13.1. Relationship between temperature and et al. 2013; Poloczanska et al. 2013) or a ver-
physiological performance of marine water-breathing
ectotherms. (A) Temperature tolerance decreases with tical temperature gradient to deeper waters
multiple environmental stressors, and (B) temperature (Dulvy et al. 2008; Pauly 2010b).
sensitivity is dependent on the characteristics of the en- Temperature may also act indirectly on
vironmental the organisms are adapted to. (Redrawn a species’ survival and distribution by in-
from Pörtner and Knust 2007.)
fluencing phenology, dispersal, predation
Organisms with small body size are also pressure, and available food supply. Change
better adapted to low-oxygen areas be- in the distribution and abundance of prey
cause of their morphology (e.g., surface and predators affects growth and mortality
area to volume ratio) and reduced energy of other species. As temperature affects the
demands (Childress and Seibel 1998; Levin rate of egg and larval development, warmer
et al. 2009; Pauly 2010b). As only species temperatures will also decrease the chance
with specialized adaptations can survive, of predation at this phase in the life cycle. In
biodiversity is therefore frequently low in addition, as the duration of the larval stage
an area with low oxygen levels, which may will determine the length of time they are
further become a “dead zone” in extreme subjected to movement by ocean currents,
cases (Levin 2003; Breitburg et al. 2018). increased temperatures will indirectly af-
Understanding organisms’ thermal tol- fect population connectivity, community
erances and the geographic pattern of cli- structure, and regional to global patterns of
mate change helps explain and predict the biodiversity (O’Connor et al. 2007).
survival and distributions of marine organ-
isms under changing climate (Cheung and
Pauly 2017). The theory of optimal foraging R E PONSE S OF M A R I N E BIODI V E R SIT Y
predicts that organisms tend to be distrib- TO C L IM AT E C H A NGE
uted in environments that maximize their
growth and reproduction, thereby provid- Global patterns of species richness of ma-
ing a pathway through which the thermal rine animals are related to habitats (e.g.,
window of an organism may affect its bio- coastal vs. open ocean), temperature, and
geography. Thus, as the ocean warms, ma- historical geographic factors (Tittensor
rine organisms respond by shifting their et al. 2010). Marine animals include zoo-
distributions and/or seasonality to main- plankton, invertebrates, fish, reptiles, birds,
tain themselves in habitats that lie within and mammals, with species richness being
their preferred temperature limits. Shifted highest in coastal and shelf seas. Although
species distribution ranges follow tempera- the richness of coastal species is highest in
ture clines from high to low, reflecting a the western Pacific, that for oceanic species
lateral gradient at the basin scale (Pinsky is highest in regions along mid-latitudinal
CLIM ATE CH A NGE A ND M A R INE BIODIV ER SIT Y 171
zones. As temperature is one of the most Macroinvertebrates and Fishes
important environmental predictors for
The extant number of marine fishes and
marine animals’ species richness, ocean
marcoinvertebrates that have been de-
warming is expected to have large impli-
scribed amount to approximately 15,000
cations for the global pattern of oceanic
species and 200,000 species, respectively
biodiversity. Thus, biological responses to
(Cheung, Pitcher, and Pauly 2005). As de-
ocean-atmospheric changes have been pro-
scribed above, marine invertebrates and
jected to lead to altered future patterns of
fishes, as water-breathing ectotherms, are
species richness (Cheung et al. 2009; Jones
physiologically and ecologically sensitive
and Cheung 2015), resulting in changes in
to changes in ocean properties. Their re-
community structure (MacNeil et al. 2010),
sponses to changes in ocean temperature,
ecosystem function (Petchey et al. 1999),
oxygen level, and acidity can therefore gen-
and marine goods and services (Cheung
erally be predicted from their physiology
and Sumaila 2008; Sumaila et al. 2011; Ma-
of thermal tolerance, oxygen capacity, and
din et al. 2012).
acid-base regulation, as well as changes in
ocean productivity.
Phytoplanktons, Macroalgae, and Seagrasses Overall, climate change and ocean acidi-
fication lead to alteration of population dy-
Ocean warming affects the growth, abun-
namics, shifts in biogeography and season-
dance, distribution, phenology, and com-
ality, and changes in species’ morphologies,
munity structure of phytoplankton. In-
such as body size and behavior. These re-
creases in temperature initially increase
sponses are potentially linked to altered ge-
the metabolic rates, and thus productivity,
netic frequencies (Thomas et al. 2001; Par-
of many phytoplankton species. However,
mesan and Yohe 2003), although evidence
when temperature at a particular location
from Pleistocene glaciations has shown that
exceeds a species’ thermal window, its
species are more likely to exhibit ecological
abundance will decrease, and it will even-
responses to climate change, such as shifts
tually be replaced by species with a higher
in range distributions, than evolutionary
thermal tolerance (Thomas et al. 2012).
responses, through local adaptation (Brad-
Thus, on the basis of the thermal windows
shaw and Holzapfel 2006).
of phytoplankton, their distributions are
Observations and theory have indicated
projected to shift poleward, with a large
that marine fish and invertebrates fre-
decline in tropical diversity (Thomas et al.
quently undergo shifts in distribution in
2012). Also, ocean warming favors smaller
response to changing environmental fac-
species, resulting in a decrease in size struc-
tors, to areas within their physiological
ture of phytoplankton community (Flom-
limits. Shifts are therefore most commonly
baum et al. 2013). The species-specific ther-
toward higher latitudes (Perry et al. 2005;
mal tolerances of macroalgae and seagrasses
Hiddink and ter Hofstede 2008; Doney et
lead to similar types of responses to ocean
al. 2012; Poloczanska et al. 2013), deeper
warming as phytoplankton, such as distri-
waters (Dulvy et al. 2008), and, in general,
bution shifts (Lima et al. 2007). Increases in
following temperature velocity (Pinsky et
CO2 levels appear to have positive effects on
al. 2013). Overall, marine examples of shift-
primary producers as a result of increased
ing distributions due to climate change are
rate of photosynthesis and productivity,
more striking than their terrestrial coun-
although some calcifying algae will be
terparts because of their greater rapidity
vulnerable to ocean acidification, and the
(Edwards and Richardson 2004; Parmesan
indirect trophic impacts from changes in
and Yohe 2003; Cheung et al. 2009). Meta-
primary production are uncertain (Martin
analyses of observed range shift in the past
and Gattuso 2009; Arnold et al. 2012).
172 W H AT DOE S THE FUTUR E HOL D?
decades have shown that zooplanktons and invertebrates and fishes in North American
fishes are moving poleward, on average, shelf seas are in directions consistent with
at rates of tens to hundreds of kilometers gradients of temperature changes (Pinsky et
per decade (Poloczanska et al. 2013) (see al. 2013).
Chapter 5, Figure 5.2A). For example, Beau- Observed changes in the species com-
grand et al. (2009) described northward position of catches from 1970 to 2006 that
movement in calanoid copepod zooplank- are partly attributed to long-term ocean
ton assemblages in the North Atlantic at a warming suggest increasing dominance of
mean rate of up to 23.16 km per year over warmer-water species in subtropical and
48 years. These changes in planktonic com- higher-latitude regions, concurrent with
munities were paralleled by a northward a reduction in the abundance of subtropi-
migration of both commercial and non- cal species in equatorial waters (Cheung
commercial fish species (Brander 2003; et al. 2013a), with implications for fisher-
Beare et al. 2004; Perry et al. 2005). In the ies. Using global fisheries catch data, an in-
European continental shelf, a response to dex called the mean temperature of catch
warming has been demonstrated in the (MTC), computed from the average pre-
abundances of 72 percent of the 50 most ferred temperature of each species reported
common species inhabiting UK waters
(Simpson et al. 2011). These shifts reflect
the influx of warmer-water-adapted marine Figure 13.2. Average mean temperature of catch (MTC)
species to regions with colder waters that and sea-surface temperature (SST) from (upper) non-
tropical and (lower) tropical large marine ecosystems
have been observed elsewhere (Arvedlund
(left panels). The right panels are schematic representation
2009; Fodrie et al. 2010). Analysis of sur- of hypotheses explaining the change in MTC over time.
vey data further shows that range shifts for (From Cheung et al. 2013.)
CLIM ATE CH A NGE A ND M A R INE BIODIV ER SIT Y 173
in fisheries data weighted by their annual phases remain synchronized with climatic
catches, was calculated for all the large ma- alterations. For example, phenophases of
rine ecosystems (LMEs) of the world from the spawning season have been shown to be
1970 to 2006. After accounting for the ef- negatively correlated with mean sea-surface
fects of fishing and large-scale oceano- temperature (SST) the preceding winter for
graphic variability, global MTC increased at 27 species in the North Sea (Greve et al.
a rate of 0.19ºC per decade between 1970 2005), and earlier spring migrations have
and 2006, whereas MTC in nontropical also been noted (Sims et al. 2001; Clarke
areas increased at a rate of 0.23ºC per de- et al. 2003). Phenological responses are
cade (Figure 13.2). In tropical areas, MTC highly taxon or species specific, resulting
increased initially because of the reduction from sensitivity to climatic fluctuations as
in the proportion of subtropical species well as factors such as temperature, light, or
catches but subsequently stabilized as scope food availability (Edwards and Richardson
for further tropicalization of communities 2004). Meta-analysis of observed phono-
became limited (Figure 13.2). logical shifts suggests that seasonal events
Range shifts for fishes and invertebrates of marine species have advanced by an av-
are projected to continue in the twenty- erage of 4.4 days per decade (Poloczanska
first century under climate change (Jones et al. 2013). Altered phenology and timing
and Cheung 2015; Cheung et al. 2009). of development may also lead to altered
Although in temperate climates, local ex- dispersal. For species whose offspring de-
tinctions may be compensated by local velop in the water column, for example, the
invasions as species move into newly suit- duration of the larval stage will determine
able habitat, thereby leading to an overall the length of time that larvae are subject to
change in community structure, tropical movement by ocean currents (O’Connor et
regions may see declines in species richness al. 2007).
as the scope for community tropicalization Both theory and empirical observations
is reached. For example, projections of dis- further support the hypothesis that warm-
tributions for over 800 exploited fishes and ing and reduced oxygen will reduce the
invertebrates using multiple species distri- body size of marine fishes and invertebrates
bution models result in a predicted average (Pauly and Cheung 2018). The preferred
poleward latitudinal range shift of 15.5– minimum oxygen tolerance threshold of an
25.6 km per decade under low and high organism varies across species, body size,
emission scenarios (Representative Concen- and life stage, and is highest for large or-
tration Pathways 2.6 and 8.0, respectively) ganisms. As fish increase in size (weight),
(Jones and Cheung 2015). This rate of shift mass-specific oxygen demand increases
is consistent with observed rates of shift in more rapidly than oxygen supply (Pauly
the twentieth century (Dulvy et al. 2008; 1997). Thus, fish reach a maximum body
Perry et al. 2005). Predicted distribution size when oxygen supply is balanced by
shifts resulted in large-scale changes in pat- oxygen demand. Moreover, the scope for
terns of species richness through species in- aerobic respiration and growth decreases
vasions (occurring in new areas) and local when size increases, with oxygen supply
extinctions (disappearing from previously per unit body weight therefore decreasing.
occurring areas) (Figure 13.3). Hotspots of The decrease in food conversion efficiency
high local invasion are common in high- that this implies, all else being equal, de-
latitude regions, whereas local extinctions creases the biomass production of fish and
are concentrated near the equator. invertebrate populations.
Climate change modifies phenology, Simulation model projections suggest
periodic biological phenomena, of ma- decreases in the maximum body size of
rine fishes and invertebrates so that critical fishes under scenarios of ocean warming,
174 W H AT DOE S THE FUTUR E HOL D?
Plate 1. Observed and simulated future climate change. Observed change in surface temperature 1901–2012 (top). Simu-
lated future change in temperature for two different emissions (radiative forcing) scenarios—RCP 2.6 and RCP 8.5
(middle). Simulated future change in precipitation and model agreement for our RCP 2.6 and RCP 8.5 (bottom). RCP 8.5
approximates a business-as-usual global emissions trajectory, whereas RCP 2.6 approximates a world in which there is
moderate to strong global policy to restrict climate change. (Figure SPM.7; Figure SPM.1 (b) from Climate Change 2014:
Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change.)
Plate 2. Coral bleaching. Corals have bleached in every major reef system on the planet due to climate change. When
regional surface water temperatures exceed a fixed threshold, corals expel their symbiotic zooxanothellae, resulting in
coral bleaching over large areas (top). Bleached corals (bottom) suffer high mortality, but some corals may recover follow-
ing bleaching. (Photographs by Ove Hoegh-Guldberg, Global Change Institute, University of Queensland.)
Plate 3. Bark beetle-killed trees (top). Warming temperatures have moved bark beetles into temperature bands in which
synchronous outbreaks occur, destroying millions of trees across huge areas of North America (bottom; red indicates
area affected by bark beetle outbreak). (Used with permission of National Geographic. Martin Gamache, NGM Staff;
Shelley Sperry. Sources: Canadian Forest Service; Barbara Bentz and Jeanne Paschke, U.S. Forest Service; Aaron McGill,
Alberta Environment and Sustainable Resource Development; Tim Ebata, British Columbia MFLNRO, University of
British Columbia; Brian Aukema, University of Minnesota.)
Plate 4. Predicted impact of warming on the thermal performance of ectotherms in 2100. (A) Impact versus latitude
for insects using thermal performance curves fit to intrinsic population growth rates measured for each species and
for a global model (red line) in which performance curves at each location are interpolated from empirical linear
relationships between seasonality and both warming tolerance and thermal safety margin. (B and C) Results from the
simplified conceptual model are shown globally for insects (B) for which performance data are most complete, and
versus latitude for three additional taxa of terrestrial ectotherms: frogs and toads, lizards, and turtles (C), for which
only warming tolerance was available. On the basis of patterns in warming tolerance, climate change is predicted to
be most deleterious for tropical representatives of all four taxonomic groups. Performance is predicted to increase in
mid- and high latitudes because of the thermal safety margins observed there for insects, and provisionally attributed
to other taxa. Note that blue denotes negative impact on performance, while performance is enhanced by temperature
change. (Redrawn from Deutsch et al. 2008.)
Plate 5. Penguin population responses to climate change. Some penguin populations in Antarctica are declining due
to warming temperatures and declining sea ice. Other penguin populations are increasing as climate change-driven
winds bring more cold water to the surface, resulting in increases in ice cover. (Courtesy of I. Gaffney.)
Plate 6. Projected climate-driven movements for 2,903 vertebrate species. Range movements from species distribution
models are averaged across 10 future climate projections. Arrows represent the direction of modeled movements from
unsuitable climates to suitable climates via routes that avoid human land uses. The sizes of the arrows represent the
number of projected species movements as a proportion of current species richness. The colors of the arrows reflect
the level of agreement in the direction of movement across species and routes. The insets are maps of (A) a high con-
centration of movements through southeastern North America and into the Appalachian Mountains and (B) areas of
movement through the Sierras de Córdoba and into the Andes and the southern Pampas. Lighter blue shading in these
topographical overlays indicates more intensive human activity. (From Lawler et al. 2013.)
Plate 7. Dynamic ocean management for Australian tuna fishery. Southern bluefin tuna (Thunnus maccoyii) (upper left) are
the target of a high-value fishery in southeast Australia. The fishery is managed using a model of tuna response to ocean
temperature that is updated every 2 weeks. This management system provides adaptation to climate change because it
responds to evolving ocean conditions. The modeling process (green arrows) begins with tagging tuna with pop-up
tags that record water temperature and depth (upper right). Based on tuna water temperature preferences, a model is con-
structed of tuna habitat. Model simulations are run every two weeks using sea surface temperature information from
satellites and near-real-time temperature recording at depth (middle). Based on the model simulations, a management
zone is defined that is accessible only to fishing vessels that hold permits for southern bluefin tuna. Fishing vessels use
GPS to ensure that they are respecting the management boundaries. Boundaries could be any shape, but for simplicity
are usually a straight line (lower right). (Figure supplied by CSIRO.)
Plate 8. Conservation planning for climate change. Models of species’ current and future ranges can be used to select
priority areas for new protection that conserve species both in their present ranges and in their likely future range.
Conservation planning software can suggest optimal solutions across hundreds or thousands of species’ present and
future ranges. In this example, 4,000 species in the northern tropical Andes are used to suggest new areas for protec-
tion that will improve representation of species’ future ranges in protected area systems of the region. The results for
the Zonation conservation planning software and Network Flow algorithm (purple areas) are compared. Note the high
level of agreement on priority areas for additional protection. (Courtesy of Patrick R. Roehrdanz.)
CLIM ATE CH A NGE A ND M A R INE BIODIV ER SIT Y 175
jected magnitude of decrease in body size is of nesting (Weishampel, Bagley, and Eh-
consistent with experimental (Forster et al. rhart 2004; Mazaris et al. 2008), whereas
2012; Cheung et al. 2013b) and field obser- increases in extreme weather events, such
vations (Baudron et al. 2014). About half of as flooding, increase nest damage and fur-
this shrinkage is due to change in distribu- ther negatively affect reproductive success
tion and abundance, and the remainder to (Van Houtan and Bass 2007). Using SST as
changes in physiological performance. The a predictor, the distribution of loggerhead
tropical and intermediate latitudinal areas turtles is projected to expand poleward
will be heavily affected, with an average re- in the Atlantic Ocean by the end of the
duction of more than 20 percent. Decreases twenty-first century relative to 1970–1989
in growth and body size are likely to reduce (Witt et al. 2010). In contrast, leatherback
the biomass production of fish populations, turtles are projected to decrease in abun-
and hence fishery catches, and potentially dance because of reduced hatching success
alter trophic interactions. with warming (Saba et al. 2012).
Ocean acidification is also expected to Climate change affects seabird popula-
affect marine animals, although the sen- tions indirectly through changes in the
sitivity varies largely between taxonomic distribution and productivity of prey and
groups. Available studies suggest that sen- directly through impacts on vulnerable
sitivity is high for organisms that form life stages. Seabirds’ foraging grounds and
calcium carbonate exoskeletons or shells their food supply are strongly affected by
(Kroeker et al. 2013; Wittman and Pörtner the distribution and availability of mid-
2013). Also, elevated CO2 levels lead to re- trophic-level prey that are in turn affected
duced tolerance to warming (Wittman and by climate change and resulting changes
Pörtner 2013). Sensitivity to ocean acidifica- in ocean productivity (Hoegh-Guldberg
tion appears to be highest for tropical coral and Bruno 2010; Trathan et al. 2014). For
reefs and mollusks, followed by crustacean example, Cassin’s auklet in the California
and cold-water corals. Their ecological re- Current ecosystem is projected to decline
sponses may become more unpredictable by 11 percent–45 percent by the end of the
when trophic interactions are considered twenty-first century due to ocean warming
(Busch et al. 2013) (see Chapter 14). (~2°C), with consideration of changes in
upwelling intensity (Wolf et al. 2010). Dis-
tributions of some seabirds are also affected
Reptiles, Mammals, and Birds
by changing habitat structure, such as sea
Some species of seabirds, marine mam- ice in the case of the emperor penguin (Je-
mals, and sea turtles are affected by nouvrier et al. 2012). Increased frequency
changes in temperature and other ocean of extreme weather events also affects
conditions. Marine reptiles, including sea penguin populations. For example, storms
turtles, sea snakes, saltwater crocodile, and are a major source of mortality of Magel-
marine iguanas (from the Galapagos Is- lanic penguin (Spheniscus magellanicus) chicks
lands), are ectotherms and thus inherently at Punta Tombo, Argentina, with storm
sensitive to ocean warming. All sea turtles intensity being positively related to chick
have demonstrated poleward distribution mortality. As storm frequency and intensity
shifts that are correlated with ocean warm- are projected to increase in the region un-
ing (Pörtner et al. 2014). As the gender of der climate change, populations of Magel-
sea turtles is also temperature dependent, lanic penguin are therefore expected to be
warming alters their sex ratio toward an negatively affected (Boersma and Rebstock
increased number of females. In addition, 2014). Warming also affects the phenol-
warming increases eggs and hatchling ogy of seabirds, frequently leading to ear-
mortality and influences the phenology lier breeding and the extension of breeding
176 W H AT DOE S THE FUTUR E HOL D?
Haward, Marcus, Julie Davidson, Michael Lockwood, Lima, F. P., P. A. Ribeiro, N. Queiroz, and S. J. Hawkins.
Marc Hockings, Lorne Kriwoken, and Robyn Allchin. 2007. “Do distributional shifts of northern and south-
2013. “Climate change, scenarios and marine bio- ern species of algae match the warming pattern?”
diversity conservation.” Marine Policy 38 (March): Global Change Biology 13: 2592–2604.
438–446. https://doi.org/10.1016/j.marpol.2012.07 MacNeil, M. Aaron, Nicholas A. J. Graham, Joshua E.
.004. Cinner, Nicholas K. Dulvy, Philip A. Loring, Simon
Henson, S. A., J. L. Sarmiento, J. P. Dunne, L. Bopp, I. Jennings, Nicholas V. C. Polunin, Aaron T. Fisk, and
D. Lima, S. C. Doney, J. John, and C. Beaulieu. 2010. Tim R. McClanahan. 2010. “Transitional states in ma-
“Detection of anthropogenic climate change in satel- rine fisheries: Adapting to predicted global change.”
lite records of ocean chlorophyll and productivity.” Philosophical Transactions of the Royal Society of London B: Biologi-
Biogeosciences 7: 621–640. cal Sciences 365 (1558): 3753–3763. https://doi.org/10
Hiddink, J. G., and R. ter Hofstede. 2008. “Climate in- .1098/rstb.2010.0289.
duced increases in species richness of marine fishes.” Madin, Elizabeth M. P., Natalie C. Ban, Zoë A. Doubleday,
Global Change Biology 14 (3): 453–460. https://doi.org/10 Thomas H. Holmes, Gretta T. Pecl, and Franz Smith.
.1111/j.1365-2486.2007.01518.x. 2012. “Socio-economic and management implications
Hoegh-Guldberg, O., and J. F. Bruno. 2010. “The impact of range-shifting species in marine systems.” Global En-
of climate change on the world’s marine ecosystems.” vironmental Change 22 (February): 137–146. https://doi
Science 328 (5985): 1523–1528. .org/10.1016/j.gloenvcha.2011.10.008.
Houtan, K. S. Van, and O. L. Bass. 2007. “Stormy oceans Martin, S., and J. P. Gattuso. 2009. “Response of Medi-
are associated with declines in sea turtle hatching.” terranean coralline algae to ocean acidification and
Current Biology 17 (15): R590–R591. elevated temperature.” Global Change Biology 15: 2089–
Jenouvrier, S., M. Holland, J. Stroeve, C. Barbraud, H. 2100.
Weimerskirch, M. Serreze, and H. Caswell. 2012. “Ef- Mazaris, A. D., A. S. Kallimanis, S. P. Sgardelis, and J.
fects of climate change on an emperor penguin popu- D. Pantis. 2008. “Do long-term changes in sea surface
lation: Analysis of coupled demographic and climate temperature at the breeding areas affect the breed-
models.” Global Change Biology 18 (9): 2756–2770. ing dates and reproduction performance of Mediter-
Jones, Miranda C., and William W. L. Cheung. 2015. ranean loggerhead turtles? Implications for climate
“Multi-model ensemble projections of climate change change.” Journal of Experimental Biology 367 (2): 219–226.
effects on global marine biodiversity.” ICES Journal of McLeod, Elizabeth, Rodney Salm, Alison Green, and
Marine Science 72: 741–752. Jeanine Almany. 2009. “Designing marine protected
Jones, Miranda C., Stephen R. Dye, John K. Pinnegar, Ra- area networks to address the impacts of climate
chel Warren, and William W. L. Cheung. 2014. “Us- change.” Frontiers in Ecology and the Environment 7 (7): 362–
ing scenarios to project the changing profitability of 370. https://doi.org/10.1890/070211.
fisheries under climate change.” Fish and Fisheries (May). Micheli, Fiorenza, Andrea Saenz-Arroyo, Ashley Green-
https://doi.org/10.1111/faf.12081. ley, Leonardo Vazquez, Jose Antonio Espinoza Mon-
Kaschner, Kristin, Derek P. Tittensor, Jonathan Ready, tes, Marisa Rossetto, and Giulio A. De Leo. 2012.
Tim Gerrodette, and Boris Worm. 2011. “Current and “Evidence that marine reserves enhance resilience to
future patterns of global marine mammal biodiver- climatic impacts.” PLOS One 7 (7): e40832. https://doi
sity.” PLOS One 6 (5): e19653. https://doi.org/10.1371/ .org/10.1371/journal.pone.0040832.
journal.pone.0019653. Millennium Ecosystem Assessment. 2005. Ecosystems and
Kroeker, K. J., R. L. Kordas, R. Crim, I. E. Hendriks, L. Human Well-being: Opportunities and Challenges for Business and
Ramajo, G. S. Singh, C. M. Duarte, and J.-P. Gattuso. Industry. World Resources Institute.
2013. “Impacts of ocean acidification on marine or- Miller, Kathleen A., Gordon R. Munro, U. Rashid Su-
ganisms: Quantifying sensitivities and interaction maila, and William W. L. Cheung. 2013. “Govern-
with warming.” Global Change Biology 19: 1884–1896. ing marine fisheries in a changing climate: A game-
Levin, L. A. 2003. “Oxygen minimum zone benthos: theoretic perspective.” Canadian Journal of Agricultural
Adaptation and community response to hypoxia.” Economics/Revue Canadienne d’agroeconomie 61 (2): 309–334.
Oceanography and Marine Biology: An Annual Review 41: 1–45. https://doi.org/10.1111/cjag.12011.
Levin, L. A., W. Ekau, A. J. Gooday, F. Jorissen, J. J. Mid- Moore, Jeffrey E., and Jay Barlow. 2011. “Bayesian state-
delburg, S. W. A. Naqvi, C. Neria, N. N. Rabalais, space model of fin whale abundance trends from a
and J. Zhang. 2009. “Effects of natural and human- 1991–2008 time series of line-transect surveys in
induced hypoxia on coastal benthos.” Biogeosciences 6 the California Current.” Journal of Applied Ecology 48 (5):
(10): 2063–2098. 1195–1205. https://doi.org/10.1111/j.1365-2664.2011
Levy, Jessica S., and Natalie C. Ban. 2013. “A method for .02018.x.
incorporating climate change modelling into marine Nikinmaa, M. 2002. “Oxygen-dependent cellular func-
conservation planning: An Indo–West Pacific exam- tions—Why fishes and their aquatic environment are
ple.” Marine Policy 38 (March): 16–24. https://doi.org/ a prime choice of study.” Comparative Biochemistry and
10.1016/j.marpol.2012.05.015. Physiology Part A: Molecular & Integrative Physiology 133: 1–16.
CLIM ATE CH A NGE A ND M A R INE BIODIV ER SIT Y 181
O’Connor, Mary I., John F. Bruno, Steven D. Gaines, Ben- in marine fish: Developing a hypothesis for cause and
jamin S. Halpern, Sarah E. Lester, Brian P. Kinlan, and effect relationships in Atlantic cod (Gadus morhua) and
Jack M. Weiss. 2007. “Temperature control of larval common eelpout (Zoarces viviparus).” Continental Shelf Re-
dispersal and the implications for marine ecology, search 21 (18–19): 1975–1997. https://doi.org/10.1016/
evolution, and conservation.” Proceedings of the National S0278-4343(01)00038-3.
Academy of Sciences 104 (4): 1266–1271. https://doi.org/ Pörtner, H. O., D. Karl, P. W. Boyd, W. Cheung, S. E.
10.1073/pnas.0603422104. Lluch-Cota, Y. Nojiri, D. N. Schmidt, and P. O. Za-
Parmesan, Camille, and Gary Yohe. 2003. “A globally vialov. 2014. “Ocean systems.” In Climate Change 2014:
coherent fingerprint of climate change impacts across Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral
natural systems.” Nature 421 (6918): 37–42. Aspects. Contribution of Working Group II to the Fifth Assessment
Pauly, D. 1997. “Geometric constraints on body size.” Report of the Intergovernmental Panel on Climate Change, ed. C.
Trends in Ecology & Evolution 12 (11): 442–443. B. Field, V. R. Barros, D. J. Dokken, K. J. Mach, M. D.
Pauly, D. 2010a. “The state of fisheries.” In Conservation Mastrandrea, T. E. Bilir, M. Chatterjee, et al., 1–138.
Biology for All, ed. S. N. S. Sodhi and P. R. Ehrlich, 118– Cambridge University Press.
120. Oxford University Press. Pörtner, H.-O., C. Bock, and F. C. Mark. 2017. “Oxygen-
Pauly, D. 2010b. Gasping Fish and Panting Squids: Oxygen, Tem- and capacity-limited thermal tolerance: Bridging ecol-
perature and the Growth of Water-Breathing Animals. Interna- ogy and physiology.” Journal of Experimental Biology 220:
tional Ecology Institute. 2685–2696.
Pauly, D., and W. W. L. Cheung. 2017. “Sound physiolog- Saba, V. S., C. A. Stock, J. R. Spotila, F. V. Paladino, and
ical knowledge and principles in Odelling shrinking P. S. Tomillo. 2012. “Projected response of an endan-
of fishes under climate change.” Global Change Biology 25: gered marine turtle population to climate change.”
e15–e26. Nature Climate Change 2 (11): 814–820.
Perry, Allison L., Paula J. Low, Jim R. Ellis, and John Salvadeo, C. J., D. Lluch-Belda, A. Gómez-Gallardo, J.
D. Reynolds. 2005. “Climate change and distribu- Urbán-Ramírez, and C. D. MacLeod. 2010. “Climate
tion shifts in marine fishes.” Science 308: 1912–1915. change and a poleward shift in the distribution of
https://doi.org/10.1126/science.1111322. the Pacific white-sided dolphin in the northeastern
Perry, R. Ian, Manuel Barange, and Rosemary E. Ommer. Pacific.” Endangered Species Research 11 (March): 13–19.
2010. “Global changes in marine systems: A social- https://doi.org/10.3354/esr00252.
ecological approach.” Progress in Oceanography 87 (1–4): Schmidtko, S., L. Stramma, and M. Visbeck. 2017. “De-
331–337. cline in global oceanic oxygen content during the past
Petchey, Owen L., P. Timon Mcphearson, Timothy M. five decades.” Nature 542: 335–339.
Casey, and Peter J. Morin. 1999. “Environmental Simpson, Stephen D., Simon Jennings, Mark P. Johnson,
warming alters food-web structure and ecosystem Julia L. Blanchard, Pieter-Jan Schön, David W. Sims,
function.” Nature 402 (November): 69–72. and Martin J. Genner. 2011. “Continental shelf-wide
Pinnegar, J., T. Watt, and K. Kennedy. 2012. “Climate response of a fish assemblage to rapid warming of
change risk assessment for the marine and fisheries the sea.” Current Biology 21 (September): 1565–1570.
sector.” In UK Climate Change Risk Assessment. Project Code https://doi.org/10.1016/j.cub.2011.08.016.
GA0204. Defra. Sims, D. W., M. J. Genner, A. J. Southward, and S. J.
Pinsky, Malin L., Boris Worm, Michael J. Fogarty, Jorge Hawkins. 2001. “Timing of squid migration reflects
L. Sarmiento, and Simon A. Levin. 2013. “Marine taxa North Atlantic climate variability.” Proceedings of the Royal
track local climate velocities.” Science 341 (6151): 1239– Society of London B: Biological Sciences 268 (1485): 2607–
1242. https://doi.org/10.1126/science.1239352. 2611. https://doi.org/10.1098/rspb.2001.1847.
Pitcher, T. J., and W. W. L. Cheung. 2013. “Fisheries: Sumaila, U. Rashid, William W. L. Cheung, Vicky W. Y.
Hope or despair?” Marine Pollution Bulletin 74: 506–516. Lam, Daniel Pauly, and Samuel Herrick. 2011. “Cli-
Planque, Benjamin, Jean-Marc Fromentin, Philippe mate change impacts on the biophysics and econom-
Cury, Kenneth F. Drinkwater, Simon Jennings, R. Ian ics of world fisheries.” Nature Climate Change (Novem-
Perry, and Souad Kifani. 2010. “How does fishing al- ber): 1–8. https://doi.org/10.1038/nclimate1301.
ter marine populations and ecosystems sensitivity to Sydeman, W. J., and S. J. Bograd. 2009. “Marine eco-
climate?” Journal of Marine Systems 79 (3–4): 403–417. systems, climate and phenology: Introduction.” Marine
https://doi.org/10.1016/j.jmarsys.2008.12.018. Ecology Progress Series 393: 185–188.
Poloczanska, Elvira S., Christopher J. Brown, William Thomas, C. D., E. J. Bodsworth, R. J. Wilson, A. D. Sim-
J. Sydeman, Wolfgang Kiessling, David S. Schoeman, mons, Z. G. Davies, M. Musche, and L. Conradt. 2001.
Pippa J. Moore, Keith Brander, et al. 2013. “Global “Ecological and evolutionary processes at expanding
imprint of climate change on marine life.” Nature Cli- range margins.” Nature 411: 577–581.
mate Change 3 (10): 919–925. https://doi.org/10.1038/ Thomas, M. K., C. T. Kremer, C. A. Klausmeier, and E.
nclimate1958. Litchman. 2012. “A global pattern of thermal adap-
Pörtner, H. 2001. “Climate induced temperature effects tation in marine phytoplankton.” Science 338: 1085–
on growth performance, fecundity and recruitment 1089. https://doi.org/10.1126/science.1224836.
182 W H AT DOE S THE FUTUR E HOL D?
Tittensor, Derek P., Camilo Mora, Walter Jetz, Heike K. mental Change 22 (February): 21–35. https://doi.org/10
Lotze, Daniel Ricard, Edward Vanden Berghe, and Bo- .1016/j.gloenvcha.2011.08.002.
ris Worm. 2010. “Global patterns and predictors of Weishampel, J. F., D. A. Bagley, and L. M. Ehrhart. 2004.
marine biodiversity across taxa.” Nature 466 (7310): “Earlier nesting by loggerhead sea turtles follow-
1098–1101. https://doi.org/10.1038/nature09329. ing sea surface warming.” Global Change Biology 10 (8):
Trathan, Phil N., Pablo García-Borboroglu, Dee Boersma, 1424–1427.
Charles-André Bost, Robert J. M. Crawford, Glenn T. Witt, M. J., L. A. Hawkes, M. H. Godfrey, B. J. Godley,
Crossin, Richard J. Cuthbert, et al. 2014. “Pollution, and A. C. Broderick. 2010. “Predicting the impacts of
habitat loss, fishing, and climate change as critical climate change on a globally distributed species: The
threats to penguins.” Conservation Biology: The Journal of the case of the loggerhead turtle.” Journal of Experimental Biol-
Society for Conservation Biology 29 (1): 31–34. https://doi ogy 213 (6): 901–911.
.org/10.1111/cobi.12349. Wittman, A. C., and H. O. Pörtner. 2013. “Sensitivities of
United Nations. 2017. The First Global Integrated Marine As- extant animal taxa to ocean acidification.” Nature Cli-
sessment: World Ocean Assessment I. Cambridge University mate Change 3. https://doi.org/10.1038/nclimate1982.
Press. https://doi.org/10.1017/9781108186148. Wolf, S. G., M. A. Snyder, W. J. Sydeman, D. F. Doak,
Vuuren, Detlef P. van, Keywan Riahi, Richard Moss, Jae and D. A. Croll. 2010. “Predicting population con-
Edmonds, Allison Thomson, Nebojsa Nakicenovic, sequences of ocean climate change for an ecosystem
Tom Kram, et al. 2012. “A proposal for a new scenario sentinel, the seabird Cassin’s auklet.” Global Change Biol-
framework to support research and assessment in dif- ogy 16 (7): 1923–1935.
ferent climate research communities.” Global Environ-
CASE STUDY 5
There is clear evidence that the distributions of many species have been changing in ways
that correspond with recent changes in climate (Chen et al. 2011; Pinsky et al. 2013). As
climates continue to change, the ranges of many species will very likely continue to shift to
track suitable climates. Some species will be able to keep pace with changing climatic condi-
tions and others will not (Nathan et al. 2011; Schloss, Nuñez, and Lawler 2012). A species’
ability to track suitable climates will depend not only on the nature of the climatic changes
but also on the species’ ability to move across landscapes—many of which are dominated
by human activity.
A recent study explored how species cies were assigned according to the values
might move across the landscapes of the of the HII, with areas of higher human in-
Western Hemisphere to track projected fluence receiving higher resistance values.
changes in climate (Lawler et al. 2013). Movement was then modeled from each
The study drew on the projections of cli- grid cell in a digital map that represented
matic niche models built for 2,903 spe- the area projected to become climatically
cies of birds, mammals, and amphibians unsuitable to the area of the map that was
(Lawler et al. 2009). The authors used the projected to be climatically suitable in the
models to identify areas that might become future. Likewise, movement was modeled
newly climatically suitable and those that from the area that is currently suitable into
might become climatically unsuitable for each of the grid cells in areas projected to
each species by the end of the twenty-first become newly climatically suitable. When
century under 10 different future climate the results from the individual species anal-
projections. They then mapped potential yses were combined and averaged across
movement routes that individuals of each the 10 climate change projections, clear
species might follow to either move into ar- patterns of potential climate-driven move-
eas projected to become newly climatically ments emerged.
suitable or out of areas projected to become There are likely to be several places in
climatically unsuitable. North and South America that will act as
To model the potential effects of patterns critical conduits for species movements in
of human activity on movement routes, the a changing climate (see Plate 6). Two ar-
authors used the Human Influence Index eas with pronounced, consistent, projected
(HII; Sanderson et al. 2002). The HII maps movements of species are in southeastern
patterns of human impact using estimates North America where species are projected
of human population densities, distance to move northeast, up into the Appalachian
to roads, land cover, and nighttime lights. Mountains and along their length to the
Movement routes were mapped using Cir- north, and in central Argentina, where spe-
cuitscape, a modeling tool that maps land- cies are projected to move southward into
scape connectivity on the basis of electri- the southern Pampas and up in the Sierras
cal circuit theory (McRae et al. 2008). de Córdoba and Andes. These are places
Circuitscape treats a landscape as a circuit
board in which each grid cell in a digital
map is a resistor with an assigned resistance Copyright © 2019 by Yale University. All rights reserved.
value. Resistance values for the 2,903 spe- ISBN 978-0-300-20611-1.
183
184 W H AT DOE S THE FUTUR E HOL D?
where the projected changes in climate are connectivity mapping efforts, it may be
likely to force species movements poleward possible to identify specific areas in which
and upward in elevation. They are also focused conservation actions will facilitate
places where the human presence on the climate-driven species movements.
landscape is likely to channel movements
in a specific way.
In general, the tundra, boreal forest and R E FE R E NCE S
taiga, north temperate broadleaf and mixed
forests, and the tropical and subtropical Brost, B., and P. Beier. 2012. “Use of land facets to de-
sign linkages for climate change.” Ecological Applications
moist broadleaf forest ecoregions are pro-
22 (1): 87–103.
jected to experience the largest number of Chen, I. C., J. K. Hill, R. Ohlemueller, D. B. Roy, and C.
movements (controlling for current species D. Thomas. 2011. “Rapid range shifts of species as-
richness). The far northern latitudes, the sociated with high levels of climate warming.” Science
Amazon Basin, and northern Paraguay and 333 (6045): 1024–1026.
Heller, N. E., and E. S. Zavaleta. 2009. “Biodiversity man-
southeastern Bolivia are projected to expe-
agement in the face of climate change: A review of 22
rience many movements and are generally years of recommendations.” Biological Conservation 142
less affected by human activity. Thus, these (1): 14–32.
areas, as well as the areas of intense antici- Lawler, J. J., S. L. Shafer, D. White, P. Kareiva, E. P. Mau-
pated movement in central Argentina and rer, A. R. Blaustein, and P. J. Bartlein. 2009. “Projected
climate-induced faunal change in the Western Hemi-
the southern Appalachians, may be ones
sphere.” Ecology 90 (3): 588–597.
where conservation efforts to protect con- Lawler, J. J., Aaron S. Ruesch, J. D. Olden, and B. H.
nectivity may serve moving species well. McRae. 2013. “Projected climate-driven faunal move-
Protecting and enhancing connectivity ment routes.” Ecology Letters 16 (8): 1014–1022.
to allow species to track changes in cli- McRae, B. H., B. G. Dickson, T. H. Keitt, and V. B. Shah.
2008. “Using circuit theory to model connectivity
mate are often-cited adaptation strategies
in ecology, evolution, and conservation.” Ecology 89:
for conserving biodiversity in the face of 2712–2724. https://doi.org/10.1890/07-1861.1.
climate change (Heller and Zavaleta 2009). Nathan, R., N. Horvitz, Y. P. He, A. Kuparinen, F. M. Sch-
Targeting connectivity efforts at areas iden- urr, and G. G. Katul. 2011. “Spread of North American
tified to be critical for facilitating species wind-dispersed trees in future environments.” Ecology
Letters 14 (3): 211–219.
movements is a first step in planning for cli-
Nuñez, T. A., J. J. Lawler, B. H. McRae, D. J. Pierce, M.
mate-driven species movements. However, B. Krosby, D. M. Kavanagh, P. H. Singleton, and J. J.
the hemispheric-scale study described here Tewksbury. 2013. “Connectivity planning to address
was conducted at a coarse spatial resolution climate change.” Conservation Biology 27 (2): 407–416.
and thus does not provide the fine-scale https:/doi.org/10.1111/cobi.12014.
Pinsky, M. L., B. Worm, M. J. Fogarty, J. L. Sarmiento,
analyses needed to target specific sites for
S. A. Levin. 2013. “Marine taxa track local climate ve-
on-the-ground conservation efforts. Sev- locities.” Science 341 (6151): 1239–1242.
eral studies have begun to explore how to Sanderson, E. W., M. Jaiteh, M. A. Levy, K. H. Red-
identify specific locations that, if protected ford, A. V. Wannebo, and G. Woolmer. 2002. “The
or restored, have the potential to facilitate human footprint and the last of the wild the hu-
man footprint is a global map of human influ-
climate-driven species movements. These
ence on the land surface, which suggests that hu-
studies have focused on linking snapshots man beings are stewards of nature, whether we like
of potentially climatically suitable habitat it or not.” BioScience 52 (10): 891–904. https://doi
through time, mapping routes that follow .org/10.1641/0006-3568(2002)052[0891:THFATL]2.0
climatic gradients (Nuñez et al. 2013), and .CO;2.
Schloss, C. A., T. A. Nuñez, and J. J. Lawler. 2012. “Dis-
connecting similar (and diverse sets of) to-
persal will limit ability of mammals to track climate
pographically defined zones (e.g., Brost and change in the Western Hemisphere.” Proceedings of the
Beier 2012). By combining projections from National Academy of Science 109: 8606–8611.
coarser resolution studies with fine-scale
CHAPTER FOURTEEN I N T RODUCTION
185
186 W H AT DOE S THE FUTUR E HOL D?
Figure 14.1. CO2 equilibrium in seawater, showing how Even though ocean acidification and its
the increase in hydrogen ion concentration (due to the impacts constitute a relatively new field of
dissociation of carbonic acid, H2CO3) leads to the con-
version of carbonate (CO32–) to bicarbonate (HCO3–).
study, the exponential growth in research
This results in a shift in relative concentrations of the provides a large body of evidence from
inorganic carbon compounds as atmospheric forcing which the question “How will ocean acidi-
increases from a CO2 concentration of 280 μatm (the fication affect marine biodiversity?” can be
preindustrial level) through a doubling (560 μatm) and addressed. This chapter first summarizes
tripling (840 μatm).
the effects of ocean acidification from a
geochemical, physiological and behavioral,
well documented. But at local scales, ocean
and biogeochemical perspective, and then
acidification occurs against a backdrop of
addresses how those effects ultimately have
many processes that can affect the CO2
an impact on marine biodiversity.
system in seawater and that operate from
hourly to seasonal and longer time scales.
Ocean acidification was originally de-
HOW AC IDIFIC ATION IM PACTS
scribed in terms of geochemical effects,
ORGA N ISMS
that is, how changes in the CO2 system in
seawater affect the formation and dissolu-
Geochemical
tion of the biomineral calcium carbonate
(CaCO3). Initially it was thought that ocean The most widely recognized consequence
acidification was primarily a problem for of ocean acidification is its effects on the
marine calcifiers, particularly those organ- precipitation and dissolution of CaCO3.
isms with only weak control over their Previous studies illustrate that the rates
skeletal-building process, such as algae, of precipitation and dissolution of abiotic
corals, and foraminifera. It is now widely CaCO3 are a function of the ion concen-
recognized that the full suite of chemical trations of calcium and carbonate. This
changes in the CO2 system impact not only relationship is demonstrated by the dis-
CaCO3 formation and dissolution, but also tribution of CaCO3 on the seafloor; CaCO3
biogeochemistry, physiology, and behav- sediments are absent where the overlying
ior (Figure 14.2), and across a wide group waters are undersaturated with respect to
of organisms, from protists to fish. Thus, aragonite and calcite. In fact, the most re-
the challenge of understanding how ocean cent analogue for a global ocean acidifica-
acidification will affect marine biodiversity tion event—the Paleocene-Eocene Ther-
is perhaps as complicated as understand- mal Maximum 55 million years ago—was
ing how temperature increases will affect identified in the geologic record as the
biodiversity. sudden disappearance of CaCO3 sediments
I M PAC T S OF O C E A N AC I DI F IC AT ION ON M A R I N E BIODI V E R SI T Y 187
Figure 14.2. Simplified representation of how the various Physiological and Behavioral
components of the CO2 system in seawater (gray boxes)
impact marine organisms, which collectively affect Ocean acidification is known to have both
biodiversity. favorable and unfavorable impacts on the
physiology of marine organisms. The
from much of the ocean floor (Zachos et most common theme in any summary of
al. 2005). Most CaCO3 production in the ocean acidification impacts on physiol-
ocean is not abiotic, however, but bio- ogy is the high degree of variability in the
logically mediated. Almost all calcifying responses. Physiologically, organisms re-
organisms exert some control over their spond to acidification with alterations in
calcification by isolating seawater and el- photosynthesis, calcification, growth, and
evating its ionic concentrations of calcium reproduction. The wide range of responses
and carbonate. Nonetheless, the rates of to ocean acidification largely reflects both
biotic precipitation in many (but not all) how an organism obtains energy and how
organisms reflect the saturation state of the it allocates it to each one of these processes.
surrounding seawater. Thus, biotic calci- If ocean acidification leads to greater en-
fication has both geochemical and physi- ergy uptake, then the organism may ben-
ological components. The physiological efit physiologically. If ocean acidification
aspects of biotic calcification are further presents an energetic demand, then the
addressed below. organism must respond with a reallocation
The geochemical impacts of ocean acidi- of energy from other processes. To com-
fication are considered a particular threat plicate matters, the responses vary within
to marine biodiversity within ecosystems single species with genotypic plasticity and
where calcium carbonate production pro- genetic diversity, with physiological state,
vides the substrate that supports biodiver- and in the presence of other environmen-
sity. Coral reefs, cold-water reefs, and oyster tal stressors.
reefs form where CaCO3 production exceeds
its dissolution (Plate 2). Dissolution is often
PR I M A RY PRODUCT ION
overlooked in its importance, but when dis-
solution rates increase, the ability to grow Photosynthetic rates of marine primary pro-
or maintain reef structures is compromised ducers that are carbon limited should bene-
even if the reef-building organisms are able fit, for example, from an increase in concen-
to maintain calcification rates. The loss of trations of CO2 and/or HCO3–, at least when
reef substrate is exacerbated by the fact that other nutrients are not limiting. Carbon fix-
ocean acidification enhances the biological ation rates in marine phytoplankton do tend
breakdown and dissolution of CaCO3 (An- to increase with ocean acidification, but
dersson and Gledhill 2013). only slightly. The genetic diversity of many
188 W H AT DOE S THE FUTUR E HOL D?
phytoplankton groups seems to explain the within the genus Acropora. Within mollusks,
wide range of results in experiments testing calcification rates of most pteropod species
the responses of phytoplankton species to are consistently reduced, but in the adults of
ocean acidification. A comprehensive study many marine gastropods and bivalves the
of 16 ecotypes of the picoplankton Ostreococ- calcification responses vary strongly. Even
cus tauri revealed that although all ecotypes within the coccolithophore Emiliania huxleyi,
experienced increases in photosynthesis the calcification response to ocean acidifica-
and growth, the responses ranged by a fac- tion varies widely among genotypes (Raven
tor of two (Schaum et al. 2013). Such stud- and Crawfurd 2012).
ies are important because they reveal the
potential of genetic variation in maintain-
R E PRODUCT ION
ing phytoplankton populations into the fu-
ture. For marine macroalgae and seagrasses, Many marine organisms invest a large
Koch et al. (2013) surmised that because amount of energy in reproduction, either
most are C3 photosynthesizers and carbon through asexual means or in the produc-
limited, their photosynthesis and growth tion of sexual gametes. Ocean acidification
rates are also likely to benefit from increases has been found to affect the fecundity and
in CO2 and HCO3– concentrations. Increases quality of gametes that are produced, fertil-
in production can alter food-web dynamics, ization success, larval survival, and recruit-
and in some species it appears that increases ment, but these impacts are certainly not
in primary production are accompanied by universal. Some species, such as copepods,
changes in carbon to nutrient stoichiom- allocate more energy toward reproduction
etry, which reduces their nutritional value with ocean acidification (Fitzer et al. 2012).
for herbivores (e.g., Verspagen et al. 2014). Reproduction in most fish species appears
For at least three species of seagrasses (Cy- to remain robust to the projected decrease
modocea nodosa, Rupia maritima, and Potamogeton in ocean pH, but the responses are ex-
perfoliatus), the benefits of increased growth tremely varied (Secretariat of the Conven-
appear to be compromised by a loss in phe- tion on Biological Diversity 2014).
nolic acid concentrations that act to deter
grazing (Arnold et al. 2012).
A N I M A L BE H AV IOR
deep-water Scleractinian corals.” Bulletin of Marine Science Nature 454 (7200): 96–99. https://doi.org/10.1038/
81 (3): 311–322. nature07051.
Chivers, D. P., M. I. McCormick, G. E. Nilsson, P. L. Mun- Hutchins, D. A., M. R. Mulholland, and F. X. Fu. 2009.
day, S. A. Watson, M. G. Meekan, M. D. Mitchell, K. C. “Nutrient cycles and marine microbes in a CO2-
Corkill, and M. C. O. Ferrari. 2014. “Impaired learn- enriched ocean.” Oceanography 22 (4): 128–145.
ing of predators and lower prey survival under ele- Inoue, S., H. Kayanne, S. Yamamoto, and H. Kurihara.
vated CO2: A consequence of neurotransmitter inter- 2013. “Spatial community shift from hard to soft cor-
ference.” Global Change Biology 20 (2): 515–522. https:// als in acidified water.” Nature Climate Change 3 (7): 683–
doi.org/10.1111/gcb.12291. 687. https://doi.org/10.1038/nclimate1855.
Connell, S. D., K. J. Kroeker, K. E. Fabricius, D. I. Kline, Kelly, Morgan W., and Gretchen E. Hofmann. 2013.
and B. D. Russell. 2013. “The other ocean acidifica- “Adaptation and the physiology of ocean acidifica-
tion problem: CO2 as a resource among competitors tion.” Functional Ecology 27 (4): 980–990. https://doi
for ecosystem dominance.” Philosophical Transactions of the .org/10.1111/j.1365-2435.2012.02061.x.
Royal Society B: Biological Sciences 368 (1627). https://doi Kerfahi, D., J. M. Hall-Spencer, B. M. Tripathi, M. Mi-
.org/10.1098/rstb.2012.0442. lazzo, J. Lee, and J. M. Adams. 2014. “Shallow water
Doropoulos, C., and G. Diaz-Pulido. 2013. “High marine sediment bacterial community shifts along a
CO2 reduces the settlement of a spawning coral on natural CO2 gradient in the Mediterranean Sea off Vul-
three common species of crustose coralline algae.” cano, Italy.” Microbial Ecology 67 (4): 819–828. https://
Marine Ecology Progress Series 475: 93–99. https://doi doi.org/10.1007/s00248-014-0368-7.
.org/10.3354/meps10096. Kleypas, J. A., R. A. Feely, V. J. Fabry, C. Langdon, C.
Duffy, J. E., and J. J. Stachowicz. 2006. “Why biodiver- L. Sabine, and L. L. Robbins. 2006. “Impacts of
sity is important to oceanography: Potential roles of ocean acidification on coral reefs and other marine
genetic, species, and trophic diversity in pelagic eco- calcifiers: A guide for future research.” Report of a
system processes.” Marine Ecology Progress Series 311: 179– workshop held 18–20 April 2005, St. Petersburg, FL,
189. https://doi.org/10.3354/meps311179. sponsored by NSF, NOAA, and the U.S. Geological
Dukes, J. S., and H. A. Mooney. 1999. “Does global Survey.
change increase the success of biological invaders?” Koch, M., G. Bowes, C. Ross, and X. H. Zhang. 2013.
Trends in Ecology & Evolution 14 (4): 135–139. https://doi “Climate change and ocean acidification ef-
.org/10.1016/s0169-5347(98)01554-7. fects on seagrasses and marine macroalgae.” Global
Fabricius, K. E., G. De’ath, S. Noonan, and S. Uthicke. Change Biology 19 (1): 103–132. https://doi.org/
2014. “Ecological effects of ocean acidification and 10.1111/j.1365-2486.2012.02791.x.
habitat complexity on reef-associated macroinver- Kroeker, K. J., R. L. Kordas, R. N. Crim, and G. G. Singh.
tebrate communities.” Proceedings of the Royal Society B: 2010. “Meta-analysis reveals negative yet variable ef-
Biological Sciences 281 (1775). https://doi.org/10.1098/ fects of ocean acidification on marine organisms.”
rspb.2013.2479. Ecology Letters 13 (11): 1419–1434. https://doi.org/
Fabricius, K. E., C. Langdon, S. Uthicke, C. Humphrey, 10.1111/j.1461-0248.2010.01518.x.
S. Noonan, G. De’ath, R. Okazaki, N. Muehllehner, M. Kroeker, K. J., F. Micheli, and M. C. Gambi. 2013. “Ocean
S. Glas, and J. M. Lough. 2011. “Losers and winners acidification causes ecosystem shifts via altered com-
in coral reefs acclimatized to elevated carbon dioxide petitive interactions.” Nature Climate Change 3 (2): 156–
concentrations.” Nature Climate Change 1 (3): 165–169. 159. https://doi.org/10.1038/nclimate1680.
https://doi.org/10.1038/NCLIMATE1122. Lindh, M. V., L. Riemann, F. Baltar, C. Romero-Oliva, P.
Fitzer, S. C., G. S. Caldwell, A. J. Close, A. S. Clare, R. S. Salomon, E. Graneli, and J. Pinhassi. 2013. “Con-
C. Upstill-Goddard, and M. G. Bentley. 2012. “Ocean sequences of increased temperature and acidification
acidification induces multi-generational decline in on bacterioplankton community composition during
copepod naupliar production with possible conflict a mesocosm spring bloom in the Baltic Sea.” Environ-
for reproductive resource allocation.” Journal of Experi- mental Microbiology Reports 5 (2): 252–262. https://doi
mental Marine Biology and Ecology 418: 30–36. https://doi .org/10.1111/1758-2229.12009.
.org/10.1016/j.jembe.2012.03.009. Lohbeck, Kai T., Ulf Riebesell, and Thorsten B. H. Re-
Hale, R., P. Calosi, L. McNeill, N. Mieszkowska, and S. usch. 2012. “Adaptive evolution of a key phytoplank-
Widdicombe. 2011. “Predicted levels of future ocean ton species to ocean acidification.” Nature Geoscience 5
acidification and temperature rise could alter com- (5): 346–351. https://doi.org/10.1038/ngeo1441.
munity structure and biodiversity in marine benthic McCormick, M. I., S. A. Watson, and P. L. Munday. 2013.
communities.” Oikos 120 (5): 661–674. https://doi “Ocean acidification reverses competition for space
.org/10.1111/j.1600-0706.2010.19469.x. as habitats degrade.” Scientific Reports 3. https://doi
Hall-Spencer, J. M., R. Rodolfo-Metalpa, S. Martin, E. .org/10.1038/srep03280.
Ransome, M. Fine, S. M. Turner, S. J. Rowley, D. Te- Motegi, C., T. Tanaka, J. Piontek, C. P. D. Brussaard, J.
desco, and M. C. Buia. 2008. “Volcanic carbon dioxide P. Gattuso, and M. G. Weinbauer. 2013. “Effect of
vents show ecosystem effects of ocean acidification.” CO2 enrichment on bacterial metabolism in an Arc-
I M PAC T S OF O C E A N AC I DI F IC AT ION ON M A R I N E BIODI V E R SI T Y 195
tic fjord.” Biogeosciences 10 (5): 3285–3296. https://doi Secretariat of the Convention on Biological Diversity.
.org/10.5194/bg-10-3285-2013. 2014. “An updated synthesis of the impacts of ocean
Munday, P. L., A. J. Cheal, D. L. Dixson, J. L. Rummer, acidification on marine biodiversity.” In Technical Series
and K. E. Fabricius. 2014. “Behavioural impairment No. 75, ed. S. Hennige, J. M. Roberts, and P. William-
in reef fishes caused by ocean acidification at CO2 son. Montreal: Secretariat of the Convention on Bio-
seeps.” Nature Climate Change 4 (6): 487–492. https://doi logical Diversity.
.org/10.1038/nclimate2195. St. Pierre, J. I., and K. E. Kovalenko. 2014. “Effect of habi-
Nielsen, L. T., H. H. Jakobsen, and P. J. Hansen. 2010. tat complexity attributes on species richness.” Ecosphere
“High resilience of two coastal plankton com- 5 (2). https://doi.org/10.1890/es13-00323.1.
munities to twenty-first century seawater acidi- Sunday, J. M., P. Calosi, S. Dupont, P. L. Munday, J. H.
fication: Evidence from microcosm studies.” Ma- Stillman, and T. B. H. Reusch. 2014. “Evolution in an
rine Biology Research 6 (6): 542–555. https://doi acidifying ocean.” Trends in Ecology & Evolution 29 (2):
.org/10.1080/17451000903476941. 117–125. https://doi.org/10.1016/j.tree.2013.11.001.
O’Gorman, E. J., and M. C. Emmerson. 2009. “Perturba- Tortell, P. D., C. D. Payne, Y. Y. Li, S. Trimborn, B. Rost,
tions to trophic interactions and the stability of com- W. O. Smith, C. Riesselman, R. B. Dunbar, P. Sedwick,
plex food webs.” Proceedings of the National Academy of Sci- and G. R. DiTullio. 2008. “CO2 sensitivity of Southern
ences of the United States of America 106 (32): 13393–13398. Ocean phytoplankton.” Geophysical Research Letters 35 (4).
https://doi.org/10.1073/pnas.0903682106. https://doi.org/10.1029/2007gl032583.
Pörtner, H. O. 2012. “Integrating climate-related stressor Uthicke, S., D. Pecorino, R. Albright, A. P. Negri, N.
effects on marine organisms: Unifying principles Cantin, M. Liddy, S. Dworjanyn, P. Kamya, M. Byrne,
linking molecule to ecosystem-level changes.” Ma- and M. Lamare. 2013. “Impacts of ocean acidifica-
rine Ecology Progress Series 470: 273–290. https://doi.org/ tion on early life-history stages and settlement of the
10.3354/meps10123. coral-eating sea star Acanthaster planci.” PLOS One 8 (12).
Raven, J. A., and K. Crawfurd. 2012. “Environmental https://doi.org/10.1371/journal.pone.0082938.
controls on coccolithophore calcification.” Marine Ecol- Verspagen, J. M. H., D. B. Van de Waal, J. F. Finke, P. M.
ogy Progress Series 470: 137–166. https://doi.org/10.3354/ Visser, and J. Huisman. 2014. “Contrasting effects of
meps09993. rising CO2 on primary production and ecological stoi-
Russell, B. D., C. D. G. Harley, T. Wernberg, N. Miesz- chiometry at different nutrient levels.” Ecology Letters 17
kowska, S. Widdicombe, J. M. Hall-Spencer, and S. D. (8): 951–960. https://doi.org/10.1111/ele.12298.
Connell. 2012. “Predicting ecosystem shifts requires Vinebrooke, R. D., K. L. Cottingham, J. Norberg, M.
new approaches that integrate the effects of climate Scheffer, S. I. Dodson, S. C. Maberly, and U. Sommer.
change across entire systems.” Biology Letters 8 (2): 164– 2004. “Impacts of multiple stressors on biodiversity
166. https://doi.org/10.1098/rsbl.2011.0779. and ecosystem functioning: The role of species co-
Schaum, E., B. Rost, A. J. Millar, and S. Collins. 2013. tolerance.” Oikos 104 (3): 451–457.
“Variation in plastic responses of a globally distrib- Weinbauer, M. G., X. Mari, and J.-P. Gattuso. 2011. “Ef-
uted picoplankton species to ocean acidification.” fects of ocean acidification on the diversity and activ-
Nature Climate Change 3 (3): 298–302. https://doi.org/ ity of heterotrophic marine microorganisms.” In Ocean
10.1038/nclimate1774. Acidification, ed. J. P. Gattuso and L. Hansson, 83–98.
Schulz, K. G., R. G. J. Bellerby, C. P. D. Brussaard, J. Oxford University Press.
Budenbender, J. Czerny, A. Engel, M. Fischer, S. Koch- Zachos, J. C., U. Rohl, S. A. Schellenberg, A. Sluijs, D.
Klavsen, S. A. Krug, S. Lischka, A. Ludwig, M. Mey- A. Hodell, D. C. Kelly, E. Thomas, M. Nicolo, I. Raffi,
erhofer, G. Nondal, A. Silyakova, A. Stuhr, and U. L. J. Lourens, H. McCarren, and D. Kroon. 2005.
Riebesell. 2013. “Temporal biomass dynamics of an “Rapid acidification of the ocean during the Paleo-
Arctic plankton bloom in response to increasing lev- cene-Eocene Thermal Maximum.” Science 308 (5728):
els of atmospheric carbon dioxide.” Biogeosciences 10 (1): 1611–1615.
161–180. https://doi.org/10.5194/bg-10-161-2013.
CHAPTER FIFTEEN OV E RV IE W
196
T ROPICA L FOR E STS IN A CH A NGING CLIM ATE 197
One way of broadly ascertaining how America and Southeast Asia are less stable
vulnerable tropical forests are going to than climates in tropical forest ecoregions
be with future climate change is to assess in Africa and Oceania (Figures 15.1 and
their “stability” at the ecoregional scale. 15.2B). When tropical forest climate sta-
Climate stability has been defined as the bility is integrated with the ecoregion’s
similarity between current and future cli- “natural integrity” (defined as the pro-
mate (in the 2050s) using a six-variable portion of intact natural vegetation found
envelope-based gauge to represent gen- in each ecoregion, and thus a function of
eral climate patterns and seasonality past land use), a measure of global ecosys-
(Watson et al. 2013). This gauge combines tem vulnerability emerges that comprises
patterns of annual mean temperature, elements of exposure to climate change
mean diurnal temperature range, mean and an ecoregions adaptive capacity (Fig-
annual temperature range, annual pre- ures 15.1 and 15.2C). It is clear that some
cipitation, precipitation seasonality, and tropical ecoregions are very vulnerable
precipitation of the driest quarter, and it to a changing climate, with East African,
assesses the degree to which regions are southeastern South American, and north-
likely to shift outside of their current cli- ern Southeast Asian forests standing out
mate envelopes. By this measure, climates as areas of particular concern.
in tropical forest ecoregions in South
If a formal carbon market for REDD+ does followed by the Americas (0.45 percent per
eventuate, it will have significant financial year) and Africa (0.28 percent per year)
implications. A recent broad global analy- (Malhi et al. 2014). Remaining tropical for-
sis also showed that between 2000 and ests not severely affected by land clearance
2005, reduced carbon emissions from the have been subject to pervasive disturbances
17.2 million ha (19.6 percent) of protected from logging, defaunation, wildfire, and
humid tropical forests were worth approxi- fragmentation. Their separate and combined
mately US$6.2–7.4 billion (Scharlemann et effects have resulted in marked forest degra-
al. 2010). Tropical forests bolster national dation, loss of biodiversity, and impairment
economies through tourism and timber of ecological processes (Figure 15.1). Many
revenues and provide essential ecosystem human-modified tropical forest landscapes
services to many thousands of local com- exist as complex mosaics of primary- and
munities (de Groot et al. 2012). second-growth forest, with a patchwork of
The level of recent tropical forest loss regenerating areas on fallow or abandoned
has been enormous. In 2000, tropical for- agricultural land. Importantly, though of-
ests covered 10 percent–13 percent of the ten ignored, tropical forested ecosystems
terrestrial surface of the earth; today, intact around the world are already experienc-
tropical forests make up less than 5 per- ing the impacts of human-forced climate
cent of the terrestrial surface of the earth change.
(Mackey et al. 2014; Watson et al. 2018). Ab- In this chapter, we briefly outline the
solute loss rates for dense forests between magnitude of impact human activities
2000 and 2012 were highest in the tropical have had on tropical forests to date. We
Americas (39,900 km2 per year), followed then describe likely responses of species
by tropical Asia (22,300 km2 per year), and and ecosystems across the tropical forest
Africa (11,000 km2 per year). Relative rates biome under climate change forecasts, fo-
were highest in Asia (0.62 percent per year), cusing on vulnerabilities of individuals and
Figure 15.1. The relationship between climate stability the average over the results from the seven GCMs (see
and mean intactness in tropical forested ecoregions by Iwamura et al. 2013 for additional details). Vegetation
different region (n = 308). Tropical forested ecoregions intactness was calculated using the GlobCover V2 data
are defined broadly as those containing a tropical or sub- set, as 1 minus the proportion of the ecoregional area
tropical moist broadleaf forests, dry broadleaf forests, where native vegetation has been transformed through
coniferous forests, and mangroves. Climate stability was agricultural development and urbanization. The intact-
defined as a measure of similarity between forecasted ness axis has been transformed to a normal distribution
climatic conditions in 2050 and the current climate for presentation purposes by taking the square root val-
of the ecoregion. The climate stability shown here is ues. (Figure and data adopted from Watson et al. 2013.)
T ROPICA L FOR E STS IN A CH A NGING CLIM ATE 199
population, communities, and ecosystem. atmospheric circulation, changes in tropi-
Although these direct impacts will be im- cal forest cover almost certainly have global
portant, we argue that the human response consequences for temperature and precipi-
to climate change in tropical forest land- tation patterns (Figure 15.2).
scapes are going to significantly increase
the vulnerability of already highly modi-
What Does the Future Look Like
fied tropical forest systems, and we describe
for Tropical Forests?
some of the likely major consequences. We
conclude with some suggested strategies Atmospheric CO2 has risen from preindus-
for conservation of tropical forests that take trial concentrations of 280 parts per million
into account both the biological and hu- (ppm) to 400 ppm in 2014, and it could to
man response to current and future climate rise to 600–1,000 ppm by 2100 under some
change. emissions scenarios (IPCC 2013). Anthro-
pogenic greenhouse gas emissions have
committed tropical forests to increased
IM PACTS OF A N T H ROPO GE N IC temperatures, altered rainfall patterns, and
C L IM AT E C H A NGE TO DAT E resulted in increases in the frequency and
severity of extreme events (Collins 2013).
Since the mid-1970s land-surface tempera- The speed and magnitude of these changes
tures in tropical rainforest regions have will have large-scale impacts on tropical
been increasing by ~0.25°C per decade as forest biodiversity, as the climates through-
a result of increases in the concentration out the tropics will fall outside of the his-
of atmospheric greenhouse gases (Malhi et toric range of variability for tropical forest
al. 2009). In regions that have been highly regions up to a decade faster than any other
deforested, this warming has been exacer- major terrestrial ecosystem and depend-
bated by the direct loss of tree cover, which ing on emission scenarios, possibly within
decreases land-surface evapotranspiration the next 25–45 years (Mora et al. 2013).
and alters cloud formation patterns (Davin This extremely rapid departure from nine-
and de Noblet-Ducoudré 2010). Observa- teenth- and twentieth-century climates will
tional studies in tropical forest mountain affect tropical forest communities in a wide
transects suggest that there has been an range of ways, and our capacity to predict
upward shift in mean species distributions those impacts is uneven.
of species (Colwell et al. 2008). There is Predicted changes in the physical climate
already direct, observational evidence of can be broken into three broad categories—
changes in the hydrological cycle; over the changes in temperature, precipitation, and
past 70 years, many tropical regions have CO2. Mean land-surface temperatures in
seen a steady increase in the seasonality of tropical forest regions are expected to in-
rainfall and a steady decrease in rainfall crease by 3°C–6°C this century (Zelazowski
predictability (Feng, Porporato, and Rodri- et al. 2011). The warming predicted is simi-
guez-Iturbe 2013). Two large-scale rainfall lar in extent to the warming of the Paleo-
exclusion experiments in Brazil have dem- cene-Eocene Thermal Maximum (PETM),
onstrated that large trees are particularly but the speed of the warming is unprec-
vulnerable to drought (da Costa et al. 2010), edented, perhaps two orders of magnitude
and field surveys across Amazonia following faster than previous periods of rapid tropi-
an intense drought in 2005 demonstrated a cal warming (Jaramillo et al. 2010). Global
widespread increase in tree mortality and precipitation is positively linked to global
a decrease in rates of carbon sequestration temperature increase, but in tropical re-
(Phillips et al. 2009). Because the tropics gions the most robust signal is for changes
play an important role in regulating global to, and increases in, rainfall seasonal-
Figure 15.2. Global distribution of tropical forest ecoregional vegetation intactness (A), climate stability (B), and an overlay of climate
stability and vegetation intactness (C). Ecoregions that have high relative climate stability and high vegetation intactness are depicted in
black. Ecoregions that have low relative climate stability but high vegetation intactness are depicted in dark gray. Ecoregions that have
both low relative climate stability and low levels of vegetation intactness are depicted in light gray. The tropical forested ecoregions of
Asia are more climatically stable but less intact, relative to the tropical forests of the Amazon and Congo Basins. The darker colors on
panel A indicate higher ecoregional intactness, while darker colors on panel B represent more relative stable climates. (Adapted from
Watson et al. 2013.)
T ROPICA L FOR E STS IN A CH A NGING CLIM ATE 201
ity. This is characterized by more intense climation capacities than their temperate
dry seasons, stronger and more frequent counterparts (Somero 2010).
droughts, and stronger, longer-lasting heat A fundamental point of concern for
waves (Malhi et al. 2014). This pattern is tropical biodiversity focuses on the rela-
particularly clear in eastern Amazonia (Col- tively narrow thermal niches of tropical
lins 2013). The latest IPCC report predicted organisms that have evolved in climates ex-
that the only tropical forest regions that will hibiting low levels of variation across sea-
experience significant declines in total rain- sonal to millennial time scales (Ghalambor
fall are Central America and northern and et al. 2006). This narrow thermal tolerance,
eastern Amazonia. coupled with the fact that current tempera-
tures in the tropics are already near or at
thermally optimal conditions for many
organisms (Deutsch et al. 2008), creates a
DI R E CT IM PACTS
situation in which a moderate increase in
temperature could lead to a marked decline
Tolerance, Adaptation, and Acclimation
in fitness for many tropical organisms. The
Species at all latitudes will respond to warm- debate around this issue is a major source
ing through a combination of physiologi- of uncertainty in predicting how tropical
cal tolerance and acclimation, evolution- forest biodiversity will respond to different
ary adaptation, migration, and dispersal. scenarios of warming (Corlett 2011), and
Our understanding of individual species’ sources of uncertainty come from the de-
resilience to rapid climate change, created gree to which shifts in mean temperature
through these different response mecha- versus extreme temperatures will drive
nisms, is far from complete, and it is now changes in population fitness, and the im-
clear that it varies considerably across taxa portance of microclimatic refugia in buff-
(Corlett 2011; Malhi et al. 2014). In plants, ering populations from climate change (see
substantial evidence is emerging that tropi- Chapter 7 in this volume).
cal trees have significantly narrower ther-
mal niches than temperate trees (Araújo et
Movement
al. 2013), and there is some evidence that
tropical trees have lower temperature ac- Although many temperate species have ac-
climation capacities compared to temperate cess to both latitudinal and elevational tem-
trees (Corlett 2011). There is also broad- perature gradients, it is thought that more
based evidence for warming-induced slow- than 50 percent of animal and plant species
downs in tropical tree growth rates (Way in tropical regions have ranges so restricted
and Oren 2010). Projecting these impacts that they lack contact with a latitudinal
into the future is complicated by uncertain- temperature gradient (Wright, Muller-Lan-
ties surrounding the degree to which CO2 dau, and Schipper 2009). The narrow ther-
fertilization may increase tropical forest mal tolerance of tropical species, combined
productivity and potentially make up for with the lack of latitudinal temperature
heat-induced growth reductions (Pan et al. gradients, suggests that a primary response
2011). In animals, narrower thermal toler- for tropical species should be elevational
ances are well established in the scientific migration (Colwell et al. 2008). Some em-
literature, with increasingly strong links pirical evidence confirms this, with upslope
between thermal tolerance and elevational migration in response to climate change
range sizes (Sheldon and Tewksbury 2014). already more common in tropical than in
Evidence on acclimation is less strong for temperate communities (Freeman and Class
animals, but there is some suggestion that Freeman 2014). This pattern is likely to
tropical animal species may have lower ac- strengthen through time as climate change
202 W H AT DOE S THE FUTUR E HOL D?
intensifies, and yet there are multiple fac- ties, low tolerance to changing tempera-
tors that may limit the positive effects of tures, and the rapid pace of warming are
mountains as buffers against the loss of spe- predicted to lead to large tropical areas in
cies under climate change. which greater than 30 percent of mammals
First, high-elevation habitat is not exten- are unable to keep pace with climate change
sive across the tropics. Even in mountain- and will find themselves in very different
ous areas, the amount of area inevitably climate envelopes over the next 75 years
declines as altitude increases. As a result, from what they experience today (Schloss,
extrapolations of species-area relationships Nuñez, and Lawler 2012).
suggest a limited capacity for mountains to The degree to which microrefugia, be-
act as refugia for the many lower-elevation havior modification, adaptation, acclima-
species seeking to take advantage of eleva- tion, and tolerance can collectively ward off
tional gradients to escape the negative ef- attrition or outright loss of lowland tropical
fects of rising temperatures. Greater than 70 species is still a matter of debate. Past re-
percent of the terrestrial surface of the earth cords of high plant diversity during previous
is below 1,000 m, and thus the amount of warming periods suggest considerable resil-
land available at higher elevations is rela- ience, but the fact that the current warming
tively small. is orders of magnitude faster than previous
Second, in tropical regions, there is a warming events (Jaramillo et al. 2010), and
greater potential for biotic attrition (Col- is coming after a prolonged period of cooler
well et. al. 2008). In temperate zones, as tropical temperatures with little evolution-
species migrate to higher elevations, lower ary selection for extreme warm tolerance
elevations presumably become available for (Corlett 2011), suggests significant risk of
colonization by species from lower lati- large scale losses in biodiversity.
tudes, but in the tropics, there are no flo-
ral and faunal assemblage preadapted to
Interactions
landscapes warmer than the current tem-
peratures in lowland tropical areas (Feeley Multiple lines of evidence highlight the im-
and Silman 2010; Colwell et al. 2008). This portance of biotic interactions in creating
has led some researchers to conclude that and sustaining biodiversity across the trop-
large-scale migration upslope will lead to ics (Schemske et al. 2009). Understanding
a significant reduction in species in lower- these interactions and how they are affected
elevation landscapes (Colwell et al. 2008). by the different components of climate
Finally, in many lowland tropical ar- change will be critical for accurate pre-
eas, upslope migration may simply not be dictions of the impacts on biodiversity as
feasible because of the distance that tropi- whole (Gilman et al. 2010). Direct changes
cal fauna would need to travel. The pace of in biotic interactions will occur as a result
climate change suggests that many tropical of changes in direct temperature, precipita-
species—particularly those with limited tion, and CO2-dependent dominance rela-
dispersal capacities or difficulties cross- tionships, shifts in phenology, and changes
ing gaps—will be unable to migrate rap- in rainfall and temperature-dependent
idly enough to keep pace with the change mutualisms and in community-level inter-
(Schloss et al. 2012). Plant populations ap- actions. Movement in the face of climate
pear to track changes in climate less well change as described above will also create
than animal populations, as they frequently changes in biotic interactions. Because el-
display longer time lags and lack of response evational migration in response to climate
compared to animals (Corlett and Westcott change will be highly variable among spe-
2013). Even in more mobile animal groups, cies, and because tropical species have both
such as mammals, limited dispersal capaci- narrower thermal tolerances and narrower
T ROPICA L FOR E STS IN A CH A NGING CLIM ATE 203
elevational ranges than their temperate by climate change across tropical forest eco-
counterparts, it is believed that upslope systems in the ways outlined above, human
movement will result in significantly higher populations that occupy and are in many
rates of community disruption, loss of spe- cases dependent on them, are also adapting
cies interactions, and species extinction to the changing climatic conditions (Max-
in tropical communities than in subtropi- well et al. 2016). Whether planned or un-
cal and temperate communities (Urban, planned, these indirect impacts of climate
Tewksbury, and Sheldon 2012). change are thought to be increasingly af-
fecting many species and ecosystems (Segan
et al. 2015). Although there are many pos-
Ecosystems
sible human responses to climate change
It is well established that vegetation cover that could lead to indirect impacts, two
change in tropical forests affects both re- stand out as very likely to occur and have
gional and global precipitation patterns significant ramifications for tropical forest
by altering the amount of water returned ecosystems across the world.
to the atmosphere via evapotranspiration.
Therefore, tropical forest loss to date is
Increased Agriculture Movement into
thought to have already affected precipita-
Tropical Forests
tion patterns, with severe drought a sig-
nificant concern in many tropical areas. The conversion of natural ecosystems is the
This feedback, coupled with increases in greatest driver of biodiversity decline glob-
drought length and severity, particularly in ally, largely because tropical forest ecosys-
the eastern Amazon, can lead to increased tems are being rapidly replaced with crop-
fire risk and reduced forest cover. Because lands that support few species (Laurance,
few tropical plant species are adapted to Sayer, and Cassman 2014). Human popula-
fire, and most tropical rainforests are in fact tion growth and increased global demand
fire sensitive (Barlow and Peres 2008), the for tropical agricultural commodities are
synergy between droughts and more fre- trends that are likely to continue in many
quent fires significantly increases the risk regions and the dynamics of each are being
of a tropical forest dieback scenario, charac- affected by climate change. For example,
terized by large-scale replacement of forest increasingly severe dry seasons in tropical
with savanna or shrub (Brando et al. 2014). forest ecosystems are likely to exacerbate
The probability of such large-scale change, the impacts of changes in land use on tropi-
especially in the Amazon forests, has been cal forest biodiversity, as prolonged or more
the subject of considerable debate in the sci- intense dry periods may allow for increased
entific literature over the past decade. High accessibility, thus removing a current bar-
levels of uncertainty surround the potential rier to resource development (Brodie, Post,
role of CO2 fertilization, which could buf- and Laurance 2012). As a consequence of
fer tropical systems from droughts through this, the economic feasibility of forest colo-
increased productivity, but there is increas- nization and logging could increase, and
ing evidence for tipping points brought on many of the last remaining “remote” for-
by heat, drought, and fire (Wright, Muller- ests could become accessible to large-scale
Landau, and Schipper 2009). exploitation. There is evidence that sever-
ity of the dry season is already a strong,
positive predictor of deforestation pressure
I N DI R E CT T H R E ATS in the Amazon (Laurance et al. 2002), im-
plying that drying trends in certain tropi-
As the dominant conditions that drive spe- cal forests could increase their vulnerabil-
cies presence and abundance are reshaped ity. While increased drought frequency in
204 W H AT DOE S THE FUTUR E HOL D?
tropical forests may create economic op- example, as human populations increase
portunities for some, past droughts have and land use changes, drought cycles in In-
also had adverse impacts on local com- donesia become increasingly coupled with
munities through impaired water quality, fire cycles (Field, van der Werf, and Shen
fish die-offs, and hindered riverine trans- 2009). Moreover, logged and fragmented
port that local communities rely on to stay tropical forests are far more vulnerable to
connected to markets. All these lead to ad- fire than intact forests, because canopy loss
ditional changes as local communities at- leads to desiccation, which allows fire to
tempt to cope with their new environment penetrate more deeply into forest remnants.
(Marengo et al. 2008). The synergies between human activities
Societal responses to climate change are and their impact on tropical forests can also
also magnifying agricultural pressures on take surprising and unexpected forms. For
tropical forests. Driven by the need to miti- instance, research from the Amazon sug-
gate greenhouse gas emissions, crop-based gests that as humans vacate rural areas, the
biofuel production has increased rapidly in ability to control fires decreases, resulting
recent years, especially in Southeast Asia in overall increased fire risk (Schwartz et
(Fargione et al. 2008). Accompanied by in- al. 2015). The complex feedback loops be-
creased demand for food, this has led to a tween human activity and physical climate
substantial expansion of agricultural lands changes necessitate that planners directly
in all tropical forested regions to create new account for their interactions when assess-
areas for biofuel production or to provide ing risk or designing interventions.
replacement sites for food production when
existing croplands are switched to biofuel
production. Recent analysis of drivers of CONC LUSIONS: CONSE QU E NC E S
deforestation globally suggest that at least F OR CONSE RVATION
half of deforestation in the past decade was
the result of agricultural pressure (Lawson Climate change will directly influence
et al. 2014). Rising demands for land to biodiversity in tropical regions in myriad
grow biofuels is likely to increase oppor- ways. The narrow thermal niches of many
tunity costs for conservation, reducing the tropical species, combined with a general
competitiveness of carbon offsets and other lack of latitudinal temperature gradients,
payment for ecosystem service programs means that uphill movement is likely to
designed to slow forest destruction. be a common adaptive response in tropi-
cal forests. For those species that lack the
evolutionary adaptive response enabling
Increasing Threats from Changing
them to track a climate gradient, rapid cli-
Fire Regimes
mate change will clearly be a serious threat.
Synergies between climate change and hu- Even for those species that are able to track
man-lit fires represent another severe threat their climate via movement, their response
to tropical forest ecosystems that is likely to over time will lead to novel ecological in-
increase in importance with future warm- teractions that are likely to reshape tropical
ing. As substantial expanses of the tropics forest ecosystems. However, it is important
could become both warmer and drier in the that we recognize that the direct impacts of
future, there is likely to be an increase in climate change are particularly challenging
the incidence, magnitude, and duration of to predict and come with huge uncertainty
human-lit fires (Nepstad et al. 2008). Im- around how individual species will adapt
portantly, human changes to land use also to a changing climate, what this is likely
increases forest vulnerability to fire, mag- to mean for community composition and
nifying the impacts of climatic drying. For ecological interactions in different tropi-
T ROPICA L FOR E STS IN A CH A NGING CLIM ATE 205
cal regions, and the synergy between these R E FE R E NCE S
changes and other important ecological
processes that drive tropical forest function Araújo, Miguel B., Francisco Ferri-Yáñez, Francisco
Bozinovic, Pablo A. Marquet, Fernando Valladares,
(Scheffers et al. 2016). This lack of knowl- and Steven L. Chown. 2013. “Heat freezes niche evolu-
edge significantly affects our ability to as- tion.” Ecology Letters 16 (9): 1206–1219. https://doi.org/
sess species risk to climate change and the 10.1111/ele.12155.
most effective actions to avert this risk, and Barlow, Jos, and Carlos A. Peres. 2008. “Fire-mediated
at the same time may lead to both overes- dieback and compositional cascade in an Amazonian
forest.” Philosophical Transactions of the Royal Society B: Biologi-
timations and underestimations of the de- cal Sciences 363 (1498): 1787–1794. https://doi.org/10
gree of risk species face when it comes to .1098/rstb.2007.0013.
climate change. Brando, Paulo Monteiro, Jennifer K. Balch, Daniel C.
Our lack of ability to predict the direct Nepstad, Douglas C. Morton, Francis E. Putz, Michael
consequences of climate change should not T. Coe, Divino Silvério, et al. 2014. “Abrupt increases
in Amazonian tree mortality due to drought-fire in-
be leveraged as an excuse for inaction. We teractions.” Proceedings of the National Academy of Sciences
are far more certain in our ability to under- 111 (17): 6347–6352. https://doi.org/10.1073/pnas
stand and predict how human communities .1305499111.
will respond in tropical systems—and it is Brodie, Jedediah, Eric Post, and William F. Laurance.
human activities that are driving the cur- 2012. “Climate change and tropical biodiversity: A
new focus.” Trends in Ecology & Evolution 27 (3): 145–150.
rent extinction crisis we face. We can use https://doi.org/10.1016/j.tree.2011.09.008.
this information to assess species risk to Collins, M., R. Knutti, J. Arblaster, J.-L. Dufresne, T.
potential indirect affects to climate change Fichefet, P. Friedlingstein, X. Gao, W. J. Gutowski,
and identify suitable conservation strate- T. Johns, G. Krinner, M. Shongwe, C. Tebaldi, A. J.
gies. Importantly, many conservation strat- Weaver, and M. Wehner. 2013. “Long-term climate
change: Projections, commitments and irreversibil-
egies are already known to be effective in ity.” In Climate Change 2013: The Physical Science Basis. Con-
protecting biodiversity from human activi- tribution of Working Group I to the Fifth Assessment Report of the
ties, including well-managed protected ar- Intergovernmental Panel on Climate Change, ed. T. F. Stoker,
eas, fire regulation, and some payment-for- D. Qin, G.-K. Plattner, M. Tignor, S. K. Allen, J. Bosc-
ecosystem-services schemes like REDD+. hung, A. Nauels, Y. Xia, V. Bex, and P. M. Midgley,
1029–1136. Cambridge University Press.
All of these can be used in ways that benefit Colwell, Robert K., Gunnar Brehm, Catherine L. Cardelus,
climate and help species overcome many Alex C. Gilman, and John T. Longino. 2008. “Global
of the short-term challenges we have out- warming, elevational range shifts, and lowland biotic
lined in this chapter. With the large amount attrition in the wet tropics.” Science 322: 258–261.
of uncertainty around species response to Corlett, Richard T. 2011. “Impacts of warming on tropical
lowland rainforests.” Trends in Ecology & Evolution 26 (11):
climate change, there are two clear over- 606–613. https://doi.org/10.1016/j.tree.2011.06.015.
arching conservation priorities for tropical Corlett, Richard T., and Richard B. Primack. 2006.
biodiversity. First, global mitigation efforts “Tropical rainforests and the need for cross-conti-
that limit the extent of climate change will nental comparisons.” Trends in Ecology & Evolution 21 (2):
be critical for species with few adaptation 104–110. https://doi.org/10.1016/j.tree.2005.12.002.
Corlett, Richard T., and David A. Westcott. 2013. “Will
options in the short term and all species plant movements keep up with climate change?” Trends
over the long term. Second, it is necessary in Ecology and Evolution 28 (8): 482–488. https://doi.org/
to prevent large-scale land clearing, wher- 10.1016/j.tree.2013.04.003.
ever possible, in intact ecosystems, espe- Costa, Antonio Carlos Lola da, David Galbraith, Samuel
cially those with latitudinal and elevational Almeida, Bruno Takeshi Tanaka Portela, Mauricio da
Costa, João de Athaydes Silva Junior, Alan P. Braga, et
gradients (Watson et al. 2018). By protect- al. 2010. “Effect of 7 yr of experimental drought on
ing large, functional tropical forest ecosys- vegetation dynamics and biomass storage of an eastern
tems, we will ensure that tropical species Amazonian rainforest.” New Phytologist 187 (3): 579–591.
can maximize their adaptive capacity to https://doi.org/10.1111/j.1469-8137.2010.03309.x.
overcome the challenges posed by future Davin, Edouard L., and Nathalie de Noblet-Ducoudré.
2010. “Climatic impact of global-scale deforestation:
climate change.
206 W H AT DOE S THE FUTUR E HOL D?
Radiative versus nonradiative processes.” Journal Jaramillo, Carlos, Diana Ochoa, Lineth Contreras, Mark
of Climate 23 (1): 97–112. https://doi.org/10.1175/ Pagani, Humberto Carvajal-Ortiz, Lisa M. Pratt, Sri-
2009JCLI3102.1. nath Krishnan, et al. 2010. “Effects of rapid global
Deutsch, Curtis A., Joshua J. Tewksbury, Raymond B. warming at the Paleocene-Eocene boundary on
Huey, Kimberly S. Sheldon, Cameron K. Ghalambor, neotropical vegetation.” Science 330 (6006): 957–961.
David C. Haak, and Paul R. Martin. 2008. “Impacts https://doi.org/10.1126/science.1193833.
of climate warming on terrestrial ectotherms across Laurance, William F., Thomas E. Lovejoy, Heraldo L. Vas-
latitude.” Proceedings of the National Academy of Sciences concelos, Emilio M. Bruna, Raphael K. Didham, Philip
105 (18): 6668–6672. https://doi.org/10.1073/pnas C. Stouffer, Claude Gascon, Richard O. Bierregaard,
.0709472105. Susan G. Laurance, and Erica Sampaio. 2002. “Eco-
Dirzo, Rodolfo, and Peter H. Raven. 2003. “Global state system decay of Amazonian forest fragments: A 22-
of biodiversity and loss.” Annual Review of Environment year investigation.” Conservation Biology 16 (3): 605–618.
and Resources 28 (1): 137–167. https://doi.org/10.1146/ https://doi.org/10.1046/j.1523-1739.2002.01025.x.
annurev.energy.28.050302.105532. Laurance, William F., Jeffrey Sayer, and Kenneth G. Cass-
Fargione, Joseph, Jason Hill, David Tilman, Stephen man. 2014. “Agricultural expansion and its impacts
Polasky, and Peter Hawthorne. 2008. “Land clear- on tropical nature.” Trends in Ecology & Evolution 29 (2):
ing and the biofuel carbon debt.” Science 319 (5867): 107–116. https://doi.org/10.1016/j.tree.2013.12.001.
1235–1238. https://doi.org/10.1126/science.1152747. Lawson, Sam, A. Blundell, Bruce Cabarle, Naomi Basik,
Feeley, Kenneth J., and Miles R. Silman. 2010. “Biotic Michael Jenkins, and Kerstin Canby. 2014. Consumer
attrition from tropical forests correcting for trun- Goods and Deforestation: An Analysis of the Extent and Nature of
cated temperature niches.” Global Change Biology 16: Illegality in Forest Conversion for Agriculture and Timber Planta-
1830–1836. tions. Forest Trends.
Feng, Xue, Amilcare Porporato, and Ignacio Rodriguez- Mackey, Brendan, Dominick A. DellaSala, Cyril Kormos,
Iturbe. 2013. “Changes in rainfall seasonality in the David Lindenmayer, Noelle Kumpel, Barbara Zim-
tropics.” Nature Climate Change 3 (9): 811–815. https:// merman, Sonia Hugh, et al. 2014. “Policy options for
doi.org/10.1038/nclimate1907. the world’s primary forests in multilateral environ-
Field, Robert D., Guido R. van der Werf, and Samuel S. mental agreements.” Conservation Letters 8 (2): 139–147.
P. Shen. 2009. “Human amplification of drought-in- https://doi.org/10.1111/conl.12120.
duced biomass burning in Indonesia since 1960.” Na- Malhi, Yadvinder. 2012. “The productivity, metabolism
ture Geoscience 2 (3): 185–188. https://doi.org/10.1038/ and carbon cycle of tropical forest vegetation.” Jour-
ngeo443. nal of Ecology 100 (1): 65–75. https://doi.org/10.1111/j
Freeman, Benjamin G, and Alexandra M. Class Freeman. .1365-2745.2011.01916.x.
2014. “Rapid upslope shifts in New Guinean birds Malhi, Yadvinder, Luiz E. O. C. Aragão, David Gal-
illustrate strong distributional responses of tropical braith, Chris Huntingford, Rosie Fisher, Przemyslaw
montane species to global warming.” Proceedings of the Zelazowski, Stephen Sitch, Carol McSweeney, and
National Academy of Sciences 111 (12): 4490–4494. https:// Patrick Meir. 2009. “Exploring the likelihood and
doi.org/10.1073/pnas.1318190111. mechanism of a climate-change-induced dieback of
Ghalambor, Cameron K., Raymond B. Huey, Paul R. the Amazon rainforest.” Proceedings of the National Acad-
Martin, Joshua J. Tewksbury, and George Wang. 2006. emy of Sciences (February). https://doi.org/10.1073/pnas
“Are mountain passes higher in the tropics? Janzen’s .0804619106.
hypothesis revisited.” Integrative and Comparative Biology 46 Malhi, Yadvinder, Toby A. Gardner, Gregory R. Gold-
(1): 5–17. https://doi.org/10.1093/icb/icj003. smith, Miles R. Silman, and Przemyslaw Zelazowski.
Gilman, S. E., M. C. Urban, J. Tewksbury, G. W. Gilchrist, 2014. “Tropical forests in the Anthropocene.” Annual
and R. D. Holt. 2010. “A framework for community Review of Environment and Resources 39 (1): 125–159. https://
interactions under climate change.” Trends in Ecology and doi.org/10.1146/annurev-environ-030713-155141.
Evolution 25: 325–331. https://doi.org/10.1016/j.tree Maxwell, S. L., O. Venter, K. R. Jones, and J. E. M. Wat-
.2010.03.002. son. 2015. “Integrating human responses to climate
Groot, Rudolf de, Luke Brander, Sander van der Ploeg, change into conservation vulnerability assessments
Robert Costanza, Florence Bernard, Leon Braat, Mike and adaptation planning.” Annals of the New York Academy
Christie, et al. 2012. “Global estimates of the value of Sciences 1355 (1): 98–116. https://doi.org/10.1111/
of ecosystems and their services in monetary units.” nyas.12952.
Ecosystem Services 1 (1): 50–61. https://doi.org/10.1016/ Marengo, J. A., C. A. Nobre, J. Tomasella, M. F. Cardoso,
j.ecoser.2012.07.005. and M. D. Oyama. 2008. “Hydro-climatic and ecolog-
IPCC. 2013. Climate Change 2013: The Physical Science Basis: ical behaviour of the drought of Amazonia in 2005.”
Working Group I Contribution to the Fifth Assessment Report of Philosophical Transactions of the Royal Society B: Biological Sciences
the Intergovernmental Panel on Climate Change. International 363 (1498): 1773–1778. https://doi.org/10.1098/rstb
Panel on Climate Change. .2007.0015.
T ROPICA L FOR E STS IN A CH A NGING CLIM ATE 207
Mora, Camilo, Abby G. Frazier, Ryan J. Longman, Ra- Segan, D. B., D. G. Hole, C. I. Donatti, C. Zganjar, S.
chel S. Dacks, Maya M. Walton, Eric J. Tong, Joseph J. Martin, S. H. M. Butchart, and J. E. M. Watson. 2015.
Sanchez, et al. 2013. “The projected timing of climate “Considering the impact of climate change on human
departure from recent variability.” Nature 502 (7470): communities significantly alters the outcome of spe-
183–187. https://doi.org/10.1038/nature12540. cies and site-based vulnerability assessments.” Diversity
Nepstad, Daniel C., Claudia M. Stickler, Britaldo Soares- and Distributions 21 (9): 1101–1111. https://doi.org/10
Filho, and Frank Merry. 2008. “Interactions among .1111/ddi.12355.
Amazon land use, forests and climate: Prospects for a Sheldon, Kimberly S., and Joshua J. Tewksbury. 2014.
near-term forest tipping point.” Philosophical Transactions “The impact of seasonality in temperature on thermal
of the Royal Society B: Biological Sciences 363 (1498): 1737– tolerance and elevational range size.” Ecology 95 (8):
1746. https://doi.org/10.1098/rstb.2007.0036. 2134–2143.
Pan, Yude, Richard A. Birdsey, Jingyun Fang, Richard Somero, G. N. 2010. “The physiology of climate change:
Houghton, Pekka E. Kauppi, Werner A. Kurz, Oliver How potentials for acclimatization and genetic adap-
L. Phillips, et al. 2011. “A large and persistent carbon tation will determine ‘winners’ and ‘losers.’” Journal of
sink in the world’s forests.” Science 333 (6045): 988– Experimental Biology 213 (6): 912–920. https://doi.org/10
993. https://doi.org/10.1126/science.1201609. .1242/jeb.037473.
Phillips, Oliver L., Luiz E. O. C. Aragao, Simon L. Lewis, Urban, Mark C., Josh J. Tewksbury, and Kimberly S.
Joshua B. Fisher, Jon Lloyd, Gabriela Lopez-Gonzalez, Sheldon. 2012. “On a collision course: Competition
Yadvinder Malhi, et al. 2009. “Drought sensitivity of and dispersal differences create no-analogue commu-
the Amazon rainforest.” Science 323 (5919): 1344–1347. nities and cause extinctions during climate change.”
https://doi.org/10.1126/science.1164033. Proceedings of the Royal Society B: Biological Sciences 279 (1735):
Scharlemann, Joern P. W., Valerie Kapos, Alison Camp- 2072–2080.
bell, Igor Lysenko, Neil D. Burgess, Matthew C. Han- Watson, J. E. M., T. Iwamura, and N. Butt. 2013. “Map-
sen, Holly K. Gibbs, Barney Dickson, and Lera Miles. ping vulnerability and conservation adaptation strate-
2010. “Securing tropical forest carbon: The contribu- gies under climate change.” Nature Climate Change 3 (11):
tion of protected areas to REDD.” Oryx 44 (3): 352–357. 989–994. https://doi.org/10.1038/nclimate2007.
Scheffers, B. R., L. De Meester, T. C. L. Bridge, A. A. Hoff- Watson, James E. M., Tom Evans, Oscar Venter, Brooke
mann, J. M. Pandolfi, R. T. Corlett, S. H. M. Butchart, Williams, Ayesha Tulloch, Claire Stewart, Ian Thomp-
et al. 2016. “The broad footprint of climate change son, et al. 2018. “The exceptional value of intact forest
from genes to biomes to people.” Science 354 (6313). ecosystems.” Nature Ecology & Evolution 2: 599–610.
https://doi.org/10.1126/science.aaf7671. Way, Danielle A., and Ram Oren. 2010. “Differential re-
Schemske, Douglas W., Gary G. Mittelbach, Howard V. sponses to changes in growth temperature between
Cornell, James M. Sobel, and Kaustuv Roy. 2009. “Is trees from different functional groups and biomes:
there a latitudinal gradient in the importance of biotic A review and synthesis of data.” Tree Physiology 30 (6):
interactions?” Annual Review of Ecology, Evolution and System- 669–688. https://doi.org/10.1093/treephys/tpq015.
atics 40 (1): 245–269. https://doi.org/10.1146/annurev Wright, S. Joseph, Helene C. Muller-Landau, and Jan
.ecolsys.39.110707.173430. Schipper. 2009. “The future of tropical species on a
Schloss, Carrie A., Tristan A. Nuñez, and Joshua J. Lawler. warmer planet.” Conservervation Biology 23 (6): 1418–1426.
2012. “Dispersal will limit ability of mammals to track https://doi.org/10.1111/j.1523-1739.2009.01337.x.
climate change in the Western Hemisphere.” Proceedings Zelazowski, Przemyslaw, Yadvinder Malhi, Chris Hunt-
of the National Academy of Sciences 109 (22): 8606–8611. ingford, Stephen Sitch, and Joshua B. Fisher. 2011.
Schwartz, Naomi B., Maria Uriarte, Victor H. Gutiér- “Changes in the potential distribution of humid tropi-
rez-Vélez, Walter Baethgen, Ruth DeFries, Katia Fer- cal forests on a warmer planet.” Philosophical Transactions
nandes, and Miguel A. Pinedo-Vasquez. 2015. “Cli- of the Royal Society A: Mathematical, Physical and Engineering
mate, landowner residency, and land cover predict Sciences 369 (1934): 137–160. https://doi.org/10.1098/
local scale fire activity in the western Amazon.” Global rsta.2010.0238.
Environmental Change 31 (March): 144–153. https://doi
.org/10.1016/j.gloenvcha.2015.01.009.
CASE STUDY 6
The Amazon forest is a giant from any perspective. It is the planet’s largest tropical for-
est drained by the largest river—the source of one-fifth of all freshwater that reaches the
oceans. It is the greatest cornucopia of biological and indigenous cultural diversity.
The Amazon forest is also a giant in Amazon rainfall system itself depends
shaping the Earth’s climate. Like an enor- on the forest. The same year-round forest
mous planetary cooling system, it influ- evapotranspiration that shapes global circu-
ences the global circulation of air and vapor lation patterns sustains rainfall patterns in
by evaporating vast amounts of water into the Amazon region (Salati and Vose 1984).
the atmosphere—converting the equato- Simply stated, the large-scale conversion
rial Sun’s intense radiant energy into latent of forest to pasture and cropland along the
heat. Even where long dry seasons are the eastern flank of the Amazon forest means
norm, the deep root systems of Amazon that there is less vapor to supply rainfall sys-
trees allow the forest to absorb soil water tems to the west and southwest. A potential
stored many meters beneath the soil surface “tipping point” could be reached if forest
to supply the leaf canopy high above the clearing becomes extensive enough to sup-
ground (Nepstad et al. 1994). When patches press rainfall below the minimum amount
of Amazon forest are replaced by cattle pas- that is necessary to sustain closed-canopy
ture, the amount of vapor pumped into the forests (Lovejoy and Nobre 2018).
atmosphere declines. A second positive feedback is between
The wood of Amazon forest trees is an- forest degradation and forest fire: once a
other reason this ecosystem shapes global forest burns, the chances of subsequent
climate. An amount of carbon equivalent to fires increase. In pre-Columbian times,
nearly a decade’s worth of global emissions Amazon forests appear to have caught fire
from human activities is stored in Amazon a few times every millennium, when se-
wood—that is, outside of the atmosphere, vere droughts occurred. Then, as now, dur-
where it would contribute to global warm- ing years of normal rainfall, most of the
ing. This biological carbon leaks into the forests of the Amazon extended like giant
atmosphere during and after droughts that firebreaks across the landscape; the dense
are severe enough to kill big Amazon trees, shade of the lofty tree canopies supplied
such as the droughts of 2005, 2007, 2010, with water by deep root systems kept the
and 2016. It also leaks out when humans forest floor too damp to carry a fire. Today,
deliberately fell the forest to grow crops or forest fire is a frequent occurrence along the
livestock, when they degrade the forest to agricultural and livestock frontiers partly
harvest timber, or when the fires they set because forests have become more suscep-
to manage the land escape into neighboring tible to fire, either because they were al-
forests, killing trees (Nepstad et al. 2008). ready subjected to an earlier fire or through
Positive feedbacks in the Amazon forest- logging operations that punch holes in the
rainfall-fire system could be important in forest canopy, allowing light to penetrate to
determining the future of this ecosystem
in the face of climate change and further Copyright © 2019 by Yale University. All rights reserved.
expansion of land-use activities. First, the ISBN 978-0-300-20611-1.
208
POSTPONING THE AMAZON TIPPING POINT 209
the forest floor. They have also increased in fire increases. Major droughts continue to
frequency because there are more ignition become more frequent and severe, increas-
sources—more people using fire to clear ing the occurrence of forest fires in vast
the land and kill shrubs and trees that in- swaths of primary and degraded forests.
vade cattle pastures. If Amazon forests burn The expansion of forage grasses, crop fields,
repeatedly, they can eventually be invaded and scrub vegetation further inhibits rain-
by flammable grasses and herbs, further in- fall. The “scrubification” of Amazon forests
creasing the likelihood of recurring fire— that have been degraded by drought, log-
in a second tipping point (Balch et al. 2015). ging, repeated burning, and invasion by
With our knowledge of Amazon forest highly flammable grasses and fires begins to
feedbacks and tipping points we can assem- appear across the eastern and southern Am-
ble two plausible scenarios for the future of azon regions (Figure CS6.1)—this is accel-
this globally significant ecosystem—one in erated when deforestation frontiers expand
which the Amazon forest becomes a major along the newly paved highways from Porto
source of added carbon emissions to the at-
mosphere as the forest-rainfall-fire system Figure CS6.1. The forests of the Amazon Basin are being
drives regional degradation. In a second, altered through severe droughts, land use deforestation,
the Amazon forest becomes a major sink of logging, and increased frequencies of forest fire. Some
atmospheric carbon, figuring prominently of these processes are self-reinforcing through positive
feedbacks and create the potential for a large-scale tip-
in our success in keeping the planet below ping point. For example, forest fire kills trees, increas-
a 1.5°C average temperature increase, as ex- ing the likelihood of subsequent burning. This effect is
pressed in the 2015 Paris Agreement. magnified when tree death allows forests to be invaded
In the first “business as usual” future by flammable grasses. Deforestation provides ignition
scenario, deforestation accelerates, revers- sources to flammable forests, contributing to this die-
back. Climate change contributes to this tipping point by
ing the recent slowdown in deforestation of increasing drought severity, reducing rainfall and rais-
the Brazilian Amazon (Nepstad et al. 2014), ing air temperatures, particularly in the eastern Amazon
as forest degradation through logging and Basin. (Figure 4-8 from Settele et al. 2014.)
210 W H AT DOE S THE FUTUR E HOL D?
Velho to Manaus, Brazil; from Santarem to cieties secure and deepen the innovations
Cachimbo, Brazil; and from Pucallpa, Peru, in public policies, law enforcement, and
to Cruzeiro do Sul, Brazil. agricultural innovation that already avoided
In a scenario of “managed resilience,” more than 6 billion tons of carbon dioxide
the slowdown in deforestation that has been emissions from deforestation in Brazil. In
taking place in the Brazilian Amazon con- the long term, however, the ecological in-
tinues to deepen and spreads to the other tegrity of this giant ecosystem will depend
Amazon forest nations. Logging continues on humanity’s success in slowing climate
but through reduced-impact practices; for- change.
est fire–control programs established across
the Amazon region successfully extinguish
forest fires soon after they ignite, and the R E FE R E NCE S
number of fires needing extinguishing de-
clines through better fire management. This Balch, J., P. Brando, D. Nepstad, et al. 2015. “Suscepti-
bility of southeastern Amazon forests to fire: Insights
fire prevention program has also allowed
from a large-scale burn experiment.” Bioscience 65 (9):
forest recovery to take place on nearly half 893–905.
of the forestlands that had been cleared for Griscom, B. W., J. Adams, P. Ellis, R. A. Houghton, G.
livestock and crops but were then aban- Lomax, D. A. Miteva, W. H. Schlesinger, D. Shoch, J. V.
doned because of inadequate soils or infra- Siikamäki, P. Smith, and P. Woodbury. 2017. “Natural
climate solutions.” Proceedings of the National Academy of Sci-
structure. This regional forest recovery re-
ences 114 (44): 11645–11650.
moves carbon dioxide from the atmosphere Lovejoy, T. E., and C. Nobre. 2018. “Amazon tipping
as it reestablishes the year-round supply of point.” Science Advances 4 (2): eaat2340.
water vapor to the atmosphere that sustains Nepstad, D., G. Carvalho, A. C. Barros, A. Alencar, J. P.
the rainfall system of the Amazon and that Capobianco, J. Bishop, P. Moutinho, P. Lefebvre, and
U. L. Silva Jr. 2001. “Road paving, fire regime feed-
shapes global circulation patterns.
backs, and the future of Amazon forests.” Forest Ecology
The Amazon forest can be part of the and Management 154: 395–407.
problem of climate change—exacerbating Nepstad, D., C. Reis de Carvalho, E. Davidson, P. Jipp,
global warming—or it can be part of the P. Lefebvre, G. Hees Negreiros, E. Silva, T. Stone, S.
solution. Tropical forests could provide a Trumbore, and S. Vieira. 1994. “The role of deep roots
in the hydrologic and carbon cycles of Amazonian
quarter or more of the emissions reductions
forests and pastures.” Nature 372: 666–669.
needed by 2030 to avoid a 2°C increase in Nepstad, D, C. M. Stickler, B. Soares-Filho, and F. Merry.
global temperature (Griscom et al. 2017). 2008. “Interactions among Amazon land use, forests
What is remarkable is that the managed and climate: Prospects for a near-term forest tipping
resilience scenario is within reach—it is point.” Philosophical Transactions of the Royal Society B—Bio-
logical Sciences. https://doi.org/10.1098/rstb.2007.0036.
a viable choice that can be achieved with
Nepstad, D., D. McGrath, C. Stickler, et al. 2014. “Slow-
political will, the right market signals, and ing Amazon deforestation through public policies and
the right types of financial investments and interventions in beef and soy supply chains.” Science
incentives for landholders. The ecological 344: 1118–1123.
and climatic integrity of the Amazon can Salati, E., and P. Vose. 1984. “The Amazon: A system in
equilibrium.” Science 225 (4648): 129–138.
be maintained for decades if Amazon so-
CHAPTER SIXTEEN OV E RV IE W
211
212 W H AT DOE S THE FUTUR E HOL D?
events have increased—these trends are temperature tolerances found in the studies
expected to continue (Melillo, Richmond, are likely overestimates, because they are
and Yohe 2014). based on thermal safety margins derived
from air temperatures, which do not reflect
the body temperatures actually experienced
by organisms. Indeed, an analysis based on
SE NSITI V IT Y TO C L IM AT E C H A NGE
estimated ectotherm body temperatures re-
A N D VA R I A BIL IT Y
veals that thermal stress events can occur at
high latitudes largely as a result of radiation
Physiological Sensitivities
spikes, which provides an explanation for
to Climate Change
the latitudinal invariance of heat tolerance
Organisms in more poleward ecosystems (Sunday et al. 2014). Studies subsequent to
have generally been thought to be more Deutsch et al. (2008) have also pointed out
vulnerable to climate change because that the initial analysis based on mean ther-
they are expected to experience a greater mal conditions may have underestimated
magnitude of temperature change (Par- the fitness detriments associated with ther-
mesan 2006). However, recent studies mal stress events in temperate areas. King-
highlight the importance of considering solver et al. (2013) predicted that climate
how seasonality influences physiological change impacts would be most severe for
performance. These studies also illustrate insects at mid-latitudes (20°–40°) due to an
the complexities involved in predicting increased frequency of heat stress events,
vulnerability. For example, insects tend which reduce fitness. However, capitalizing
to evolve more specialized thermal physi- on an extended growing season may enable
ology in constant tropical environments organisms to counter some of the projected
relative to seasonally variable temperate fitness loss. Similarly, Vasseur et al. (2014)
(and boreal) environments (Deutsch et al. found that fitness detriments associated
2008). Compilations of thermal tolerance with anticipated increases in the incidence
data for both ectotherms (Sunday et al. of extreme events will outweigh fitness
2014) and endotherms (Khaliq et al. 2014) increases associated with shifts in mean
confirm that the breadth of thermal toler- temperatures and make temperate and bo-
ances increases with latitude. Declines in real organisms most vulnerable to climate
lower thermal limits with latitude and in- change. These results suggest that behav-
variant upper thermal limits tend to pro- ioral buffering will be essential to ectother-
duce this pattern, which yields increases mic organisms across latitudes to alleviate
in thermal safety margins (the difference the impacts of climate change (Sunday et al.
between heat tolerance and habitat air tem- 2014).
peratures) with latitude. An analysis based Studies on plants show some similar
on this relationship between temperature trends and complexities. A recent meta-
and fitness suggested that, as a result of analysis (Way and Oren 2010), for example,
these physiological differences, climate showed that warmer temperatures are ex-
change will decrease the fitness of tropical pected to increase plant performance (mea-
ectotherms while increasing the fitness of sured as growth or physiological responses)
temperate and boreal ectotherms (Deutsch to a greater extent in temperate and boreal
et al. 2008). ecosystems than in tropical forests, where
However, studies using similar ap- growth declines with increased warm-
proaches have cautioned against conclud- ing was often seen. Moreover, growth re-
ing that temperate and boreal organisms sponses to increased temperatures were less
are less sensitive to climate change. For one, variable for tropical than temperate trees,
the latitudinal gradients in the breadth of suggesting that tropical species have nar-
TEMPER ATE A ND BOR E A L R E SPONSE S TO CLIM ATE CH A NGE 213
rower thermal tolerances. This meta-analy- pathogens (see the section “Interacting and
sis also suggested that deciduous trees were Indirect Stressors” below). A key response
more stimulated by increased temperatures among organisms that are dormant during
than are evergreen trees. Finally, other the winter will be shifts in energy use. For
studies demonstrate that water availabil- endothermic animals below their thermal
ity (influenced by temperature) will play a neutral zone, warm temperatures will re-
large role in determining the sensitivity of duce the energy required to maintain stable
tree species to increased temperatures, with body temperatures. Ectothermic animals
increases in growth primarily expected in will experience increased metabolic rates,
locations where water availability does not and the higher energy use may threaten
decline. depletion of overwintering energy reserves,
Several factors complicate this simple and have been broadly observed as warm-
prediction and are likely to lead to com- ing extends the growing season (Boisvenue
plex biome shifts at tree line. First, the pri- and Running 2006). However, productivity
mary constraint on tree growth at some increases associated with the longer grow-
tree lines (both altitudinal and latitudinal) ing season may be countered by thermal
is likely to be water availability, not grow- and drought stress in the summer and in-
ing season length, which implies that bi- creased respiration in the fall. Similarly, cli-
ome shifts will depend on changes in both mate change impacts on disturbance (e.g.,
temperature and water availability (Littell, fire regimes, pests) may also limit the in-
Peterson, and Tjoelker 2008). Indeed, a re- creased productivity we would expect from
cent meta-analysis found only 52 percent of warming-induced increases in growing
monitored tree lines showed a significant season length (Kurz et al. 2008), although
advance, despite recent warming (Harsch the movement of less flammable temperate
et al. 2009). A second factor constraining species into boreal forests may offset poten-
tree-line shifts is dispersal limitation, espe- tial wildfire increases in response to warm-
cially latitudinally (Aitken et al. 2008). Tree ing (Terrier et al. 2013).
cover is likely to increase in grasslands and The effects of climate change on temper-
shrublands where moisture is not limit- ate and boreal biomes and disturbance re-
ing (Grimm et al. 2013). Other, moisture- gimes may also result in positive feedbacks
limited temperate woodlands may become to the climate system. Shifts in vegetation
grasslands (Breshears et al. 2005). types induced by climate change could am-
Temperate tree species are also expected plify local warming through albedo and
to shift their ranges poleward into boreal transpiration effects, as is predicted for
forests in response to warming. Increases boreal regions (Bonan and Pollard 1992).
in the recruitment of temperate seedlings Climate change–induced changes to dis-
and saplings and decreases in recruitment turbance regimes (e.g., fire, pests) may also
of boreal species have been observed as bo- have consequences for carbon dynamics
real forests warm (Fisichelli, Frelich, and (Running 2008). For example, warming-
Reich 2013), but altered trophic interactions induced changes to the life cycle of the
may constrain the expansion of temperate mountain pine beetle Dendroctonus ponderosae
forests (Frelich et al. 2012). Shifts in the lati- caused a region in British Columbia to go
tude and elevation of temperate and boreal from being a carbon sink to a carbon source
forests are widely anticipated (Grimm et al. (Kurz et al. 2008).
2013; Melillo, Richmond, and Yohe 2014),
and the ecotone between temperate and
Interacting and Indirect Stressors
boreal forests has already shifted upward
and Feedbacks
in some regions (Beckage et al. 2008). Ex-
perimental warming conducted over three The responses to climate change discussed
growing seasons found that photosynthesis here will interact, often synergistically,
and growth were reduced for boreal tree with changing natural disturbance re-
species near their warm-edge range limit gimes. Climate change is expected to shift
but enhanced for temperate species near the frequency, intensity, duration, and tim-
their cold range limit (Reich et al. 2015). ing of natural disturbances, which shape
This study suggests that leaf-level responses temperate and boreal forests. These natu-
of photosynthesis and respiration will be ral disturbances include both abiotic (e.g.,
indicative of whole-plant responses. fire, drought, storms) and biotic (e.g., in-
Ecosystem changes associated with veg- troduced species, insect and pathogen out-
etation shifts will also likely affect biodi- breaks) factors and may be further exacer-
versity. Productivity increases are expected bated by human stressors such as changes
TEMPER ATE A ND BOR E A L R E SPONSE S TO CLIM ATE CH A NGE 217
in land use (Dale et al. 2001). For example, this synchrony in many systems with var-
wildfire is central to the dynamics of tem- ied outcomes, as described below.
perate and boreal forests because it drives
tree mortality, successional dynamics, nu-
SH I F T S I N SE A SONA L I T Y
trient cycling, and hydrology (Dale et al.
2001). Warmer spring and summer temper- Increases in parasite growth and repro-
atures and earlier spring snowmelt are ex- duction may accelerate their transmission
pected to increase fire frequency and inten- cycles such that outbreaks occurring ev-
sity along with extending wildfire seasons ery several years may become an annual
(Westerling et al. 2006). Climate change is occurrence (Figure 16.3A). Mild winters
also likely to increase the severity and du- may enable overwinter survival of para-
ration of drought, which can combine, of- sites and extend the transmission season
ten synergistically, with other physiological (Figure 16.3B). However, warming may
stresses such as increased temperature to lead summer temperatures to exceed the
increase tree mortality above baseline lev- upper thermal tolerance of parasites and
els. This tree mortality can in turn facilitate depress transmission in summer (Figure
insect and pathogen outbreaks (Allen et al. 16.3C). Shifts in seasonality may also result
2010). Some widespread tree die-offs, such from shifts in host behavior. Milder win-
as pine mortality spanning over a million ters may drive organisms that previously
hectares in the southwestern United States, underwent long-distance migrations to re-
have already been linked to anomalous main resident, potentially elongating the
droughts associated with climate change transmission season and leading to parasite
(Breshears et al. 2005). accumulation. For example, monarch but-
Climate change can also have direct im-
pacts on diseases through the temperature Figure 16.3. The seasonal dynamics of disease transmis-
dependence of host immunity and pest sion may shift in numerous ways following warming
depending on thermal tolerance and phenological cues.
and parasite growth and reproduction (re- (A) Warming may increase rates of growth and develop-
viewed in Altizer et al. 2013). These two ment and accelerate semi-annual cycles to annual cycles.
processes may either increase or decrease (B) Warmed summer temperatures may exceed the
the incidence or severity of a disease de- thermal tolerance of pathogens, shifting the peak trans-
pending on how the thermal optimum of mission season to the spring and fall. (C) Warming may
enable overwinter survival, leading to year-round trans-
the parasites and their hosts compares to mission. (D) Warming may drive phenological shifts,
baseline environmental temperatures and which could disrupt synchrony with hosts and expose
those following climate change. The rela- more sensitive life stages to transmission.
tive shifts in host immunity and parasite
growth and reproduction will determine
disease dynamics. Rates will be influenced
both by shifts in mean environmental con-
ditions and increases in climate variability,
which will tend to decrease host immunity
as a result of climate warming (Altizer et
al. 2013).
Climate change will also alter the sea-
sonality of immunity and transmission
(Figure 16.3). Temperature seasonality in
temperate and boreal systems drives evolu-
tion to synchronize the population growth
and virulence of parasites with the suscep-
tibility of hosts. Climate change will disrupt
218 W H AT DOE S THE FUTUR E HOL D?
terfly (Danaus plexippus) populations that have r Broad thermal tolerances associated
become year-round residents of the United with seasonal climates are likely to
States as a result of milder winters exhibit allow many temperate and boreal
higher rates of parasite infection than mi- species to benefit from warming,
grating populations (reviewed in Altizer et especially if they are able to avoid
al. 2013). Conversely, augmented host mi- thermal or drought stress events (for
gration or range shifts may enable release animals and plants, respectively). In
from parasites. all, we expect fewer negative di-
rect responses to climate change for
temperate and boreal species than for
SH I F T S I N SY NCH RON Y
tropical species.
Hosts and parasites may differ in their phe-
nological responses to temperature and r Impacts on biodiversity in temper-
photoperiod, shifting synchrony following ate and boreal regions are likely to
climate change (Figure 16.3D). For example, be mediated by changes to season-
mild winters can increase the synchrony of ality and temperature variability.
Lyme disease vectors (ticks) and their hosts Additionally, negative biodiversity
(mice) and increase disease incidence (Ost- consequences resulting from direct
feld 2010). Another mechanism by which responses of species to climate change
climate change can increase disease trans- may stem primarily from the loss of
mission is exposing a more sensitive life synchrony with interacting species
stage to parasites. and the disruption of communities
An increased prevalence of insect pests rather than a decline in population
will be another important form of eco- viability.
logical disturbance. Recent temperature in-
creases have enabled bark beetles in North r Many climate change impacts driven
American temperate and boreal forests to by seasonality and temperature vari-
survive winter, complete additional gen- ability will occur in winter rather
erations in a season, and expand their dis- than the more thoroughly studied
tributions; these shifts have expanded the growing season.
areas of recent infestations beyond those
observed over the past 125 years (Raffa et r Impacts on diversity and ecosystem
al. 2008). Climate change will continue to processes will likely be driven by eco-
redistribute insect pests and enable them to logical disturbances associated with
invade new habitats and forest types (Lo- climate change and their interaction
gan, Regniere, and Powell 2003). with warming rather than warm-
ing directly. Specifically, increases in
the incidence of disease, insect pests,
CONC LUSIONS : BIODI V E R SIT Y and drought may lead to biodiversity
CONSE QU E NC E S declines and decreased productivity.
More prominent wildfires also have
Warming will directly influence biodi- the potential to diminish biodiversity,
versity in temperate and boreal regions but fire may alternately increase bio-
through altering energetics, rates of growth diversity by reverting forests to earlier
and development (including primary pro- successional stages.
ductivity), seasonal timing, and thermal
stress. Effects will likely be complex, but we r Biodiversity changes in temperate and
outline several generalizations here that we boreal systems will be particularly
believe are likely to occur: challenging to predict because they
TEMPER ATE A ND BOR E A L R E SPONSE S TO CLIM ATE CH A NGE 219
will depend on the incidence of ex- in global leaf phenology.” Nature Climate Change 5:
treme and rare events. Seasonality and 364–368.
Crimmins, S. M., S. Z. Dobrowski, J. A. Greenberg, J.
thermal variability have led temper- T. Abatzoglou, and A. R. Mynsberge. 2011. “Changes
ate and boreal species to evolve broad in climatic water balance drive downhill shifts in
environmental tolerances, which plant species’ optimum elevations.” Science 331 (6015):
may lessen the impacts of warming. 324–327.
However, climate change may push Dale, Virginia H., Linda A. Joyce, Steve McNulty, Ronald
P. Neilson, Matthew P. Ayres, Michael D. Flannigan,
rare events beyond the environmen- Paul J. Hanson, et al. 2001. “Climate change and for-
tal tolerance of species and result in est disturbances: Climate change can affect forests by
thermal stress. altering the frequency, intensity, duration, and timing
of fire, drought, introduced species, insect and patho-
gen outbreaks, hurricanes, windstorms, ice storms, or
landslides.” BioScience 51 (9): 723–734.
R E FE R E NCE S Deutsch, C. A., J. J. Tewksbury, R. B. Huey, K. S. Sheldon,
C. K. Ghalambor, D. C. Haak, and P. R. Martin. 2008.
Aitken, S. N., S. Yeaman, J. A. Holliday, T. Wang, and S. “Impacts of climate warming on terrestrial ecto-
Curtis-McLane. 2008. “Adaptation, migration or ex- therms across latitude.” Proceedings of the National Academy
tirpation: Climate change outcomes for tree popula- of Sciences 105 (18): 6668–6672.
tions.” Evolutionary Applications 1 (1): 95–111. Dobrowski, S. Z., S. M. Crimmins, J. A Greenberg, J. T
Allen, Craig D., Alison K. Macalady, Haroun Chen- Abatzoglou, and A. R. Mynsberge. 2011. “Response
chouni, Dominique Bachelet, Nate McDowell, Michel to comments on ‘Changes in climatic water balance
Vennetier, Thomas Kitzberger, et al. 2010. “A global drive downhill shifts in plant species’ optimum eleva-
overview of drought and heat-induced tree mortality tions.’” Science 334 (6053): 177.
reveals emerging climate change risks for forests.” For- Ettinger, Ailene K., Kevin R. Ford, and Janneke Hille-
est Ecology and Management 259 (4): 660–684. RisLambers. 2011. “Climate determines upper, but not
Altermatt, F. 2009. “Climatic warming increases volt- lower, altitudinal range limits of Pacific Northwest
inism in European butterflies and moths.” Proceedings of conifers.” Ecology 92 (6): 1323–1331.
the Royal Society B 277: 1281–1287. Fisichelli, Nicholas A., Lee E. Frelich, and Peter B. Reich.
Altizer, Sonia, Richard S. Ostfeld, Pieter T. J. Johnson, 2013. “Climate and interrelated tree regeneration
Susan Kutz, and C. Drew Harvell. 2013. “Climate drivers in mixed temperate-boreal forests.” Landscape
change and infectious diseases: From evidence to a Ecology 28 (1): 149–159.
predictive framework.” Science 341 (6145): 514–519. Frelich, Lee E., Rolf O. Peterson, Martin Dovčiak, Pe-
Beckage, Brian, Ben Osborne, Daniel G. Gavin, Caro- ter B. Reich, John A. Vucetich, and Nico Eisenhauer.
lyn Pucko, Thomas Siccama, and Timothy Perkins. 2012. “Trophic cascades, invasive species and body-
2008. “A rapid upward shift of a forest ecotone dur- size hierarchies interactively modulate climate change
ing 40 years of warming in the Green Mountains of responses of ecotonal temperate-boreal forest.” Philo-
Vermont.” Proceedings of the National Academy of Sciences 105 sophical Transactions of the Royal Society B: Biological Sciences 367
(11): 4197–4202. (1605): 2955–2961.
Boisvenue, Céline, and Steven W. Running. 2006. “Im- Grimm, Nancy B., F. Stuart Chapin III, Britta Bierwa-
pacts of climate change on natural forest productivity: gen, Patrick Gonzalez, Peter M. Groffman, Yiqi Luo,
Evidence since the middle of the 20th century.” Global Forrest Melton, et al. 2013. “The impacts of climate
Change Biology 12 (5): 862–882. change on ecosystem structure and function.” Frontiers
Bonan, Gordon B., and David Pollard. 1992. “Effect of in Ecology and the Environment 11 (9): 474–482.
boreal forest vegetation on global climate.” Nature 359: Harsch, Melanie A., Philip E. Hulme, Matt S. McGlone,
716–718. and Richard P. Duncan. 2009. “Are treelines advanc-
Breshears, David D., Neil S. Cobb, Paul M. Rich, Kevin ing? A global meta-analysis of treeline response to
P. Price, Craig D. Allen, Randy G. Balice, William H. climate warming.” Ecology Letters 12 (10): 1040–1049.
Romme, et al. 2005. “Regional vegetation die-off in Khaliq, Imran, Christian Hof, Roland Prinzinger, Ka-
response to global-change-type drought.” Proceedings trin Böhning-Gaese, and Markus Pfenninger. 2014.
of the National Academy of Sciences 102 (42): 15144–15148. “Global variation in thermal tolerances and vulner-
Buckley, Lauren B., Joshua J. Tewksbury, and Curtis A. ability of endotherms to climate change.” Proceedings of
Deutsch. 2013. “Can terrestrial ectotherms escape the the Royal Society B: Biological Sciences 281 (1789): 20141097.
heat of climate change by moving?” Proceedings of the Kingsolver, Joel G., Sarah E. Diamond, and Lauren B.
Royal Society B: Biological Sciences 280 (1765): 20131149. Buckley. 2013. “Heat stress and the fitness conse-
Buitenwerf, Robert, Laura Rose, and Steven I. Higgins. quences of climate change for terrestrial ectotherms.”
2015. “Three decades of multi-dimensional change Functional Ecology 27 (6): 1415–1423.
220 W H AT DOE S THE FUTUR E HOL D?
Kurz, Werner A., C. C. Dymond, G. Stinson, G. J. Ram- Reich, Peter B., Kerrie M. Sendall, Karen Rice, Roy L.
pley, E. T. Neilson, A. L. Carroll, T. Ebata, and L. Saf- Rich, Artur Stefanski, Sarah E. Hobbie, and Rebecca
ranyik. 2008. “Mountain pine beetle and forest car- A. Montgomery. 2015. “Geographic range predicts
bon feedback to climate change.” Nature 452 (7190): photosynthetic and growth response to warming in
987–990. co-occurring tree species.” Nature Climate Change. http://
Littell, Jeremy S., David L. Peterson, and Michael Tjo- www.nature.com/nclimate/journal/vaop/ncurrent/
elker. 2008. “Douglas-fir growth in mountain ecosys- full/nclimate2497.html.
tems: Water limits tree growth from stand to region.” Running, Steven W. 2008. “Ecosystem disturbance, car-
Ecological Monographs 78 (3): 349–368. bon, and climate.” Science 321 (5889): 652–653.
Loarie, S. R., P. B. Duffy, H. Hamilton, G. P. Asner, C. B. Schloss, C. A., T. A. Nuñez, and J. J. Lawler. 2012. “Dis-
Field, and D. D. Ackerly. 2009. “The velocity of cli- persal will limit ability of mammals to track climate
mate change.” Nature 462 (7276): 1052–1055. change in the Western Hemisphere.” Proceedings of the
Logan, Jesse A., Jacques Regniere, and James A. Powell. National Academy of Sciences 109 (22): 8606–8611.
2003. “Assessing the impacts of global warming on Sunday, Jennifer M., Amanda E. Bates, Michael R. Ke-
forest pest dynamics.” Frontiers in Ecology and the Environ- arney, Robert K. Colwell, Nicholas K. Dulvy, John T.
ment 1 (3): 130–137. Longino, and Raymond B. Huey. 2014. “Thermal-
Melillo, Jerry M., T. C. Richmond, and G. W. Yohe. safety margins and the necessity of thermoregulatory
2014. Climate Change Impacts in the United States: The Third behavior across latitude and elevation.” Proceedings of the
National Climate Assessment. U.S. Global Change Research National Academy of Sciences 111: 5610–5615.
Program. Terrier, Aurélie, Martin P. Girardin, Catherine Périé,
Millar, Constance I., Robert D. Westfall, Diane L. Delany, Pierre Legendre, and Yves Bergeron. 2013. “Potential
John C. King, and Lisa J. Graumlich. 2004. “Response changes in forest composition could reduce impacts of
of subalpine conifers in the Sierra Nevada, California, climate change on boreal wildfires.” Ecological Applica-
USA, to 20th-century warming and decadal climate tions 23 (1): 21–35.
variability.” Arctic, Antarctic, and Alpine Research 36 (2): Vasseur, David A., John P. DeLong, Benjamin Gilbert,
181–200. Hamish S. Greig, Christopher D. G. Harley, Kevin
Moritz, C., J. L. Patton, C. J. Conroy, J. L. Parra, G. C. S. McCann, Van Savage, Tyler D. Tunney, and Mary
White, and S. R. Beissinger. 2008. “Impact of a cen- I. O’Connor. 2014. “Increased temperature variation
tury of climate change on small-mammal commu- poses a greater risk to species than climate warming.”
nities in Yosemite National Park, USA.” Science 322 Proceedings of the Royal Society B: Biological Sciences 281 (1779):
(5899): 261–264. 20132612.
Ostfeld, Richard. 2010. Lyme Disease: The Ecology of a Complex Visser, M. E., and C. Both. 2005. “Shifts in phenology due
System. Oxford University Press. to global climate change: The need for a yardstick.”
Parmesan, C. 2006. “Ecological and evolutionary re- Proceedings of the Royal Society B 272 (1581): 2561–2569.
sponses to recent climate change.” Annual Review of Ecol- Way, Danielle A., and Ram Oren. 2010. “Differential re-
ogy, Evolution and Systematics 37: 637–669. sponses to changes in growth temperature between
Parmesan, C. 2007. “Influences of species, latitudes trees from different functional groups and biomes:
and methodologies on estimates of phenological re- A review and synthesis of data.” Tree Physiology 30 (6):
sponse to global warming.” Global Change Biology 13 (9): 669–688.
1860–1872. Westerling, Anthony L., Hugo G. Hidalgo, Daniel R.
Raffa, Kenneth F., Brian H. Aukema, Barbara J. Bentz, Cayan, and Thomas W. Swetnam. 2006. “Warming
Allan L. Carroll, Jeffrey A. Hicke, Monica G. Turner, and earlier spring increase western US forest wildfire
and William H. Romme. 2008. “Cross-scale drivers activity.” Science 313 (5789): 940–943.
of natural disturbances prone to anthropogenic am- Williams, Caroline M., Hugh A. L. Henry, and Brent J.
plification: The dynamics of bark beetle eruptions.” Sinclair. 2014. “Cold truths: How winter drives re-
Bioscience 58 (6): 501–517. sponses of terrestrial organisms to climate change.”
Biological Reviews 90: 214–235.
CHAPTER SEVENTEEN SE T TI NG T H E SC E N E : C L IM AT E
C H A NGE I N MOU N TA I N A R E AS
Climate Change Mountains cover 12.3 percent of the ter-
Impacts on restrial area outside Antarctica (Table 17.1;
Körner et al. 2011) and harbor a propor-
Mountain tionally higher amount of biodiversity
than lowlands, including several biodi-
Biodiversity versity hotspots, such as tropical cloud
forests (Dimitrov et al. 2012) or endemic-
ANTOINE A. GUISAN, OLIVIER rich alpine grasslands (Engler et al. 2011).
BROENNIMANN, ALINE BURI, High-elevation habitats present particular
CARMEN CIANFR ANI, MANUELA D’AMEN, ecological challenges for life (e.g., low tem-
VALERIA DI COLA, RUI FERNANDE S, peratures, short windows for reproduction,
SAR AH M. GR AY, RUBÉN G. MATEO, fluctuation in food availability, high solar
ERIC PINTO, JEAN-NICOL AS PR ADERVAND, radiation, hypoxia), restricting colonization
DANIEL SCHER RER, PASCAL VITTOZ, to adapted life forms (Scridel 2014). Moun-
ISALINE VON DÄNIK EN, AND tain regions seem also to be warming at a
ERIK A YASHIRO higher rate than other regions (Rangwala
et al. 2013), which makes these ecosystems
quite sensitive to natural and anthropo-
genic impacts (e.g., climate, fires, land-use
changes; Beniston 1994; Beniston et al.
1997).
Mountains were identified early as sensi-
tive to climate change (e.g., Beniston 1994;
Guisan et al. 1995) for being (1) climatically
sensitive areas (Nogués-Bravo et al. 2007);
(2) “islands” separated by lowland areas
(Pauchard et al. 2009); (3) sensitive high-
elevation ecosystems, with no escape for
species toward higher elevations (Theuril-
lat and Guisan 2001); and (4) prone to con-
flict between nature conservation and use
of mountains to provide services to humans
(Gret-Regamey et al. 2012).
According to the IPCC (Beniston et al.
1996 and following reports; see http://
www.ipcc.ch), mountains are already af-
fected by climate change, as shown in the
past decades by the following:
221
222 W H AT DOE S THE FUTUR E HOL D?
Area
Thermal belts (Mio km2) M (%) T (%)
Nival (<3.5°C, season <10d) 0.53 3.24 0.40
Upper alpine (<3.5°C, 10d< season <54d) 0.75 4.53 0.56
Lower alpine (<6.4°C, 54d< season <94d) 2.27 13.74 1.68
Tree line
Upper montane (>6.4°C ≤10°C) 3.29 20.53 2.51
Lower montane (>10°C ≤15°C) 3.74 22.64 2.78
Remaining mountain area with frost (<15°C) 1.34 8.11 0.99
Remaining mountain area without frost (>15°C) 4.49 27.22 3.34
Total 16.51 100.00 12.26
Note: See http://www.mountainbiodiversity.org. Temperatures refer to growth season mean air tempera-
tures: M (%) = percentage of total mountain area (100% = 16.5 Mio km2); T (%) = percentage of total
terrestrial area outside Antarctica (100% = 134.6 Mio km2). From Körner, Paulsen, and Spehn (2011).
r Shrinkage of most mountain glaciers, 2008), and these trends are expected to
with greatest ice losses in Patagonia, intensify in the future. Here, we review
Alaska, northwestern United States, the changes already observed and future
southwestern Canada, and the Euro- model-based projections in mountain re-
pean Alps (Arendt et al. 2012), and gions worldwide for many different groups
glaciers in mountains of tropical areas and ecosystems (Figure 17.1).
also particularly affected (Thompson
et al. 2006).
MOU N TA I N F OR E STS A N D T R E E L I N E S
r Significant change in snow cover,
mainly toward overall reduction and About 78 percent of the world’s mountain
greater seasonal variations (Brown areas are below the natural tree line and
and Robinson 2011), resulting in therefore potentially forested (Paulsen and
greater snow-free growing seasons Körner 2014). In the past 100 years the vast
(~5 days per decade; Choi et al. 2010). majority of tree lines around the globe ad-
vanced upward (Harsch et al. 2009), corre-
Predictions of future climate change for lated with an increasing growth rate of es-
mountain ranges worldwide involve an av- tablished trees (Salzer et al. 2009). However,
erage temperature change of 2°C–3°C by the advance of the tree line is not necessarily
2070 and 3°C–5°C by the end of the century tracking the current warming rate, likely due
(Nogués-Bravo et al. 2007), with greater in- to the slow tree growth and very sensitive
creases for mountains in northern latitudes recruitment to climatic conditions. Hence,
than in temperate and tropical climates. An it may take a long time before a new equi-
additional threat is represented by invasive librium is reached between climate and tree
species (e.g., plants), which are predicted to line (Körner 2012). Despite such lag in tree-
increasingly invade mountains under cli- line advance, both mechanistic and correla-
mate change (Pauchard et al. 2009; Petitpi- tive modeling approaches predict an advance
erre et al. 2016). of the tree line and a consequent reduction of
Many mountain species are already re- the alpine and nival (i.e., higher than alpine,
sponding to the effects of ongoing climate where vegetation becomes patchy and scarce)
warming (Pounds et al. 1999; Lenoir et al. areas (Gehrig-Fasel et al. 2007; Körner 2012).
Figure 17.1. Summaries of the main observed (O) and
predicted (P) trends for the responses of the different
biodiversity groups to climate change in mountain areas.
224 W H AT DOE S THE FUTUR E HOL D?
environments are still scarce. Future projec- chian Mountains, with significant declines
tions of butterfly species suggest a potential in suitable habitat, especially for those spe-
altitudinal shift of 650 m by the year 2100 cies with southern and/or smaller ranges.
(Merrill et al. 2008). Pradervand et al. (2014) Taking a broader perspective, Lawler et al.
predicted community homogenization and (2010) projected a very high species turn-
species-specific responses of bumblebees over (exceeding 60 percent) in the Andes
to climate change along an elevation gra- under the 2071–2100 future scenarios.
dient, with the most affected species being Other studies predicting climate change risk
those currently restricted to high elevation. in European countries projected a potential
For the same mountain area, Descombes et high range contraction for many moun-
al. (2015) projected a potential reduction tain amphibians and mountain areas (e.g.,
between 37 percent and 50 percent of the D’Amen et al. 2011; Popescu et al. 2013).
suitable areas for butterflies, depending on
their faculty to broaden or not their diet to
new host plants. MOU N TA IN R E PTIL E S
Descombes, P., J.-N. Pradervand, J. Golay, A. Guisan, and Gehrig-Fasel, J., A. Guisan, and N. E. Zimmermann.
L. Pellissier. 2015. “Simulated shifts in trophic niche 2007. “Tree line shifts in the Swiss Alps: Climate
breadth modulate range loss of alpine butterflies un- change or land abandonment?” Journal of Vegetation Sci-
der climate change.” Ecography 39: 796–804. ence 18: 571–582.
Di Rosa, I., F. Simoncelli, A. Fagotti, and R. Pascolini. Gottfried, M., H. Pauli, A. Futschik, M. Akhalkatsi, P.
2007. “The proximate cause of frog decline?” Nature Barančok, J. L. B. Alonso, G. Coldea, J. Dick, B. Er-
447 (E4–E5): 44. schbamer, and G. Kazakis. 2012. “Continent-wide
Dimitrov, D., D. Nogues-Bravo, and N. Scharff. 2012. response of mountain vegetation to climate change.”
“Why do tropical mountains support exceptionally Nature Climate Change 2: 111–115.
high biodiversity? The eastern arc mountains and the Gret-Regamey, A., S. H. Brunner, and F. Kienast. 2012.
drivers of Saintpaulia diversity.” PLOS One 7 (11): e48908. “Mountain ecosystem services: Who cares?” Mountain
http://doi.org/10.1371/journal.pone.0048908. Research and Development 32: S23–S34.
Dullinger, S., A. Gattringer, W. Thuiller, D. Moser, N. E. Guisan, A., J. I. Holten, R. Spichiger, and L. Tessier, eds.
Zimmermann, A. Guisan, W. Willner, C. Plutzar, M. 1995. Potential Ecological Impacts of Climate Change in the Alps
Leitner, and T. Mang. 2012. “Extinction debt of high- and Fennoscandian Mountains. Conservatoire et Jardin Bota-
mountain plants under twenty-first-century climate nique de Genève.
change.” Nature Climate Change 2: 619–622. Guizado-Rodriguez, M. A., C. Ballesteros-Barrera, G. Ca-
Eby, L. A., O. Helmy, L. M. Holsinger, and M. K. Young. sas-Andreu, V. L. Barradas-Miranda, O. Tellez-Valdes,
2014. “Evidence of climate-induced range contrac- and I. H. Salgado-Ugarte. 2012. “The impact of global
tions in bull trout Salvelinus confluentus in a Rocky Moun- warming on the range distribution of different cli-
tain watershed, USA.” PLOS One 9: e98812. matic groups of Aspidoscelis costata costata.” Zoological Science
Elkin, C., A. G. Gutierrez, S. Leuzinger, C. Manusch, 29: 834–843.
C. Temperli, L. Rasche, and H. Bugmann. 2013. “A Hagvar, S., and K. Klanderud. 2009. “Effect of simulated
2 degrees C warmer world is not safe for ecosystem environmental change on alpine soil arthropods.”
services in the European Alps.” Global Change Biology 19: Global Change Biology 15: 2972–2980.
1827–1840. Harsch, M. A., P. E. Hulme, M. S. McGlone, and R. P.
Elumeeva, T. G., V. G. Onipchenko, A. V. Egorov, A. B. Duncan. 2009. “Are treelines advancing? A global
Khubiev, D. K. Tekeev, N. A. Soudzilovskaia, and J. H. meta analysis of treeline response to climate warm-
C. Cornelissen. 2013. “Long-term vegetation dynamic ing.” Ecology letters 12: 1040–1049.
in the Northwestern Caucasus: Which communities Hegland, S. J., A. Nielsen, A. Lazaro, A. L. Bjerknes, and
are more affected by upward shifts of plant species?” O. Totland. 2009. “How does climate warming af-
Alpine Botany 123: 77–85. fect plant-pollinator interactions?” Ecology Letters 12:
Engler, R., C. F. Randin, W. Thuiller, S. Dullinger, N. 184–195.
E. Zimmermann, M. B. Araújo, P. B. Pearman, et al. Hobbs, W. O., R. J. Telford, H. J. B. Birks, J. E. Saros,
2011. “21st century climate change threatens moun- R. R. Hazewinkel, B. B. Perren, É. Saulnier-Talbot,
tain flora unequally across Europe.” Global Change Biology and A. P. Wolfe. 2010. “Quantifying recent ecological
17: 2330–2341. changes in remote lakes of North America and Green-
Fan, J., J. Li, R. Xia, L. Hu, X. Wu, and G. Li. 2014. “As- land using sediment diatom assemblages.” PLOS One 5:
sessing the impact of climate change on the habitat e10026.
distribution of the giant panda in the Qinling Moun- Inouye, D. W., B. Barr, K. B. Armitage, and B. D. Inouye.
tains of China.” Ecological Modelling 274: 12–20. 2000. “Climate change is affecting altitudinal mi-
Fisher, M. C., D. A. Henk, C. J. Briggs, J. S. Brownstein, L. grants and hibernating species.” Proceedings of the National
C. Madoff, S. L. McCraw, and S. J. Gurr. 2012. “Emerg- Academy of Sciences 97: 1630–1633.
ing fungal threats to animal, plant and ecosystem Isaak, D. J., C. C. Muhlfeld, A. S. Todd, R. Al-Chokhachy,
health.” Nature 484: 186–194. J. Roberts, J. L. Kershner, and S. W. Hostetler. 2012.
Freeman, B. G., and A. M. C. Freeman. 2014. “Rapid “The past as prelude to the future for understand-
upslope shifts in New Guinean birds illustrate strong ing 21st-century climate effects on Rocky Mountain
distributional responses of tropical montane species trout.” Fisheries 37: 542–556.
to global warming.” Proceedings of the National Academy of Jacobsen, D., A. M. Milner, L. E. Brown, and O. Dangles.
Sciences 111: 4490–4494. 2012. “Biodiversity under threat in glacier-fed river
Fujimura, K. E., K. N. Egger, and G. H. Henry. 2008. systems.” Nature Climate Change 2: 361–364.
“The effect of experimental warming on the root-as- Körner, C. 2012. Alpine Treelines: Functional Ecology of the Global
sociated fungal community of Salix arctica.” ISME Journal High Elevation Tree Limits. Springer Science and Business
2: 105–114. Media.
Füreder, L., R. Ettinger, A. Boggero, B. Thaler, and H. Körner, C., J. Paulsen, and E. M. Spehn. 2011. “A defi-
Thies. 2006. “Macroinvertebrate diversity in Alpine nition of mountains and their bioclimatic belts for
lakes: Effects of altitude and catchment properties.” global comparisons of biodiversity data.” Alpine Botany
Hydrobiologia 562: 123–144. 121: 73–78.
C L I M AT E C H A NGE I M PAC T S ON MOU N TA I N BIODI V E R SI T Y 231
Kuffner, M., B. Hai, T. Rattei, C. Melodelima, M. versity hotspot as a consequence of projected global
Schloter, S. Zechmeister-Boltenstern, R. Jandl, A. climate change.” PLOS One 5: e12189.
Schindlbacher, and A. Sessitsch. 2012. “Effects of sea- Nogués-Bravo, D., M. B. Araújo, M. P. Errea, and J. P.
son and experimental warming on the bacterial com- Martínez-Rica. 2007. “Exposure of global mountain
munity in a temperate mountain forest soil assessed systems to climate warming during the 21st Century.”
by 16S rRNA gene pyrosequencing.” FEMS Microbiology Global Environmental Change 17: 420–428.
Ecology 82: 551–562. Parkinson, D., M. McLean, and S. Scheu. 2004. “Impact
Lawler, J. J., S. L. Shafer, B. A. Bancroft, and A. R. Blaus- of earthworms on other biota in forests soils, with
tein. 2010. “Projected climate impacts for the amphib- some emphasis on cool temperate montane forests.”
ians of the Western Hemisphere.” Conservation Biology In Earthworm Ecology, 2nd ed., ed. C. A. Edwards, 241–
24: 38–50. 259. CRC Press.
Lenoir, J., J. C. Gégout, P. A. Marquet, P. De Ruffray, and Pauchard, A., C. Kueffer, H. Dietz, C. C. Daehler, J. Alex-
H. Brisse. 2008. “A significant upward shift in plant ander, P. J. Edwards, J. R. Arevalo, et al. 2009. “Ain’t no
species optimum elevation during the 20th century.” mountain high enough: Plant invasions reaching new el-
Science 320: 1768–1771. evations.” Frontiers in Ecology and the Environment 7: 479–486.
Lenoir, J., J. C. Gegout, J. C. Pierrat, J. D. Bontemps, and Pauli, H., M. Gottfried, S. Dullinger, O. Abdaladze, M.
J. F. Dhote. 2009. “Differences between tree species Akhalkatsi, J. L. B. Alonso, G. Coldea, J. Dick, B. Ersch-
seedling and adult altitudinal distribution in moun- bamer, and R. F. Calzado. 2012. “Recent plant diver-
tain forests during the recent warm period (1986– sity changes on Europe’s mountain summits.” Science
2006).” Ecography 32: 765–777. 336: 353–355.
Lurgi, M., B. C. Lopez, and J. M. Montoya. 2012. “Climate Paulsen, J., and C. Körner. 2014. “A climate-based model
change impacts on body size and food web structure to predict potential treeline position around the
on mountain ecosystems.” Philosophical Transactions of the globe.” Alpine Botany 124: 1–12.
Royal Society B: Biological Sciences 367: 3050–3057. Pellissier, L., E. Pinto-Figueroa, H. Niculita-Hirzel, M.
MacDonald, R. J., S. Boon, J. M. Byrne, M. D. Robinson, Moora, L. Villard, J. Goudet, N. Guex, M. Pagni, I.
and J. B. Rasmussen. 2014. “Potential future climate Xenarios, I. Sanders, and A. Guisan. 2013. “Plant spe-
effects on mountain hydrology, stream temperature, cies distributions along environmental gradients: Do
and native salmonid life history.” Canadian Journal of Fish- belowground interactions with fungi matter?” Fron-
eries and Aquatic Sciences 71: 189–202. tiers in Plant Science 4. https://doi.org/10.3389/fpls.2013
Maggini, R., A. Lehmann, M. Kery, H. Schmid, M. Benis- .00500.
ton, L. Jenni, and N. Zbinden. 2011. “Are Swiss birds Pellissier, L., H. Niculita-Hirzel, A. Dubuis, M. Pagni, N.
tracking climate change? Detecting elevational shifts us- Guex, C. Ndiribe, N. Salamin, et al. 2014. “Soil fungal
ing response curve shapes.” Ecological Modelling 222: 21–32. communities of grasslands are environmentally struc-
Maggini, R., A. Lehmann, N. Zbinden, N. E. Zimmer- tured at a regional scale in the Alps.” Molecular Ecology
mann, J. Bolliger, B. Schroder, R. Foppen, H. Schmid, 23: 4274–4290.
M. Beniston, and L. Jenni. 2014. “Assessing species Peñuelas, J., and M. Boada. 2003. “A global change in-
vulnerability to climate and land use change: The case duced biome shift in the Montseny Mountains (NE
of the Swiss breeding birds.” Diversity and Distributions 20: Spain).” Global Change Biology 9: 131–140.
708–719. Petitpierre, B., K. McDougall, T. Seipel, O. Broennimann,
Margesin, R., S. Minerbi, and F. Schinner. 2014. “Long- A. Guisan, and C. Kueffer. 2016. “Will climate change
term monitoring of soil microbiological activities in increase the risk of plant invasions into mountains?”
two forest sites in South Tyrol in the Italian alps.” Mi- Ecological Applications 26: 530–544.
crobes and Environments 29: 277–285. Pettorelli, N., F. Pelletier, A. von Hardenberg, M. Festa-
Martin, K. 2013. “The ecological values of mountain en- Bianchet, and S. D. Cote. 2007. “Early onset of vegeta-
vironments and wildlife: The impacts of skiing and tion growth vs. rapid green-up: Impacts on juvenile
related winter recreational activities on mountain mountain ungulates.” Ecology 88: 381–390.
environments.” In The Impacts of Skiing and Related Winter Ponce-Reyes, R., E. Nicholson, P. W. J. Baxter, R. A.
Recreational Activities on Mountain Environments, ed. C. Rixen Fuller, and H. Possingham. 2013. “Extinction risk in
and A. Rolando, 3–29. Bentham E-Books. cloud forest fragments under climate change and hab-
Menéndez, R. 2007. “How are insects responding to itat loss.” Diversity and Distributions 19: 518–529.
global warming?” Tijdschrift voor Entomologie 150: 355. Popescu, V. D., L. Rozylowicz, D. Cogǎlniceanu, I. M.
Merrill, R. M., D. Gutierrez, O. T. Lewis, J. Gutierrez, S. Niculae, and A. L. Cucu. 2013. “Moving into protected
B. Diez, and R. J. Wilson. 2008. “Combined effects areas? Setting conservation priorities for Romanian
of climate and biotic interactions on the elevational reptiles and amphibians at risk from climate change.”
range of a phytophagous insect.” Journal of Animal Ecology PLOS One 8: e79330.
77: 145–155. Pounds, J. A., M. P. L. Fogden, and J. H. Campbell. 1999.
Milanovich, J. R., W. E. Peterman, N. P. Nibbelink, and “Biological response to climate change on a tropical
J. C. Maerz. 2010. “Projected loss of a salamander di- mountain.” Nature 398: 611–615.
232 W H AT DOE S THE FUTUR E HOL D?
Pounds, J. A., M. R. Bustamante, L. A. Coloma, J. A. Con- Singh, B. K., R. D. Bardgett, P. Smith, and D. S. Reay.
suegra, M. P. L. Fogden, P. N. Foster, E. La Marca, et 2010. “Microorganisms and climate change: Terres-
al. 2006. “Widespread amphibian extinctions from trial feedbacks and mitigation options.” Nature Reviews
epidemic disease driven by global warming.” Nature Microbiology 8: 779–790.
439: 161–167. Slemmons, K. E. H., J. E. Saros, and K. Simon. 2013. “The
Pradervand, J.-N., L. Pellissier, C. F. Randin, and A. influence of glacial meltwater on alpine aquatic eco-
Guisan. 2014. “Functional homogenization of bum- systems: A review.” Environmental Science: Processes and Im-
blebee communities in alpine landscapes under pro- pacts 15: 1794–1806.
jected climate change.” Climate Change Responses 1 (1). Smol, J. P., A. P. Wolfe, H. J. B. Birks, M. S. V. Doug-
https://doi.org/10.1186/s40665-014-0001-5. las, V. J. Jones, A. Korhola, R. Pienitz, K. Rühland, S.
Rangwala, I., E. Sinsky, and J. R. Miller. 2013. “Amplified Sorvari, and D. Antoniades. 2005. “Climate-driven
warming projections for high altitude regions of the regime shifts in the biological communities of arc-
northern hemisphere mid-latitudes from CMIP5 mod- tic lakes.” Proceedings of the National Academy of Sciences 102:
els.” Environmental Research Letters 8: 024040. 4397–4402.
Roberts, J. J., K. D. Faush, D. P. Peterson, and M. B. Hoo- Swift, M. J., O. Andrén, L. Brusaard, M. Briones, M.-M.
ten. 2013. “Fragmentation and thermal risks from Couteaux, K. Ekschmitt, A. Kjoller, P. Loiseau, and P.
climate change interact to affect persistence of native Smith. 1998. “Global change, soil biodiversity, and
trout in the Colorado River basin.” Global Change Biology nitrogen cycling in terrestrial ecosystems: three case
19: 1383–1398. studies.” Global Change Biology 4: 729–743.
Rosset, V., and B. Oertli. 2011. “Freshwater biodiver- Sylvain, Z. A., and D. H. Wall. 2011. “Linking soil bio-
sity under climate warming pressure: Identifying the diversity and vegetation: Implications for a changing
winners and losers in temperate standing waterbod- planet.” American Journal of Botany 98: 517–527.
ies.” Biological Conservation 144: 2311–2319. Theurillat, J. P., and A. Guisan. 2001. “Potential impact
Rubidge, E. M., J. L. Patton, M. Lim, A. C. Burton, J. S. of climate change on vegetation in the European Alps:
Brashares, and C. Moritz. 2012. “Climate-induced A review.” Climatic Change 50: 77–109.
range contraction drives genetic erosion in an alpine Thompson, L. G., E. Mosley-Thompson, H. Brecher, M.
mammal.” Nature Climate Change 2: 285–288. Davis, B. Leon, D. Les, P.-N. Lin, T. Mashiotta, and
Salerno, F., S. Gambelli, G. Viviano, S. Thakuri, N. K. Mountain. 2006. “Abrupt tropical climate change:
Guyennon, C. D’Agata, G. Diolaiuti, C. Smiraglia, F. Past and present.” Proceedings of the National Academy of Sci-
Stefani, and D. Bocchiola. 2014. “High alpine ponds ences 103: 10536–10543.
shift upwards as average temperatures increase: A case Thompson, P. L., M.-C. St.-Jacques, and R. D. Vine-
study of the Ortles-Cevedale mountain group (South- brooke. 2008. “Impacts of climate warming and ni-
ern Alps, Italy) over the last 50 years.” Global and Plan- trogen deposition on alpine plankton in lake and pond
etary Change 120: 81–91. habitats: An in vitro experiment.” Arctic, Antarctic, and
Scherrer, D., and C. Korner. 2011. “Topographically con- Alpine Research 40: 192–198.
trolled thermal-habitat differentiation buffers alpine Vittoz, P., D. Cherix, Y. Gonseth, V. Lubini, R. Maggini,
plant diversity against climate warming.” Journal of Bio- N. Zbinden, and S. Zumbach. 2013. “Climate change
geography 38: 406–416. impacts on biodiversity in Switzerland: A review.”
Scheurer, K., C. Alewell, D. Baenninger, and P. Bur- Journal for Nature Conservation 21: 154–162.
khardt-Holm. 2009. “Climate and land-use changes Wake, D. B., and V. T. Vredenburg. 2008. “Are we in the
affecting river sediment and brown trout in alpine midst of the sixth mass extinction? A view from the
countries: A review.” Environmental Science and Pollution Re- world of amphibians.” Proceedings of the National Academy of
search 16: 232–242. Sciences 105: 11466–11473.
Scridel, D. 2014. “Ecology and conservation of birds in Wasserman, T. N., S. A. Cushman, J. S. Littell, A. J. Shirk,
upland and alpine habitats: A report on the BOU’s An- and E. L. Landguth. 2013. “Population connectiv-
nual Conference held at the University of Leicester, ity and genetic diversity of American marten (Mar-
1–3 April 2014.” Ibis 156: 896–900. tes americana) in the United States northern Rocky
Sekercioglu, C. H., S. H. Schneider, J. P. Fay, and S. R. Loa- Mountains in a climate change context.” Conservation
rie. 2008. “Climate change, elevational range shifts, Genetics 14: 529–541.
and bird extinctions.” Conservation Biology 22: 140–150. Weidman, P. R., D. W. Schindler, P. L. Thompson, and
Sheldon, A. L. 2012. “Possible climate-induced shift R. D. Vinebrooke. 2014. “Interactive effects of higher
of stoneflies in a southern Appalachian catchment.” temperature and dissolved organic carbon on plank-
Freshwater Science 31: 765–774. tonic communities in fishless mountain lakes.” Fresh-
Sinervo, B., F. Mendez-de-la-Cruz, D. B. Miles, B. Heu- water Biology 59: 889–904.
lin, E. Bastiaans, M. V. S. Cruz, R. Lara-Resendiz, et al. Wilhelm, L., G. A. Singer, C. Fasching, T. J. Battin, and K.
2010. “Erosion of lizard diversity by climate change Besemer. 2013. “Microbial biodiversity in glacier-fed
and altered thermal niches.” Science 328: 894–899. streams.” ISME Journal 7: 1651–1660.
C L I M AT E C H A NGE I M PAC T S ON MOU N TA I N BIODI V E R SI T Y 233
Williams, S. E., E. E. Bolitho, and S. Fox. 2003. “Climate to warming and increased precipitation in a semiarid
change in Australian tropical rainforests: an impend- temperate steppe.” Applied Soil Ecology 84: 200–207.
ing environmental catastrophe.” Proceedings of the Royal Zimmerli, S., D. Bernet, P. Burkhardt-Holm, H. Schmidt-
Society B: Biological Sciences 270: 1887–1892. Posthaus, P. Vonlanthen, T. Wahli, and H. Segner.
Wilson, R. J., D. Gutierrez, J. Gutierrez, and V. J. Mon- 2007. “Assessment of fish health status in four Swiss
serrat. 2007. “An elevational shift in butterfly species rivers showing a decline of brown trout catches.”
richness and composition accompanying recent cli- Aquatic Sciences 69: 11–25.
mate change.” Global Change Biology 13: 1873–1887. Zinger, L., B. Shahnavaz, F. Baptist, R. A. Geremia, and
Winder, M., J. E. Reuter, and S. G. Schladow. 2009. “Lake P. Choler. 2009. “Microbial diversity in alpine tundra
warming favours small-sized planktonic diatom spe- soils correlates with snow cover dynamics.” ISME Jour-
cies.” Proceedings of the Royal Society B: Biological Sciences 276: nal 3: 850–859.
427–435.
Wu, T. J., F. L. Su, H. Y. Han, Y. Du, C. D. Yu, and S.
Q. Wan. 2014. “Responses of soil microarthropods
CASE STUDY 7
High-altitude ecosystems, including the Rocky Mountains in North America, are experienc-
ing some of the strongest effects of climate change (Imtiaz, Sinsky, and Miller 2013; Ped-
erson, Betancourt, and McCabe 2013; Rangwala and Miller 2012), including both warming
of temperatures and changes in precipitation. At 2,900 m (9,500 ft) in the Colorado Rocky
Mountains, at the Rocky Mountain Biological Laboratory, the strongest trend for warming
has been in April minimum temperatures (3ºC change from 1973), which can influence
spring snowmelt. Winter snowfall (measured since 1975 and reconstructed from stream
runoff in the period 1935–1974) has been highly variable, but there is a declining trend
since 1935. The date of snowmelt is similarly variable, for example differing by 40 days
between 1980 and 1981, but also trending toward earlier dates, and the year-to-year varia-
tion (measured by the moving 3-year range) is increasing. The trends toward both warmer
springs and earlier snowmelt dates can have important consequences for the plant commu-
nity and animal species that interact with plants.
Although snowmelt dates are advanc- depend on, and consumers of seeds such as
ing, the date of the last spring frost has not insects and birds will also lose those sources
been changing, still occurring around June of nutrition. Animals attracted to disperse
10. If snowmelt is late, plants have not had seeds by the presence of the fruits contain-
much time to develop before the frost, so ing them will also lose access to those re-
there is little damage. But if snow melts in sources. Herbivores that have evolved to
April, as now sometimes happens, by June take advantage of early spring leaves, such
many plants have already expanded leaves as many species of caterpillars, may starve
and initiated flower buds, or even begun to to death before the trees releaf. Although
flower. In such years, plants that are sensi- in many cases these consequences are easy
tive to frost can suffer significant damage. to understand, in only a few instances have
In some cases, only new growth, such as they been quantified in an ecological con-
the young leaves on aspen trees or new nee- text. One example is the long-term study
dles on conifers, or reproductive parts, such of frost effects on flowers of Erigeron speciosus
as buds or fruits, are sensitive to frost. Some and on the butterfly species Speyeria mormonia,
species can develop new leaves to replace which is highly dependent on nectar from
the damaged ones, but short-term replace- those flowers in order to reproduce (Boggs
ment of reproductive parts does not seem and Inouye 2012). In this case 84 percent of
to be an option in this environment. So in the variation in population growth rate of
those years that do have early snowmelt and the butterfly was explained by how many
late frosts, there may be no flowers, fruits, flowers avoided frost damage.
or seeds produced by frost-sensitive plants. Flower buds of the aspen sunflower
In addition to the potential effects of loss (Helianthella quinquenervis) are highly sensitive
of flowers, fruits, and seeds for the affected to frost, and in many years close to 100
plants, the consequences of the loss of these percent of the buds die off (Figure CS7.1).
resources for consumers such as pollinators, Flower buds produce extrafloral nectar that
seed predators, and seed dispersers can be
significant (Inouye 2008). Pollinators will Copyright © 2019 by Yale University. All rights reserved.
not have access to the pollen and nectar they ISBN 978-0-300-20611-1.
234
CL IM ATE CH A NGE A ND FROST EFFECTS IN ROCK Y MOUNTA IN PL A NTS 235
attracts ants, which defend them from flies (the coldest in 126 years; “South African
(two species of Tephritidae and one of Ag- frosts” 2013), and in 2014 in Argentina and
romyzidae) that lay eggs on the heads (In- Chile (Frank 2013).
ouye and Taylor 1979). The flowers are vis- These incidents of frost damage to wild-
ited by bumblebees and flies for nectar and flowers and crops may continue to increase
pollen, and the flower heads are popular in frequency unless the low temperatures in
food for large herbivores such as deer and late spring begin to warm significantly. But
sheep. Thus, the loss of flower heads to frost the fact that some models of climate change
affects a variety of consumer species as well predict increased variability suggests that
as their parasites and predators. The Erigeron perhaps frost will continue to influence
flower heads that are so important for but- ecological communities and agriculture for
terflies for nectar are also used by tephritid quite some time.
flies, whose larvae eat developing seeds, and
their populations probably also suffer from
the effects of frost on flower abundance. R E FE R E NCE S
The pattern of frost damage I have ob-
served in wildflowers in Colorado is also Ault, T. R., G. M. Henebry, K. M. de Beurs, M. D. Schwartz,
J. L. Betancourt, and D. Moore. 2013. “The false spring
occurring in agriculture there, where the
of 2012, earliest in North American record.” Eos,
loss of apricots and cherries, which flower
early, seems to be happening more fre-
quently. The phenomenon is global, too. Figure CS7.1. Aspen sunflower (Helianthella quinquenervis) is
For example, a warm spell in early spring a common wildflower near the Rocky Mountain Bio-
logical Laboratory, but flower abundance is extremely
2007 was followed by a hard freeze that variable. The plants themselves are long-lived (decades-
caused more than $2 billion in damage to long) perennials, but the flower buds are sensitive to
agriculture in the eastern United States (Gu frost. As snowmelt gets earlier (with the date of the
et al. 2008; NOAA/USDA 2008). A similar last frost still around June 10), the frequency of frost
2012 event in Michigan (Ault et al. 2013) kills of buds is becoming more common. These data
for counts of unfrosted flowers come from two per-
was responsible for $223 million in losses manent plots at RMBL. (Courtesy of David W. Inouye,
(Parker 2013). Wine grape harvests were data collected with funding from NSF grant DEB-
devastated by frost in South Africa in 2013 1354104.)
236 W H AT DOE S THE FUTUR E HOL D?
Transactions American Geophysical Union 94 (20): 181–182. Parker, R. 2013. “Michigan fruit losses documented
https://doi.org/10.1002/2013eo200001. in 2012 USDA report.” Michigan State University.
Boggs, Carol L., and David W. Inouye. 2012. “A single http://www.mlive.com/news/kalamazoo/index
climate driver has direct and indirect effects on insect .ssf/2013/02/michigan_fruit_losses_document.html.
population dynamics.” Ecology Letters 15 (5): 502–508. Pederson, Gregory T., Julio L. Betancourt, and Gregory
https://doi.org/10.1111/j.1461-0248.2012.01766.x. J. McCabe. 2013. “Regional patterns and proximal
Frank, M. 2013. “Chilean wine producers hit by frost.” causes of the recent snowpack decline in the Rocky
Wine Spectator. http://www.winespectator.com/ Mountains, U.S.” Geophysical Research Letters 40 (9): 1811–
webfeature/show/id/49055. 1816. https://doi.org/10.1002/grl.50424.
Gu, Lianhong, Paul J. Hanson, W. Mac Post, Dale P. Kai- Rangwala, Imtiaz, and James R. Miller. 2012. “Cli-
ser, Bai Yang, Ramakrishna Nemani, Stephen G. Pal- mate change in mountains: A review of elevation-
lardy, and Tilden Meyers. 2008. “The 2007 eastern US dependent warming and its possible causes.” Climatic
spring freeze: Increased cold damage in a warming Change 114 (3–4): 527–547. https://doi.org/10.1007/
world?” BioScience 58: 253–262. s10584-012-0419-3.
Inouye, David W. 2008. “Effects of climate change on Rangwala, Imtiaz, E. Sinsky, and J. R. Miller. 2013.
phenology, frost damage, and floral abundance of “Amplified warming projections for high altitude re-
montane wildflowers.” Ecology 89 (2): 353–362. gions of the northern hemisphere mid-latitudes from
Inouye, David W., and Orley R. Taylor. 1979. “A tem- CMIP5 models.” Environmental Research Letters 8: 024040.
perate region plant-ant-seed predator system: Conse- “South African frosts mean job and volume loss for
quences of extrafloral nectar secretion by Helianthella grape sector.” 2013. FreshFruitPortal.com. https://
quinquenervis.” Ecology 60 (1): 1–7. www.freshfruitportal.com/news/2013/10/16/south
NOAA/USDA. 2008. “The Easter freeze of April 2007: A -african-frosts-mean-job-and-volume-loss-for-grape
climatological perspective and assessment of impacts -sector.
and services.” https://www1.ncdc.noaa.gov/pub/
data/techrpts/tr200801/tech-report-200801.pdf.
CHAPTER EIGHTEEN I N T RODUCTION
237
238 W H AT DOE S THE FUTUR E HOL D?
extinction, with climate change a major out, and triggering floods, droughts, and
contributing factor. The signal is evident even earthquakes on a schedule dictated by
even if we exclude the loss in whole or human, not whole, systems. Hydropower
part of lacustrine fish species flocks such and irrigation projects have been com-
as those in Lakes Victoria (Africa), Lanao pleted for ages as the human population
(Philippines), and Tana (Ethiopia). For ex- and global economy mushroomed. Yet the
ample, using a flow model, Xenopoulos et key to understanding climate change im-
al. (2005) estimated that up to 75 percent pacts on freshwater fishes is to first grasp
of freshwater fish species are likely to be the ricochet dynamic of climate and society.
negatively affected, directly or indirectly, Climate change most obviously threatens
by climate change, with the lion’s share of biodiversity hotspots, a few of which have
species occurring in developing world na- been classified on every continent except
tions. Added to the freshwater extinctions Antarctica. According to a recent analysis
already under way due to dams, pollution, (Newbold et al. 2016), biodiversity hotspots
and other insults, the prospect from a mass harbor 55 percent of the world’s freshwater
extinction of freshwater species is all but fishes, 29 percent as endemics. A different
ensured (e.g., Winemiller et al. 2016). The analysis would be required to assess the
degree to which any particular species will conservation of evolutionary uniqueness,
be affected is a function of its evolutionary as many spots with high freshwater fish
history, the location and topography of the endemism are dominated by recent radia-
landscape in which it lives, its life history, tions (e.g., Lakes Victoria and Malawi, des-
and the degree of cumulative human im- ert springs). Hotspots aside, climate change
pacts within its range. also threatens species-poor but highly val-
There may be no freshwater fish that ued fish assemblages of high-latitude lakes
is threatened only by climate change, but and rivers.
likely no freshwater fish will escape at least Watersheds are broadly similar in func-
some impact from climate change. Climate tion and so are the ways in which they re-
change effects are superimposed on other spond to climate change. Differences arise
anthropogenic stressors, a situation similar from interactions among precipitation,
to that for coral reefs: climate change con- latitude, topography, and soils, interactions
stitutes an existential threat, but it acts by that determine vegetation patterns and so
amplifying the threats of pollution, over- fish habitats as structured by plants grow-
fishing, and habitat destruction. In fresh- ing both above and below the waterline.
water systems an added—and indeed, prin- The impacts of climate change on freshwa-
cipal—human impact is hydroengineering ter fishes in any one watershed cannot be
to stabilize and maximize water availability predicted accurately—it is difficult to down-
for human use and to optimize its distribu- scale climate models, and local processes,
tion in time and space to suit human needs. particularly species interactions, are fraught
We draw enormous quantities of water with stochasticity—but the kinds of climate-
from lakes, rivers, and aquifers for domes- related impact are generalizable (Table 18.1).
tic, agricultural, and industrial applications,
so much that water tables fall, rivers dry
before reaching the sea, and deep ancient T H R E E F ISH Y BR IE F S : A M E R IC A , E AST
stores of fossil water are depleted. As fresh- A F R IC A , A N D E AST ASI A
water becomes scarce, people arrest control
of what remains through technology— Freshwater fish diversity is distributed un-
damming or diverting it, fighting wars over evenly, but each of the world’s great land-
it, storing or releasing it, driving it into the masses hosts at least one area of high species
ground to squeeze the last bits of fossil fuel diversity. Most of these areas are home to
Table 18.1. Examples of climate change effects on freshwater habitats and fishes
1. Loss of alpine snowpack and glaciers. Reduced precipitation at high altitude and increased
melting and sublimation cause net loss of snowpack and glaciers. This reduces the
seasonal pulse flow on which migratory fishes key their upstream spawning runs,
reduces downstream drift important in delivering fry to first-feeding grounds,
and can cause intermittent to zero dry-season flow through all or portions of a
watershed.
2. Changes in tundra ponds. Tundra ponds on permafrost are changing in size and connec-
tivity (mostly shrinking), altering Arctic fish communities.
3. Reduction in base flow. A product of reduced precipitation or fog capture anywhere in a
watershed but particularly in the headwaters. Effects similar to No. 1, especially in
causing ever-flowing streams to become seasonal, intermittent, or completely dry.
4. Reduction in maximum flood levels. All freshwater systems occupy some position along an
axis from zero to high flood-pulse amplitude. In classic flood-pulse systems like
the Mekong and Amazon Rivers, submerged basin at peak flood serves as a cru-
cial young-of-year nursery area, with a food web partially powered by bacterial
activity associated with decomposing terrestrial vegetation. If the seasonal floods
are reduced (as by regional drying and dams) and flow stabilized overall (as by
dams) the contribution of flood-pulse dynamics to fish recruitment and growth are
eliminated.
5. High-altitude habitat “squeeze.” Warming shifts habitat zones to higher elevations, elimi-
nating the highest as they run out of mountainside to be on. This will eliminate
low-latitude mountaintop refugia for high-latitude species and threatens distinctive
formations such as tropical cloud forest.
6. Mixing of historically isolated populations and species.
a. Lowering of mountain passes. These are famously “higher in the tropics” (i.e.,
the range in climate from valley to peak is greatest where valley climate is the
warmest). Climate warming may increase connectivity of adjacent freshwater
systems, reducing isolation.
b. Wets increasing connectivity in swamp systems, dries reconnecting island and
mainland populations. Extreme floods and droughts will drive rapid shifts in
population structure and species composition (e.g., spread of exotic species) in
subtropical and tropical watershed and lake systems.
7. Loss of habitat area.
a. Species compression: species interactions in shallow lakes and wetlands (and
some river systems) will be intensified in drought and relaxed in flood (Itz-
kowitz paper pupfish/gambusia, Lake Chapala), causing species shuffling and
potentiating extinctions.
b. Habitat reduction and loss. Fishes of ephemeral water bodies (e.g., “annual”
cyprinodontiforms) such as the rivuline killifishes will be threatened as the
size, abundance, and duration of seasonal ponds are reduced due to increased
evaporation.
(continued)
240 W H AT DOE S THE FUTUR E HOL D?
numerous threatened fishes (and some spe- side broadly distributed species (largely
cies are extinct already); about one-third of the Mississippi Basin fauna), yielding
of freshwater species are listed by the In- both high alpha (in any one spot) and beta
ternational Union for Conservation of Na- (across all habitats) diversity plus an unpar-
ture (IUCN) as vulnerable to critically en- alleled regional (γ) diversity for a temper-
dangered. Threat status aside, even species ate ecosystem: 493 fish species, or about 62
ranked as “least concern” may be at risk percent of the United States’ total. Bivalve
from climate change. mollusks are also phenomenally diverse in
this region, with about 269 species, consti-
tuting 91 percent of the US tally. Many of
North America
these bivalves are gravely endangered, and
Although it is mostly high-latitude and low- nearly all depend on fishes as hosts and dis-
species density, North America boasts three persal agents for their glochidia larvae (one
concentrations of freshwater fish diversity. species uses a large salamander, the mud-
One is in Mexico, a country that spans puppy Necturus maculosus, as its host). Species
the temperate-tropical divide, whereas the densities are lower on the vast southeastern
other two are a curious study in contrasts: coastal plain, although there is appreciably
the southern Appalachian temperate broad- high endemism (e.g., among pygmy sun-
leaf forests and the arid Southwest. In the fishes of the genus Elassoma and killifishes of
Appalachians, narrow endemics swim be- the Cyprinodontidae).
F I N A L A R B I T E R O F T H E M A S S E X T I N C T I O N O F F R E S H WAT E R F I S H E S 241
Climate change plays a role in what has freshwater fishes have been assessed; the
been called an “extinction vortex” in the IUCN Red List is woefully incomplete for
southeastern United States (Freeman et al. the entire continent.
2012). We know that the initial trigger for Loss of fish species in South America
the observed declines in freshwater species can be attributed to climate change of two
is the often-combined impacts of defores- sorts: regional (caused by mass deforesta-
tation, mining, sedimentation, hydroen- tion) and global. These act together and
gineering (chiefly dams), and pollution. are supercharged by strengthening climate
Climate change overlaid on these other fac- cycles such as that of El Niño–Southern Os-
tors is expected to include a warming of cillation. Global climate change will result
3°C–4°C. There may be an increase in pre- in warming and possibly drying of wa-
cipitation, mostly on the coastal plain, con- tercourses in the region. Regional climate
centrated into more intense and clustered change driven by deforestation and conse-
summer storms. Increased evapotranspira- quent loss of moisture recycling is particu-
tion will decrease runoff and stream flow, larly important in the Amazon.
except in Florida and along the Gulf Coast, The Amazon Basin exhibits an extreme
where precipitation may exceed evapo- annual cycle of hydrology; the entire basin
transpiration and perhaps increase stream is a massive flood-pulse system, and fish life
flow, some of which will present as surges histories are geared to it. Any change in the
and floods. Both spike against a lower base- timing and volume of available water, or in
line flow and extended droughts are likely. the rate of change in its availability, will af-
Warmer waters, reduced water quality, fect a large fraction of the more than 2,500
lower oxygen tensions during the summer, fish species that occur there. Such changes
and intense flushing add impetus to incipi- come naturally with every El Niño cycle,
ent mass extinctions in these areas, which but recent droughts have been of historic
host the most species-rich of all the world’s severity. Droughts were so bad in 2005,
temperate freshwater fish communities. 2010, and 2014 that the upper reaches of
many tributaries dried out completely—ob-
viously a problem for fishes, but also a cause
South America
of immense human suffering as lines of
Freshwater fishes in South America, and transportation and commerce disappeared,
Brazil in particular, are imperiled by dams, literally, into thin air. The entire region en-
pollution, deforestation, exotic intro- compassing the eastern Amazon and Mata
ductions, and rapid industrial and urban Atlântica (Atlantic Forest) is exquisitely
growth. In the west are the vulnerable (or sensitive to climate change. The severity of
extinct) fishes endemic to altiplano lakes recent droughts has been tied to the com-
and streams in Chile, Peru, and Bolivia, the bined effects of deforestation and warming
Galaxiids (southern smelts) of Argentina (Xenopoulos et al. 2005).
and Chile, and the poorly known fishes of Evapotranspiration from the Amazon
the western Andean slopes. To the east is forest is the wellspring of an aerial cur-
the Amazon, with the highest gamma di- rent of water vapor, or “flying river,” which
versity (regional species pool) of any fresh- contributes to life-giving precipitation in
water system; about 8 percent of all the fish other parts of South America (Arraut and
species on earth occur there. Even more at Nobre 2012). As trees fall, so do the levels
risk, the fishes of the Atlantic Forest, re- of reservoirs that feed São Paulo, Rio de Ja-
duced to 7 percent of its original extent, neiro, and other heavily populated south-
are poorly known as a result of incomplete eastern Brazilian cities, and agricultural
sampling, particularly in northern portions lands in Brazil and other nations to the west
of the forest domain. Few South American and the south. Unique assemblages of fish
242 W H AT DOE S THE FUTUR E HOL D?
species isolated in the last glacial period volumes are immense, yet these lakes and
have diversified through allopatric specia- their diverse fishes are varyingly vulnerable
tion in these watersheds, placing a wealth to climate change (Hecky et al. 2010). This
of endemic species at risk of extinction in variation in vulnerability owes much to the
the same mountains that provide the water way endemics evolved and to interactions
for the great cities of Brazil. It is important with nonclimate stressors, such as demands
to maintain forest cover in the Amazon to for freshwater by people seeking electric-
maintain regional climate stabilization of ity, food, and other economic returns. In
the Amazonian forest, which can reduce Lake Victoria, climate-related shifts in sur-
the likelihood of a downward spiral of dry- face temperature, seasonality, and winds
ing in response to global climate change. aggravate a history of overfishing, exotic
introductions, and eutrophication that col-
lectively have led to the mass extinction of
East Africa
at least half of the lake’s endemic species
East Africa’s Great Lakes boast the highest of 500–600 haplochromine cichlids. The
concentrations of endemic vertebrate spe- primary climate-related impacts to Africa’s
cies on earth; most of these endemics are Great Lakes are lake-level flux and stratifica-
fishes, with amphibians likely running a tion. Intensive rains can cause a rapid rise in
close second, and most of the endemic fishes lake level, or transgression, that submerges
are cichlids. The three largest lakes alone new habitats. Some of these habitats, such
host a species pool near equal to those of as rocky reefs, can harbor founder popula-
the whole of the Amazon Basin or Mekong tions of philopatric fishes that subsequently
Basin; with Tanganyika at 280, Malawi at speciate; mouth-brooding cichlids are fa-
approximately 1,000, and Victoria at about mous for this. Even so, rising waters may
600. Levels of endemism range from about drown shallow-water species, which may
80 percent to 95 percent. Lakes Turkana, lose important shallow-water habitat as
Albert, Edward, and George also claim ap- lake levels rise and spill over other kinds of
preciable diversity, as do the swamps, satel- bottom (e.g., Kaufman 1997, 2003; Sturm-
lite, and soda lakes associated with them. baur et al. 2000). Falling lake levels can also
The three greatest lakes sit atop three great pose a conservation challenge, leaving fish
watersheds, the Nile (Victoria), the Congo populations stranded in peripheral or satel-
(Tanganyika), and the Zambezi (Malawi); lite lakes (Greenwood 1965; Chapman et al.
each is home to endemic fishes and diverse 1996).
fish species assemblages. Miniradiations ex- In the Lake Kyoga Basin of central
ist in Lake Tana (the Blue Nile) and in hy- Uganda, a system of interconnected papy-
persaline soda lakes astride the Rift, such rus swamp lakes are the most important
as Lake Natron. Ironically, Africa is a high, refugium from the threat of mass extinc-
dry continent: arid conditions prevail over tion in Lake Victoria precipitated by climate
most it. Microendemic species occur in the change, habitat destruction, and invasive
tiny places where water is found, as in the species (Mwanja et al. 2001). A rise in Lake
deserts of North America, the Middle East, Kyoga caused it to merge with adjacent Lake
and Australia, complete with an endemic Nawampassa, exposing endangered cichlids
desert sinkhole cichlid (Tilapia guinasana) to the possibility of invasion and predation
and ephemeral pond rivulines (e.g., Notho- by the introduced Nile perch (Lates niloticus).
branchius spp.). All are at risk from changes in Connectivity was brief and Nawampassa
the frequency and intensity of precipitation. was has not been invaded . . . yet. Compari-
The African Great Lakes would seem to son between lakes in this system inhabited
be among those places least vulnerable to by, as compared to free of, perch show how
warming because their contained water the introduction of perch could result in the
F I N A L A R B I T E R O F T H E M A S S E X T I N C T I O N O F F R E S H WAT E R F I S H E S 243
extinction of endemic fishes (Chapman et possible to estimate, the two interact in
al. 1996; Schwartz et al. 2006). a nonadditive manner. This compounds
It is natural for lakes to stratify during systemic ills that arise from lack of atten-
warm periods and turn over in high winds tion to our dependence on natural systems.
or cooler weather, but global warming can Indeed, even if anthropogenic climate
heat surface waters and lock down strati- change were arrested immediately, fresh-
fication. If accompanied by eutrophication, water fishes would still be ailing. Hence (4)
the result will be a shallow, hyperproduc- the systemic nature of the problem chal-
tive epilimnion and a deadly, hypoxic hy- lenges species-centric conservation efforts,
polimnion, a recipe for fish die-offs from implying that the most realistic and prac-
the upwelling of deep, oxygen-poor water tical strategy for averting the global mass
to the shallows. Lake Victoria, not a rift extinction of freshwater fish species would
lake, comparatively shallow (87 m at most), be ecosystem based. That is, all proposed
and set in a broad, gradually sloping basin, developments must be reviewed in the
was dry as recently as 12,500 years ago, and context of an ecosystem-based manage-
even modest lake-level flux can shift shore- ment plan for human activities, rational
line position. In the deeper and steeper rift limits to growth, and most important, a
lakes, Tanganyika, Malawi, Edward, Albert, climate adaptation strategy, particularly for
and Turkana, the notion of an entire lake the freshwater supply.
drying up completely is unlikely, but these Freshwater fish conservation requires ac-
lakes have only modest inputs and outputs, tion at three scales—species, habitat, and
gaining water predominantly from rainfall ecosystem—but the ecosystem-level needs
and losing it mostly through evaporation. have been consistently ignored. Most cur-
Consequently, the fishes they harbor likely rent attention is directed toward the species
will respond to climate change, particularly level, ranging from protective legislation
species that occupy the littoral zone of lakes to breeding and reintroduction programs.
in shallow basins or lakes with shallow Examples of protective legislation include
shelf areas. the Convention on International Trade in
Endangered Species, special requirements
for possession or distribution, and national
CONC LUSION and local endangered species acts. Legisla-
tion is useful in avoiding unintended harm
These three fishy stories exemplify four and in preventing trade from becoming a
conclusions that can be drawn regarding hindrance to recovery. Captive breeding
the role of anthropogenic climate change programs now exist for the desert fishes
in determining the conservation status of of North America, the Lake Victoria cich-
freshwater fishes: (1) Human demand for lids, Cambodia’s dragonfish, Appalachian
water for personal consumption, agricul- stream fishes, and quite a few others. We
ture, industry, and waste disposal by itself have learned that such programs can work
poses an existential threat to a large per- to reverse extinction from the wild, that
centage of surviving freshwater fish spe- they are immensely difficult and expensive
cies. (2) Anthropogenic climate change to complete from first captivity to final re-
is altering the distribution, quality, and lease, that they only make sense where in-
availability of surface freshwater habitats tact habitat remains in order to justify the
and by itself threatens a large percentage considerable expense of propagation and
of freshwater fish species. (3) Although the reintroduction, and that we need a great
percentage of threatened or endangered many such programs in play, way too many
freshwater fishes attributable to water use to make this a practical primary approach
versus climate change is difficult or im- to safeguarding freshwater fish diversity.
244 W H AT DOE S THE FUTUR E HOL D?
Most concentrations of threatened fresh- tem services.” Ecology and Society 19 (4): 31. http://dx.doi
water fish diversity such as those discussed .org/10.5751/ES-06965-190431.
Freeman, M. C., G. R. Buel, L. E. Hay, W. B. Hughes, R.
in this chapter occupy areas that are already B. Jacobson, J. W. Jones, S. A. Jones, et al. 2012. “Link-
heavily populated and/or undergoing rapid ing river management to species conservation using
development. This greatly elevates the im- dynamic landscape-scale models.” River Research and Ap-
portance that development be carefully plications 29: 906–918.
planned, monitored, and guided in an adap- Greenwood, P. H. 1965. “The cichlid fishes of Lake
Nabugabo, Uganda.” Bulletin of the British Museum 12 (9).
tive fashion. There are very few examples in Hansen, J., P. Kharecha, M. Sato, V. Masson-Delmotte,
the world, or in history, of a human society F. Ackerman et al. 2013. “Dangerous climate change:
behaving in this manner. Conservation or- Required reduction of carbon emissions to protect
ganizations, watershed management coun- young people, future generations, and nature.” PLOS
cils, and place-based systems of governance One 8 (10): 1–26.
Hecky, R., R. Mugidde, P. S. Ramlal, M. R. Talbot, and G.
must focus their efforts on learning how to W. Kling. 2010. “Multiple stressors cause rapid eco-
do this properly. Otherwise, all the well-in- system change in Lake Victoria.” Freshwater Biology 55
tended legislative and captive breeding pro- (S1): 19–42.
grams are doomed to fail. In the immedi- Hulsey, C. D., J. Marks, D. A. Hendrickson, C. A. Wil-
ate future the question is one of how much liamson, A. E. Cohen, and M. J. Stephens. 2006. “Feed-
ing specialization in Herichthys minckleyi: A trophically
of the wealth of freshwater fishes we can polymorphic fish.” Journal of Fish Biology 68: 1–12.
carry through the hysteresis in global cli- Kaufman, L. S. 1992. “Catastrophic change in species-
mate change. Scientists have written a clear rich freshwater ecosystems: The lessons of Lake Victo-
prescription for sustainable, ecosystem- ria.” Bioscience 42: 846–858.
based approaches to human development. Kaufman, L. S. 1997. “Asynchronous taxon cycles in hap-
lochromine fishes of the greater Lake Victoria region.”
These are ways of living that seek to main- South African Journal of Science 93: 601–606.
tain species, restore ecosystem services, and Kaufman, L. S. 2003. “Evolutionary footprints in eco-
put the lid on the runaway growth of hu- logical time: Water management and aquatic con-
man demands and on the economies that servation in African lakes.” In Conservation, Ecology, and
stoke these demands beyond reason. The Management of African Freshwaters, ed. T. L. Crisman, L. J.
Chapman, C. A. Chapman, and L. S. Kaufman, 460–
medicine is sugarcoated with strong logic, 490. University Press of Florida.
powerful incentives, and the love of our Lin, Hsien-Yung, Alex Bush, Simon Linke, Hugh P. Pos-
children. We have but to take it to heart. singham, and Christopher J. Brown. 2017. “Climate
change decouples marine and freshwater habitats of
a threatened migratory fish.” Diversity and Distributions 23
(7): 751–760.
R E FE R E NCE S Lundberg, J. G., J. P. Sullivan, R. Rodiles-Hernandez, and
D. A. Hendrickson. 2007. “Discovery of African roots
Arraut, M. A., C. Nobres, H. M. J. Barbosa, G. Obregon for the Mesoamerican Chiapas catfish, Lacantunia enig-
and J. Marengo. 2012. “Aerial rivers and lakes: Look- matica, requires an ancient intercontinental passage.”
ing at large-scale moisture transport and its relation to Proceedings of the Academy of Natural Sciences of Philadelphia 156:
Amazonia and to subtropical rainfall in South Amer- 39–53. https://doi.org/10.1635/0097-3157(2007)156
ica.” Journal of Climate 25: 543–556. [39: DOARFT]2.0.CO;2.
Chapman, L. J., C. A. Chapman, R. Ogutu-Ohwayo, M. Mulholland, P. J., G. R. Best, C. C. Coutant, G. M. Horn-
Chandler, L. Kaufman, and A. E. Keiter. 1996. “Refu- berger, J. L. Meyer, P. J. Robinson, J. R. Steinberg, R.
gia for endangered fishes from an introduced preda- E. Turner, F. Vera-Herrera, R. G. Wetzel. 1997. “Ef-
tor in Lake Nabugabo, Uganda.” Conservation Biology 10: fects of climate change on freshwater ecosystems of
554–561. the southeastern United States and the Gulf coast of
Comte, L., and G. Grenouillet. 2013. “Do stream fishes Mexico.” Hydrological Processes 11: 949–970.
track climate change? Assessing distribution shifts in Mwanja, W. W., A. S. Armoudlian, S. B. Wandera, L.
recent decades.” Ecography 36 (11): 1236–1246. Kaufman, L. Wu, G. C. Booton, and P. A. Fuerst. 2001.
Downing, A. S., E. Van Nes, J. Balirwa, J. Beuving, P. “The bounty of minor lakes: The role of small satellite
Bwathondi, L. J. Chapman, I. J. M. Cornelissen, et al. water bodies in evolution and conservation of fishes.”
2014. “Coupled human and natural system dynamics Hydrobiologia 458: 55–62.
as key to the sustainability of Lake Victoria’s ecosys-
F I N A L A R B I T E R O F T H E M A S S E X T I N C T I O N O F F R E S H WAT E R F I S H E S 245
Newbold, Tim, Lawrence N. Hudson, Andrew P. Arnell, Silsbe. 2015. “The Nile perch invasion in Lake Vic-
Sara Contu, Adriana De Palma, Simon Ferrier, and Sa- toria: Cause or consequence of the haplochromine
mantha L. Hill. 2016. “Has land use pushed terrestrial decline?” Canadian Journal of Fisheries and Aquatic Sciences.
biodiversity beyond the planetary boundary? A global https://doi.org/10.1139/cjfas-2015-0130.
assessment.” Science 353 (6296): 288–291. Winemiller, Kirk O., P. B. McIntyre, L. Castello, E. Fluet-
Schwartz, J. D. M., M. J. Pallin, R. H. Michener, D. Mba- Chouinard, T. Giarrizzo, S. Nam, I. G. Baird, et al.
bazi, and L. S. Kaufman. 2006. “Effects of Nile perch, 2016. “Balancing hydropower and biodiversity in
Lates niloticus, on functional and specific fish diversity the Amazon, Congo, and Mekong.” Science 351 (6269):
in Uganda’s Lake Kyoga system.” African Journal of Ecology 128–129.
44 (2): 145–156. Xenopoulos, M., D. M. Lodge, J. Alcamo, M. Marker, K.
Shapiro, J. 2001. Mao’s War Against Nature: Politics and the Schultze, and D. P. Van Vuurens. 2005. “Scenarios of
Environment in Revolutionary China. Cambridge University freshwater fish extinction from climate change and
Press. water withdrawal.” Global Change Biology 11: 1–8.
Van Zwieten, P., J. Kolding, M. J. Plank, R. E. Hecky, T.
B. Bridgeman, S. MacIntyre, O. Seehausen, and G. M.
CHAPTER NINETEEN I N T RODUCTION
246
T H E A S Y M M E T R IC A L I M PAC T S OF C L I M AT E C H A NGE ON FO OD W E B S 247
CONSU M E R-R E SOU RC E DY NA M IC S sion efficiency e) by the consumer, and the
A N D T H E CONSE QU E NC E S OF consumer mortality rate (m) (Figure 19.1).
ASY M M E T R IC A L-T R A IT R E SPONSE S Hereafter, these parameters are referred
to as functions of temperature to empha-
Temperature affects nearly every biologi- size that it is their response to tempera-
cal process. Scientists relate effects of tem- ture that collectively determines biomass
perature on the outcomes of cellular meta- and dynamic responses to climate warm-
bolic processes to whole animal functions ing. Individual trait performances manifest
using thermal performance curves (TPCs) in the overall expression of each aspect to
(Huey and Stevenson 1979). TPCs depict determine how energy and material flow
how biological traits respond across a range through the interaction and thus determine
of relevant temperatures, describing the dynamic outcomes of climate warming.
breadth of thermal tolerance and providing Climate-driven asymmetries make a trait-
estimates of “optimal” temperatures that based approach seem daunting; however,
yield maximal trait performance (Huey and the theoretical framework proposed by Gil-
Kingsolver 1989). Asymmetries arise at the bert et al. (2014) uses bioenergetic reason-
trait level when multiple traits differentially ing to aggregate these key traits to assess the
respond to temperature (Dell et al. 2013). temperature dependence of trophic interac-
For example, a study of experimental algae tions, and thus the structural (biomass ra-
(Chlorella vulgaris) populations found that the tios) and dynamic (stability) implications
intrinsic per capita growth rate (r) showed of change. An understanding of which dy-
a hump-shaped response to temperature, namics drive changes in abundance is es-
whereas carrying capacity (K) remained sential to incorporate temperature-depen-
unchanged across biologically relevant dent rates into estimates of how consumers
temperature ranges (Jarvis et al. 2016). In and resource abundance or biomass will be
this example, growth rate is the dominant altered by future climate.
trait to act on the population at any given Imagine that the biomass of the con-
temperature, although multiple traits, with sumer in Figure 19.1 is a balloon. The size of
various TPCs, may interact to yield a vari- this balloon (i.e., biomass of the consumer)
ety of population dynamics under changing depends on three things: how much air is
thermal stress. Other empirical studies have blown into the balloon; the elasticity of the
identified asymmetries in consumer and re- balloon (i.e., how easily it expands); and
source growth, activity rates, metabolism, the size of the leak in the balloon. Note that
ingestion, and feeding rates—asymmetries production (via growth rate r, or carrying
that shift interactions over a thermal gradi-
ent (Rall et al. 2010; O’Connor 2009; Rose
Figure 19.1. Three key components of production (R
and Caron 2007; Dell et al. 2011; Dell et al.
growth), energy transfer (from R to C) and loss (mortal-
2013; Monaco et al. 2017). Recognizing this ity or metabolic costs) contribute to the size of the con-
dynamic, Gilbert et al. (2014) developed a sumer biomass or the size of the consumer “balloon.”
simple theory to identify how asymmetri-
cal trait responses influence the dynamics
and composition of consumer-resource
interactions.
In consumer-resource interactions,
warming affects the production of the re-
source (growth rate r and carrying capac-
ity K), the consumption and assimilation
of the resource (attack rate a and conver-
248 W H AT DOE S THE FUTUR E HOL D?
capacity K), energy transfer (via attack rate expect a priori that one parameter will al-
a or conversion efficiency e) and energy loss ways dominate; similarly, there is no reason
(i.e., mortality m) respectively govern each to expect all parameters to be affected iden-
of these attributes. As this chapter proceeds
to navigate the theory behind trait (r, K, a, e,
Figure 19.2. Hypothetical cases of trait responses to tem-
and m) responses to climate, keep in mind perature (left) and subsequent changes to C-R interac-
that if relatively more air is entering the tions (right). In (A), the thermal performances of ae
“consumer biomass” balloon than is leaving (solid line) and m (dotted line) are equal, thus the influx
it, the balloon will expand, and vice versa. of energy to the consumer (ae) is similar to the energy
Understanding the effects of production, lost (m) across increasing temperatures and temperature
has no impact on the C-R interaction. (B) is showing
energy transfer, and energy loss on con- a case where both ae and m have the same optimum
sumer biomass can help explain how the temperature, but warming has a stronger effect on ae
implications of climate warming ultimately than m such that higher temperatures increase the C-R
depend on whether the traits are affected interaction. In the final case (C), ae and m have differen-
symmetrically (Figure 19.2A) or asymmet- tial temperature responses in that they peak at different
temperatures, causing stronger interactions at interme-
rically (Figure 19.2B). There is no reason to diate temperatures and weaker interactions at relatively
high temperatures.
T H E A S Y M M E T R IC A L I M PAC T S OF C L I M AT E C H A NGE ON FO OD W E B S 249
tically. Figure 19.2 outlines three different e[Temp]/m[Temp]) varies with temperature,
scenarios. and whether asymmetries are common
To put asymmetrical trait responses into in one direction or another, is poorly un-
context, a recent study by Bost et al. (2015) derstood. Increases in consumer standing
found that temperature had an impact on stocks may often be an early but not lasting
the population sizes of king penguins (Ap- signal of warming (Rose and Caron 2007;
tenodytes patagonicus) by altering their forag- O’Connor et al. 2011).
ing behavior (and thus their attack rate). The example in Figure 19.1 illustrates
Temperature shifts associated with tropical how understanding relative thermal per-
anomalies pushed prime foraging habitats formance of traits can explain how the
farther south and increased the depth of the abundance of a species in a C-R interaction
ocean’s thermocline (the depth at which responds to temperature. Abundance or bio-
temperature decreases rapidly). These shifts mass typically and tractably are measured
forced the penguins to travel farther and in experiments or in the field, and changes
dive deeper for food, consequently increas- through time are often used to infer future
ing the total time spent foraging. Though trajectories of persistence and extinction,
not part of the study, subsequent reductions although a dynamic approach reminds us
in attack rate and conversion efficiency (re- that temporal trajectories in abundance
call energy transfer) means less energy is alone are insufficient to predict the future.
available to be allocated to reproduction. For example, compare the patterns of rela-
Assuming that mortality rates remain con- tive abundance of consumers and resources
stant, warmer temperatures asymmetrically in Figures 19.2. The decrease in consumer
altered consumer traits (reduced attack biomass at high temperatures (Figure 19.2)
rates but not mortality) and, in this case, reflects the reversal in the temperature de-
decreased population sizes. pendence of the underlying ratio of ae/m,
This framework suggests that under- even though the C-R biomass ratio (C/R) at
standing the relationship between tempera- lower temperatures was indistinguishable
ture dependence of consumer traits, a(Temp) from the pattern in Figure 19.2. The bio-
and e(Temp), and consumer mortality rates, energetic framework outlined here, with
m(Temp), provides insight into how dynamic information on the thermal asymmetry
systems respond to warming, even with- of ae/m, could be used to predict temporal
out information on values (a, e, m, r, and trends in C/R with warming through time.
K) or thermal dependence of each param- So far the effects of temperature on in-
eter (a(Temp), e(Temp), m(Temp), r(Temp), and teractions have been considered without
K(Temp)). Broadly synthesized empirical evi- the consideration of temperature-depen-
dence suggests that across species, mortal- dent resource population growth (r and
ity and attack rates both increase similarly K), each of which may vary with tempera-
with temperature (McCoy and Gillooly ture; and there is little reason to think they
2008; Englund et al. 2011), although Dell are thermally symmetric with consumer
et al. (2011) suggest a moderate asymme- rates (a and m). For plant-grazer (or algal-
try in which mortality is less sensitive to grazer) C-R interactions, primary produc-
warming. For the outcome of dynamic sys- tion tends to be less sensitive to warming
tems, the relative temperature dependences than secondary production, which suggests
of these rates will reflect not only broad an asymmetry between consumer and re-
metabolic constraints but also species-spe- source rates (Allen et al. 2005; O’Connor
cific temperature performance curves for a et al. 2011). Another source of consumer-
given system (Englund et al. 2011; Dell et al. resource thermal asymmetry is an animal’s
2013). Within C-R systems, how the ratio velocity (Dell et al. 2011, 2013). Differences
of energy gain to energy loss (i.e., a[Temp] in the mobility of consumers and their prey
250 W H AT DOE S THE FUTUR E HOL D?
Coefficient of Variation
CV
(CV)
and production (r, K) to form ratios that de-
λMAX
termine biomass flux across temperature
gradients, thus enabling us to track the
C:R Biomass Ratio C:R Biomass Ratio
flow of energy between interacting species
Figure 19.3. In a simple C-R model (e.g., type I func- and the impact of change on relative abun-
tional response), the C:R biomass ratio correlates to in- dances and stability.
creased top-down control. Here, shown by measuring
the change in R density (called log ratio; see Shurin et al.
2002) with and without the consumer (A). A top-heavy
biomass ratio drives a large cascade. Similarly, high C:R C L I M AT E C H A NGE A N D ASY M M E T R IC
biomass ratios also tend to drive heightened instability COM PA RT M E N T R E SPONSE S
(B). For more details on these results in numerous mod-
els, see McCann (2012) and Gilbert et al. (2014). The theoretical exercise suggested a rela-
tively straightforward way to understand
can generate thermal asymmetries in the how a given C-R interaction can be affected
rates underlying consumption, and when by warming. To scale up to whole food
these differ from population growth rates webs is difficult. Certain properties inher-
the framework here suggests shifts in inter- ent in real food webs may allow an infor-
action outcomes. mative macroscopic glimpse of how climate
Several important ecological proper- warming ought to affect the energy flow of
ties are altered with biomass accumulation whole webs. However, different habitats of-
changes. First, increasing consumer bio- ten have unique characteristics (i.e., organ-
mass in simple C-R theory tends to equate ismal sizes, habitat conditions) that likely
with the strength of top-down control mediate the impacts of climate change dif-
(Figure 19.3A). That is, the larger the C:R ferently. If true, then climate change may
biomass ratio, the greater the release of fundamentally alter energy flow through
resource biomass (represented by the re- whole ecosystems.
source log ratio; Shurin et al. 2002) when Empirical results suggest that food webs
the consumer is removed. In other words, typically have a gross architecture in which
with increased C:R ratios, the strength of lower trophic level habitats are compart-
the trophic cascade and top-down control mentalized (species relatively isolated) and
grows (Gilbert et al. 2014; DeLong et al. ultimately are coupled through foraging to
2015). Such a change in consumer biomass higher-trophic-level predators, which tend
affects the stability of the interaction (Fig- to be more mobile (Figure 19.4). Mobile
ure 19.3B): there is a strong tendency for organisms tend to be generalist foragers
top-heavy interactions (i.e., high C:R bio- capable of consuming a variety of distinct
mass ratios) to be less stable (McCann 2012; prey types because, all else equal, mobil-
Gilbert et al. 2014). ity exposes consumers to a greater number
It is evident that specific impacts of of habitats and prey types (i.e., encounter
warming are context dependent; however, probability is positively related to mobil-
two important points are addressed here: ity; Pyke et al. 1977). This correlation be-
First, the relative response of entire suites tween size and generalism suggests that
of traits govern general dynamic outcomes each progressively higher trophic position
of climate change, and second, climate will increasingly use different habitats
change can alter the structure or architec- (Pimm and Lawton 1977). Increased gen-
tural framework of systems and subsystems eralism at higher trophic levels increases
via interacting traits. To simplify the vex- the likelihood for omnivory, defined here
ing problem of simultaneously responding as feeding at more than one trophic level
T H E A S Y M M E T R IC A L I M PAC T S OF C L I M AT E C H A NGE ON FO OD W E B S 251
ASY M M E T R IC A L F O OD -W E B
P COM PA RT M E N T R E SPONSE S TO
C L I M AT E C H A NGE
pler’s” behavior. Altered behavior could creased the absolute temperature at each
fundamentally alter the flow of energy even elevation of the vegetation canopy and set
if things like compartmental productivity new spatial constraints, thus altering the
or energy transfers between trophic levels habitat domain of the thermally sensitive
do not change. active hunting spider (vulnerable to the in-
Taken together, climate change is ex- creased temperatures at higher elevations)
pected to have asymmetric impacts on dif- and causing overlap of the two species. This
ferent food web compartments that could resulted in intraguild predation and extinc-
drastically change energy flow. Below are tion of the active species (Figure 19.5B).
some simple empirical examples—one Climate warming drives asymmetries in
aquatic and one terrestrial—to show that compartmentalized food webs by differen-
some emerging data agree with this food- tially warming alternate habitats and affect-
web-scale perspective on climate change. ing trait-mediated responses of species at
both the physiological and the behavioral
levels. Warming can alter the architectural
E M E RGI NG E M PIR IC A L E X A M PL E S framework of food webs via changes in
OF ASY M M E T R IC W HOL E biomass production (e.g., increased growth
F O OD -W E B I M PAC TS rates, thermal refugia for prey species) and
top-down control.
In lake ecosystems, as air temperature in-
creases, shallow, near-shore (littoral)
habitats warm more quickly than deep, DIS C US SION
offshore (pelagic) habitats. Tunney et al.
(2014) found that cold-adapted predatory Context-dependent, TPC-driven response
lake trout (Salvelinus namaycush) are increas- of C-R interactions to climate change can
ingly restricted from accessing the littoral yield complex outcomes (see Figure 19.2C).
carbon pool when temperatures exceed that This suggests that warming associated with
of optimal foraging. Across 50 boreal lake climate change acts on asymmetric thermal
ecosystems, warming had a “decoupling” responses of individual species traits to al-
effect, meaning that lake trout diet became ter whole “compartment” responses that
increasingly derived from pelagic habitats affect the gross architecture of entire food
in warmer lakes, independent of prey abun- webs due to habitat heterogeneity and the
dance in either habitat (Figure 19.5A; see physiology and behavior of resident species
Guzzo et al. 2017 for similar patterns found (Figure 19.4).
temporally). This differential warming and Empirical patterns in gross food-web
subsequent shifts in species behavior are energy flows suggest that webs are often
not confined to aquatic ecosystems. Barton compartmentalized at lower trophic levels
and Schmitz (2009) documented a similar and coupled by the behavior of higher-
behavioral shift for a predatory spider spe- order generalists. This structure is often
cies in a grassland food web with a vertical “adaptive” and stabilizing to changing con-
temperature differential in the vegetation ditions, as one compartment serves as a
canopy (temperatures increase from the buffer to the pressures that act on another
ground surface to the top of the canopy). (e.g., overconsumption, seasonal fluxes).
They examined two spider species that dif- Evolution has equipped species for habitat-
fered in foraging strategy (ambush sit-and- specific conditions, and this compartmen-
wait vs. active hunting) and principal habi- talized or channeled food web structure
tat (the ambush species, Pisaurina mira, resided sets up the possibility for food webs to be
lower in the canopy than the active hunter, affected asymmetrically by change and so
Phidippus rimator). Experimental warming in- potentially altered in terms of their gross
T H E A S Y M M E T R IC A L I M PAC T S OF C L I M AT E C H A NGE ON FO OD W E B S 253
−2 0
−2
−4
−6 −4
15 16 17 18 19 20 21 15 16 17 18 19 20 21
C D
Trophic position (mean and max)
7 1.0
(Proportion of total catch <6m)
Near-shore lake trout captures
6 0.8
5 0.6
4 0.4
3 0.2
2 0
15 16 17 18 19 20 21 15 16 17 18 19 20
ο
Summer air temperature ( C)
Figure 19.5. Empirical data from Tunney et al. (2014), gross architecture of many aquatic and ter-
showing the impacts of climate warming on food web restrial webs suggests that their inherently
structure. (A) shows that there is evidence for a decrease
in near- and offshore coupling by lake trout, (B) shows
spatial structuring can generate different re-
a decrease in the contribution of nearshore fish to lake sponses that completely reroute energy and
trout diet, (C) shows an increase in trophic position, and carbon flow in whole ecosystems.
(D) shows a decrease in lake trout presence in nearshore
habitats with increasing temperatures.
R E FE R E NCE S
energy flow patterns (due to changes in,
for example, primary production or con- Allen, A. P., J. F. Gillooly, and H. Brown. 2005. “Link-
sumption rates). Changes can alter the be- ing the global carbon cycle to individual metabolism.”
Functional Ecology 19: 202–213. https://doi.org/10.1111/j
havior of larger, mobile species, potentially .1365-2435.2005.00952.
restricting accessibility to alternate habitats Barton, B. T., and O. J. Schmitz. 2009. “Experimen-
and eliminating the buffering or stabilizing tal warming transforms multiple predator effects in
properties of compartmentalized webs. a grassland food web.” Ecology Letters 12: 1317–1325.
Food webs are complex, and the fate of https://doi.org/10.1111/j.1461-0248.2009.01386.x.
Bost, C. A., C. Cotté, P. Terray, C. Barbraud, C. Bon, K.
our ecosystems in response to climate change Delord, O. Gimenez, Y. Handrich, Y. Naito, C. Gui-
depends not only on individual species or in- net, and H. Weimerskirch. 2015. “Large-scale climatic
dividual interactions but also on the collec- anomalies affect marine predator foraging behaviour
tive response of the whole system to change. and demography.” Nature Communications 6. https://doi
This chapter shows that a first glimpse of the .org/10.1038/ncomms9220.
254 W H AT DOE S THE FUTUR E HOL D?
Dell, A. I., S. Pawar, and V. M. Savage. 2013. “Tempera- ecosystem and implications for stability.” Marine Ecol-
ture dependence of trophic interactions are driven by ogy Progress Series 482: 17–28. https://doi.org/10.3354/
asymmetry of species responses and foraging strat- meps10278.
egy.” Journal of Animal Ecology 83: 70–84. https://doi.org/ Monaco, C. J., C. D. McQuaid, and D. Marshall. 2017.
10.1111/1365-2656.12081. “Decoupling of behavioural and physiological ther-
Dell, A. I., S. Pawar, and V. M. Savage. 2011. “Systematic mal performance curves in ectothermic animals: A
variation in the temperature dependence of physi- critical adaptive trait.” Oecologia 185: 583–593. https://
ological and ecological traits.” Proceedings of the National doi.org/10.1007/s00442-017-3974-5.
Academy of Sciences 108: 10591–10596. https://doi.org/10 O’Connor, M. I. 2009. “Warming strengthens an herbi-
.1073/pnas.1015178108. vore-plant interaction.” Ecology 90: 388–398. https://
Englund, G., G. Ohlund, C. L. Hein, and S. Diehl. 2011. doi.org/10.1890/08-0034.1.
“Temperature dependence of the functional re- O’Connor, M. I., B. Gilbert, and C. J. Brown. 2011. “The-
sponse.” Ecology Letters 14: 914–921. https://doi.org/10 oretical predictions for how temperature affects the
.1111/j.1461-0248.2011.01661.x. dynamics of interacting herbivores and plants.” Ameri-
Fussmann, K. E., F. Schwarzmuller, U. Brose, A. Jousset, can Naturalist 178: 626–638. https://doi.org/10.1086/
and B. C. Rall. 2014. “Ecological stability in response 662171.
to warming.” Nature Climate Change 4: 206–210. https:// Pimm, S. L., and J. H. Lawton. 1977. “Number of tro-
doi.org/10.1038/nclimate2134. phic levels in ecological communities.” Nature 268:
Gilbert, B., T. D. Tunney, K. S. McCann, J. P. DeLong, D. 329–331. https://doi.org/10.1038/268329a0.
A. Vasseur, V. M. Savage, J. B. Shurin, A. Dell, B. T. Pyke, G. H., H. R. Pulliam, and E. L. Charnov. 1977. “Op-
Barton, C. D. G. Harley, P. Kratina, J. L. Blanchard, C. timal foraging: A selective review of theory and tests.”
Clements, M. Winder, H. S. Greig, and M. I. O’Connor. Quarterly Review of Biology 52: 137–154. http://www.jstor
2014. “A bioenergetics framework for the temperature .org/stable/2824020.
dependence of trophic interaction strength.” Ecology Rall, B. C., O. Vucic-Pestic, R. B. Ehnes, M. Emmerson,
Letters 17: 1–12. https://doi.org/10.1111/ele.12307. and U. Brose. 2010. “Temperature, predator-prey
Gillooly, J. F., J. H. Brown, G. B. West, V. M. Savage, and interaction strength and population stability.” Global
E. L. Charnov. 2001. “Effects of size and temperature Change Biology 16: 2145–2157. https://doi.org/10.1111/
on metabolic rate.” Science 293: 2248–2251. https://doi j.1365–2486.2009.02124.x.
.org/10.1126/science.1061967. Rooney, N., K. S. McCann, and J. C. Moore. 2008. “A land-
Guzzo, M., P. J. Blanchfield, and M. D. Rennie. “Behav- scape theory for food web architecture.” Ecology Let-
ioural responses to annual temperature variation alter ters 11: 867–881. https://doi.org/10.1111/j.1461-0248
the dominant energy pathway, growth, and condi- .2008.01193.x.
tion of a cold-water predator.” PNAS. https://10.1073/ Rooney, N., K. S. McCann, G. Gellner, and J. C. Moore.
pnas.1702584114. 2006. “Structural asymmetry and the stability of di-
Huey, R. B., and J. G. Kingsolver. 1989. “Evolution of verse food webs.” Nature 442: 265–269. https://doi
thermal sensitivity of ectotherm performance.” Tree .org/10.1038/nature04887.
4: 131–135. https://doi.org/10.1016/0169-5347(89) Rose, J. M., and D. A. Caron. 2007. “Does low tempera-
90211-5. ture constrain the growth rates of heterotrophic pro-
Huey, R. B., and R. D. Stevenson. 1979. “Integrating ther- tists? Evidence and implications for algal blooms in
mal physiology and ecology of ectotherms: A discus- cold waters.” Limnology and Oceanography 52: 886–895.
sion of approaches.” American Zoologist 19: 357–366. https://doi.org/10.4319/lo.2007.52.2.0886.
Jarvis, L., K. S. McCann, T. Tunney, G. Gellner, and J. Shurin, J. B., E. T. Borer, E. W. Seabloom, K. Anderson,
Fryxell. 2016. “Early warning signals detect critical C. A. Blanchette, B. Broitman, S. D. Cooper, and B.
impacts of experimental warming.” Ecology and Evolution S. Halpem. 2002. “A cross-ecosystem comparison of
6: 6097–6106. https://doi.org/10.1002/ece3.2339. trophic cascades.” Ecology Letters 5: 785–791. https://doi
Kraemer, B. M., S. Chandra, A. I. Dell, M. Dix, E. Kuu- .org/10.1046/j.1461-0248.2002.00381.x.
sisto, D. M. Livingstone, S. G. Schladow, E. Silow, L. Tunney, T., K. S. McCann, N. Lester, and B. Shuter. 2014.
M. Sitoki, R. Tamatamah, and P. B. McIntyre. 2016. “Effects of differential habitat warming on complex
“Global patterns in lake ecosystem responses to communities.” Proceedings of the National Academy of Sci-
warming based on the temperature dependence of ences. 111: 8077–8082. https://doi.org/10.1073/pnas
metabolism.” Global Change Biology. https://doi.org/10 .1319618111.
.1111/gcb.13459. Vadeboncouer, Y. M., M. J. Vander Zanden, and D. M.
McCann, K. S. 2012. Food Webs. Princeton University Press. Lodge. 2002. “Putting the lake back together: Reinte-
McCoy, M. W., and J. F. Gillooly. 2008. “Predicting natu- grating benthic pathways into lake food web models.”
ral mortality rates of plants and animals.” Ecology Let- BioScience 52: 44–54. https://doi.org/10.1641/0006
ters 11: 710–716. https://doi.org/10.1111/j.1461-0248 -3568(2002)052[0044:PTLBTR]2.0.CO;2.
.2008.01190.x.
McMeans, B. C., N. Rooney, M. T. Arts, and A. T. Fisk.
2013. “Food web structure of a coastal Arctic marine
CASE STUDY 8
Global oceans have warmed over the past 100 years by an average of approximately 0.6°C,
however, ocean temperatures are not increasing homogeneously, as some areas are warming
even more rapidly than the average (Hobday and Pecl 2014). These rapidly warming areas
are concentrated against coastal margins where western boundary currents move water
poleward. This long-term warming moves isotherms poleward and shifts ocean habitats—
as a result, many changes in species’ distribution have been reported from these regions. In
areas with a strong north-south current axis, this “climate scale” habitat shift also occurs
seasonally, with the advance and retreat of currents. For example, the seasonal variation in
the East Australian Current off southeastern Australia is superimposed on the long-term
southward extension, where warming is about four times the global average. Such dynamic
and fast-warming areas represent an obvious location for developing and testing climate-
proof management approaches (see Hobday et al. 2016).
255
256 W H AT DOE S THE FUTUR E HOL D?
Authority. The distribution of likely south- that underpin this dynamic management
ern bluefin tuna (SBT) habitat, which can approach are not mission-critical-supported
change rapidly with the movement of the products, and occasional failures have pre-
East Australian Current, has been used vented on-time delivery of southern bluefin
to dynamically regulate fisher access to tuna habitat reports. This risk is mitigated
East Coast fishing areas, by dividing the by close engagement with the authority
ocean into a series of zones based on the to ensure that, in the event of disruption,
expected distribution of southern bluefin historical patterns in the distribution of the
tuna. A habitat preference model was used habitat zones can be coupled with the sea-
to provide near-real-time advice to man- sonal forecasts and the management system
agement about likely tuna habitat, deliv- is not compromised.
ered in the form of regular reports (every In operational use by the Australian
1–2 weeks) to the authority during the Fisheries Management Authority from
fishing season (Hobday et al. 2010). Man- 2003 to 2015, this habitat model evolved
agers used these habitat preference reports from one based on surface temperature
to frequently update spatial restrictions to to an integrated surface and subsurface
fishing grounds. These restrictions limit model, before finally including a seasonal
unwanted interactions by fishers that do forecasting element to aid managers and
not hold SBT quota (SBT cannot be landed fishers in planning for future changes in
without quota and in that situation must the location of the habitat zones (Hob-
be discarded) and allow access to those day et al. 2016). The continual improve-
that do have SBT quota to operate effi- ment and adaptation of the system saw
ciently (Hobday et al. 2010), and they are new oceanographic products tested and
enforced by vessel monitoring systems and included in the operational model. Stake-
fisheries observers. holder engagement has been critical to the
The data requirements for this dynamic development, validation, and refinement
habitat prediction system are informa- of the system (Hobday et al. 2010). For ex-
tion on the temperature preference profile ample, incorporating a seasonal forecast-
from SBT in the study area provided from ing component was an important step in
electronic tags, near-real-time sea surface informing and encouraging managers and
temperature from satellites, and near-real- fishers to think about decisions on longer
time subsurface temperatures from an op- time scales (Hobday et al. 2016). The dy-
erational ocean model. Efficient computing namic approach reduced the need for clo-
systems support real-time delivery, for ex- sures of large areas while still meeting the
ample, with temperature preference profiles management goal, but required that more
obtained from a tagging database of up-to- flexible fishing strategies be developed,
date data from electronic tags deployed on including planning vessel movements,
southern bluefin tuna in the study area. The home-port selection, and quota purchase.
habitat prediction system integrates these From 2015, the authority has opted to use
components to produce a two-dimensional video monitoring of the fleet to ensure
map of the habitat preference in near real- accurate reporting of catch of all species,
time (see Plate 7). Following quality con- SBT included, and reliance on the habitat
trol, reports on the current southern bluefin model has declined.
tuna habitat prediction in the fishery area This example from Australia’s East Coast
are provided to fisheries managers (see Plate is one of a growing set of dynamic ocean
7), who make a decision about changing management applications for fisheries, con-
management zones within hours of receiv- servation, and shipping. Under dynamic
ing the reports. The ocean data products management, rather than closing off large
DY NA M IC SPAT I A L M A NAGE M E N T I N A N AUST R A L I A N T U NA F I SH E RY 257
areas, managers can integrate information R E FE R E NCE S
from habitat models, recent interactions,
and historical knowledge to dynamically Hobday, A. J., J. R. Hartog, T. Timmis, and J. Fielding.
2010. “Dynamic spatial zoning to manage southern
close smaller areas for shorter periods of bluefin tuna capture in a multi-species longline fish-
time. At longer time scales, similar strate- ery.” Fisheries Oceanography 19 (3): 243–253.
gies can be implemented to minimize over- Hobday, A. J., and G. T. Pecl. 2014. “Identification of
laps of protected and desired fish species. global marine hotspots: Sentinels for change and
Climate-aware conservation planners may vanguards for adaptation action.” Reviews in Fish Biol-
ogy and Fisheries 24: 415–425. https://doi.org/410.1007/
also consider the use of mobile protected s11160-11013-19326-11166.
areas, defined by dynamic habitat models, Hobday, A. J., C. M. Spillman, J. P. Eveson, and J. R. Har-
to afford protection to species changing tog. 2016. “Seasonal forecasting for decision support
their distribution in response to long-term in marine fisheries and aquaculture.” Fisheries Oceanog-
ocean change. raphy 25 (S1): 45–56.
CHAPTER TWENTY I N T RODUCTION
I N VASI V E SPE C IE S A N D
C L I M AT E C H A NGE
258
I N VA S I V E S P E C I E S A N D C L I M AT E C H A N G E 259
to 45 percent in the western United States, the distribution of existing populations of
increasing the risk of wildfires across the invasive species, (4) altering the impacts of
continent and reducing native biodiversity existing populations of invasive species, and
(Bradley 2009). In India, the geographi- (5) altering the effectiveness of manage-
cal distribution of the invasion range of ment of invasive species. We explore each
the giant African snail (Achatina fulica) is not of five dimensions proposed by Hellmann
expected to change, but already-invaded et al. (2008) in turn, with an eye to litera-
areas are expected to be more prone to fu- ture emerging since 2008. Specifically, we
ture invasions (Sarma et al. 2015). In the explore new evidence, counterexamples,
oceans, international shipping has been an and more nuanced expectations that have
important vector for invasive species, with arisen for each dimension since Hellmann
84 percent of marine ecoregions report- et al. (2008) was published. This chapter is
ing marine invasions (Molnar et al. 2008). intended to provide readers with an intro-
Climate change impacts on global shipping duction to the impacts of invasive species
routes may increase the propagule pressure and climate change on biodiversity; it is not
in the oceans, exacerbating current inva- a comprehensive review. We conclude the
sion problems. chapter with some possible solutions for
Although these impacts have been clearly mitigating the climate change–related im-
documented for select invasive species, the pacts of invasive species.
impacts of climate change for others are less
clear. Invasive ant populations are expected
to respond in a variety of ways to climate A LT E R E D I N T RODUCTION
change, with climate suitability models A N D T R A NSPORT
predicting that some species will expand
their ranges and others will contract (Ber- Climate change may alter the way that
telsmeier et al. 2014; Figure 20.1). Ecologi- invasive species are introduced to novel
cal niche models predict that the potential environments through changes in trans-
of weedy European plant species to invade port patterns or novel geographic linkages
new areas will increase across the Southern (Hellmann et al. 2008). As climate change
Hemisphere but decrease in the Northern alters local weather patterns, changes to
Hemisphere across four emissions scenar- commerce and tourism routes may change
ios (Peterson et al. 2008), suggesting that both the introduction rate and the local sur-
climate change will not necessarily work vival probabilities of invasive species.
synergistically with invasive species ev- In their global assessment of marine in-
erywhere. These variable responses to cli- vaders, Molnar et al. (2008) found that in-
mate change mean that management strate- ternational shipping was the major means
gies will have to be flexible, adaptive, and of introduction of marine invaders. Sev-
largely region-specific in order to respond eral recent studies have revealed new trade
to changing invasion pressures. routes that may emerge under climate
In this chapter, we present possible con- change. For example, Smith and Stephen-
sequences of climate change for invasive son (2013) found that historic lows in Arctic
species impacts, using the stages of inva- summer sea ice will allow for increased and
sion as a guide to evaluate invasion risk. Ac- new routes across the North Pole for ves-
cording to Hellmann et al. (2008), climate sel shipping. This new route will increase
change is expected to impact the success of ship traffic, increasing the introduction
invasive species in a variety of ways, includ- rates of nonnative species to novel habitats
ing (1) altering the mechanisms of trans- through ballast water and on hulls of ships.
port and introduction, (2) altering climatic Further, Ware et al. (2016) found that cur-
constraints on invasive species, (3) altering rent ballast water management practices in
260 W H AT DOE S THE FUTUR E HOL D?
Figure 20.1. Spatial shift of suitable areas for five different temperature and environmental similar-
ant species. Darkest gray: suitable currently, but not in ity to more ports. Increasing tempera-
the future. Lighter gray: suitable currently and in the fu-
ture (to 2080). Medium gray: suitable only in the future.
tures and temperature oscillations have
(Reproduced from Bertelsmeier et al. 2014.) allowed for the range expansion of previ-
ously extant species in the Arctic (Berge
et al. 2005), suggesting that conditions
the Arctic do not prevent the introduction are becoming more favorable for warmer
of nonindigenous zooplankton and marine adapted species.
invertebrates. In Europe, nonnative spider Intentional introductions for recreation
introduction has been shown to track with or conservation purposes may also facili-
shipping routes, with the highest introduc- tate the transfer of invasive species in novel
tions coming from the areas with the high- habitats. With the increased ability to ship
est amounts of trade (Kobelt and Nentwig plants and animals in the global pet trade,
2008). there is an increased probability of intro-
Increased propagule pressure and ductions through intentional releases. Car-
warming temperatures may also make rete and Tella (2008) found a link between
these novel environments more hospitable increasing demand for exotic birds in the
for nonnatives. Ware et al. (2014) found pet trade and increased invasion risk of
that the Arctic ecosystem is at greater risk wild-caught bird species. Similarly, Strecker
for invasion due to increased sea surface et al. (2011) found that approximately 2,500
I N VA S I V E S P E C I E S A N D C L I M AT E C H A N G E 261
exotic fish are released into Puget Sound, a temperate species into higher latitudes
temperate aquatic system, annually. (Walther et al. 2009; Chen et al. 2011), thus
Invasive species proliferation may also leading to range expansions. Additionally,
be an unintended consequence of other increased extreme weather events may fa-
policies aimed at mitigating the effects of cilitate disturbance—a process that creates
climate change. Recent studies suggest that openings for invaders.
plants meeting the criteria for ideal biofuels In countries with a high Human Devel-
often also have traits that can be classified as opment Index (HDI), invasion threat was
“weedy” (Mainka and Howard 2010); with found to coincide with areas where climate-
the increased demand for biofuels, we may driven biome shifts are projected (Early et
be inadvertently introducing new invaders al. 2016). Invasive cheatgrass (Bromus tecto-
into our agricultural systems. California’s rum) growth and success has been found to
low carbon fuel standard (LCFS), part of the increase with temperature (Blumenthal et
state’s Assembly Bill 32, provides benefits al. 2016), with implications for the success
to the transportation industry through the of native semiarid grasslands as tempera-
use of biofuel blends but does not provide tures rise. Bertelsmeier et al. (2014) found
guidelines for mitigating potential inva- that 5 of 15 invasive ant populations are
sive species issues from biofuel production expected to expand their ranges up to 35.8
(Pyke et al. 2008; Chapman et al. 2016). percent. Sanders et al. (2003) have shown
Finally, climate change may allow for that the invasive Argentine ant (Linepithema
extended recreational seasons or open humile) has the potential to disassemble na-
new areas to recreational activities. These tive ant communities in northern California
changes in recreation patterns may lead to by reducing species aggregations and shift-
an increase in introduction opportunities ing the community from structured to ran-
for invasive species. Zebra mussels (Dreissena dom, consequently reducing biodiversity
polymorpha) have been introduced to novel and altering total community structure. If
freshwater environments via recreational Argentine ants or cheatgrass expand with
boaters, resulting in costly damage and climate change, their disruption of com-
mitigation programs. A recent meta-anal- munity dynamics can drive larger losses in
ysis found that abundance and richness of biodiversity.
nonnative species are higher in sites with Extreme weather events—predicted to
high tourist activity (Anderson et al. 2015), worsen under climate change scenarios—
suggesting that sites where people from di- may reduce the ability of native species to
verse areas congregate can also be impor- resist establishment of and compete with
tant introduction sites for invasive species. invaders. Through direct physiological
stress as well as indirect mechanisms, cli-
mate change has been shown to undermine
A LT E R E D C L I M ATIC CONST R A I N TS the resilience of aquatic systems to invaders.
In their meta-analysis on different systems,
Climate change can facilitate invasion in Sorte et al. (2013) found that increases in
the colonization, establishment, and spread temperature and CO2 inhibited native spe-
phases of the invasion process. In addition cies in aquatic systems. In some cases, cli-
to changes in transport pathways discussed mate change is expected to work in tandem
earlier, altered climatic constraints may with the invasion process, with positive
provide more hospitable environments for feedbacks for invasive species, which are
introduced propagules or allow currently driven by extreme weather events. This ap-
rare species to establish and colonize an pears to be the case for salt cedar (Tamarix
area (Sorte et al. 2013). Milder winter tem- spp.). The species is better able to withstand
peratures can increase survival for many drought conditions than native flora and is
262 W H AT DOE S THE FUTUR E HOL D?
the effectiveness of these policies. Keller et al. 2017). One of the most fascinating
and Springborn (2014) found that imper- things about climate change is how it calls
fect screening is still a better method than into question the very definition of native
an open-door policy to invasive species. and invasive. As species reorganize under a
Latombe et al. (2016) outlined a modular changing climate—when they can—new
approach to monitoring invasive species species will appear in new areas, and land
globally through information transfer at the managers will have to decide whether they
country level. Overall, coordinated policies are welcome arrivals.
at the international, national, and regional
level are needed to protect biodiversity
from invasive species. CONC LUSION
disturbance-driven species to thrive. This Berge, Jørgen, Geir Johnsen, Frank Nilsen, Bjørn Gul-
suggests that climate change may act as a liksen, and Dag Slagstad. 2005. “Ocean temperature
oscillations enable reappearance of blue mussels Myti-
filter to promote a weedier world. lus edulis in Svalbard after a 1000 year absence.” Marine
As trade becomes increasingly global- Ecology Progress Series 303: 167–175. https://doi.org/10
ized, better management and policies .3354/meps303167.
should be put into place to help mitigate Bertelsmeier, Cleo, Gloria M. Luque, Benjamin D. Hoff-
the impacts of invasive species and reduce mann, and Franck Courchamp. 2014. “Worldwide ant
invasions under climate change.” Biodiversity and Conser-
introduction pressure. Changes to climatic vation 24 (1): 117–128. https://doi.org/10.1007/s10531
constraints, distributions, and impacts with -014-0794-3.
warming show that much more needs to Blackburn, Tim M., Franz Essl, Thomas Evans, Philip E.
be done to mitigate and reduce the im- Hulme, Jonathan M. Jeschke, Ingolf Kühn, Sabrina
pacts of invasive species. More research Kumschick, et al. 2014. “A unified classification of
alien species based on the magnitude of their environ-
is still needed to reveal the circumstances mental impacts.” PLOS Biology 12 (5). https://doi.org/10
when climate change enhances versus re- .1371/journal.pbio.1001850.
duces the risk of invasion, but data from the Blumenthal, Dana M., Julie A. Kray, William Ortmans,
past decade give considerable credence to Lewis H. Ziska, and Elise Pendall. 2016. “Cheatgrass
the expectation that climate change could is favored by warming but not CO2 enrichment in a
semi-arid grassland.” Global Change Biology 22 (9): 3026–
increase the abundance and spread of pest 3038. https://doi.org/10.1111/gcb.13278.
species without thoughtful ecosystem man- Bradley, Bethany A. 2009. “Regional analysis of the im-
agement informed by an understanding of pacts of climate change on cheatgrass invasion shows
the invasion pathway. potential risk and opportunity.” Global Change Biology
15 (1): 196–208. https://doi.org/10.1111/j.1365-2486
.2008.01709.x.
Brooks, Matthew L., Carla M. D. Antonio, David M. Rich-
R E FE R E NCE S ardson, James B. Grace, E. Keeley, Joseph M. Ditomaso,
Richard J. Hobbs, et al. 2010. “Effects of invasive alien
Aitken, Sally N., Sam Yeaman, Jason A. Holliday, Tongli plants on fire regimes.” BioScience 54 (7): 677–688.
Wang, Sierra Curtis-McLane. 2008. “Adaptation, mi- Brower, Andrew D., J. Bradley St. Clair, and Vicky Er-
gration or extirpation: climate change outcomes for ickson. 2014. “Generalized provisional seed zones for
tree populations.” Evolutionary Applications: 95–111. native plants.” Ecological Applications 24: 913–919. https://
Anderson, Lucy G., Steve Rocliffe, Neal R. Haddaway, doi.org/10.1890/13-0285.1.
and Alison M. Dunn. 2015. “The role of tourism and Brownstein, John, Theodore Holford, and Durland Fish.
recreation in the spread of non-native species: A sys- 2005. “Effect of climate change on Lyme disease risk
tematic review and meta-analysis.” PLOS One. https:// in North America.” Ecohealth 2 (1): 38–46. https://doi
doi.org/10.1371/journal.pone.0140833. .org/10.1007/s10393-004-0139-x.Effect.
Bailie, Jonathan, Craig Hilton-Taylor, and S. N. Stuart, Burgiel, Stanley W. 2015. “From global to local: Integrat-
eds. 2004. 2004 IUCN Red List of Threatened Species: A Global ing policy frameworks for the prevention and man-
Species Assessment. International Union for Conservation agement of invasive species.” In Invasive Species in a Global-
of Nature (IUCN). https://portals.iucn.org/library/ ized World: Ecological, Social, and Legal Perspectives on Policy, ed.
node/9830. Reuben P. Keller, Marc W. Cadotte, and Glenn Sandi-
Barbet-Massin, Morgane, Quentin Rom, Claire Vil- ford, 283–302. University of Chicago Press.
lemant, and Franck Courchamp. 2018. “Can species Byrne, Maria, Mailie Gall, Kennedy Wolfe, and Antonio
distribution models really predict the expansion of Agüera. 2016. “From pole to pole: The potential for
invasive species?” PLOS One 13 (3): e0193085. the Arctic seastar Asterias amurensis to invade a warming
Battisti, Andrea, Michael Stastny, Emiliano Buffo, and Stig Southern Ocean.” Global Change Biology (March). https://
Larsson. 2006. “A rapid altitudinal range expansion in doi.org/10.1111/gcb.13304.
the pine processionary moth produced by the 2003 Carrete, Martina, and Jose L. Tella. 2008. “Wild-bird
climatic anomaly.” Global Change Biology 12: 662–671. trade and exotic invasions: A new link of conserva-
https://doi.org/10.1111/j.1365-2486.2006.01124.x. tion concern?” Frontiers in Ecology and the Environment 6 (4):
Bellard, Céline, Cleo Bertelsmeier, Paul Leadley, Wil- 207–211. https://doi.org/10.1890/070075.
fried Thuiller, and Franck Courchamp. 2012. “Im- Chapin, F. S., E. S. Zavaleta, V. T. Eviner, R. L. Naylor, P.
pacts of climate change on the future of biodiversity.” M. Vitousek, H. L. Reynolds, D. U. Hooper, et al. 2000.
Ecology Letters 15 (4): 365–377. https://doi.org/10.1111/j “Consequences of changing biodiversity.” Nature 405
.1461-0248.2011.01736.x. (6783): 234–242. https://doi.org/10.1038/35012241.
I N VA S I V E S P E C I E S A N D C L I M AT E C H A N G E 267
Chapman, Daniel S., László Makra, Roberto Alber- Conservation Biology 22 (3): 534–543. https://doi.org/10
tini, Maira Bonini, Anna Páldy, Victoria Rodinkova, .1111/j.1523-1739.2008.00951.x.
Branko Šikoparija, Elżbieta Weryszko-Chmielewska, Hewitt, N., N. Klenk, A. L. Smith, D. R. Bazely, N. Yan,
and James M. Bullock. 2016. “Modelling the introduc- S. Wood, J. I. Maclellan, C. Lipsig-mumme, and I.
tion and spread of non-native species: International Henriques. 2011. “Taking stock of the assisted migra-
trade and climate change drive ragweed invasion.” tion debate.” Biological Conservation 144 (11): 2560–2572.
Global Change Biology 22 (9): 3067–3079. https://doi.org/ https://doi.org/10.1016/j.biocon.2011.04.031.
10.1111/gcb.13220. Huang, Dingcheng, Robert A. Haack, and Runzhi Zhang.
Chen, I-ching, Jane K. Hill, Ralf Ohlemüller, David B. 2011. “Does global warming increase establishment
Roy, and Chris D. Thomas. 2011. “Rapid range shifts rates of invasive alien species? A centurial time se-
of species of climate warming.” Science 333: 1024– ries analysis.” PLOS One 6 (9). https://doi.org/10.1371/
1027. journal.pone.0024733.
Colautti, Robert I., and Spencer C. Barrett. 2013. “Rapid Jeschke, Jonathan M., and David L. Strayer. 2008. “Use-
adaptation to climate facilitates range expansion of an fulness of bioclimatic models for studying climate
invasive plant.” Science 342. https://doi.org/10.1126/ change and invasive species.” Annals of the New York
science.1242121. Academy of Sciences 1134: 1–24. https://doi.org/10.1196/
Courchamp, Franck, Alice Fournier, Céline Bellard, Cleo annals.1439.002.
Bertelsmeier, Elsa Bonnaud, Jonathan M. Jeschke, and Jore, Solveig, Sophie O. Vanwambeke, Hildegunn
James C. Russell. 2017. “Invasion biology: Specific Viljugrein, Ketil Isaksen, Anja B. Kristoffersen, Zerai
problems and possible solutions.” Trends in Ecology and Woldehiwet, Bernt Johansen, et al. 2014. “Climate and
Evolution 32 (1): 13–22. https://doi.org/10.1016/j.tree environmental change drives Ixodes ricinus geographical
.2016.11.001. expansion at the northern range margin.” Parasites and
Early, Regan, Bethany A. Bradley, Jeffrey S. Dukes, Vectors 7 (11): 1–14.
Joshua J. Lawler, Julian D. Olden, Dana M. Blumen- Keller, Reuben P., and Michael R. Springborn. 2014.
thal, Patrick Gonzalez, et al. 2016. “Global threats “Closing the screen door to new invasions.” Conserva-
from invasive alien species in the twenty-first century tion Letters 7 (3): 285–292. https://doi.org/10.1111/conl
and national response capacities.” Nature Communications .12071.
7: 12485. https://doi.org/10.1038/ncomms12485. Knops, M., D. Tilman, N. M. Haddad, S. Naeem, C. E.
Evans, Thomas, Sabrina Kumschick, and Tim M. Black- Mitchell, J. Haarstad, M. E. Ritchie, et al. 1999. “Ef-
burn. 2016. “Application of the Environmental Impact fects of plant species richness on invasion dynamics,
Classification for Alien Taxa (EICAT) to a global as- disease outbreaks, insect abundances and diversity.”
sessment of alien bird impacts.” Diversity and Distributions Ecology Letters 2: 286–293. https://doi.org/10.1046/j
22 (9). https://doi.org/10.1111/ddi.12464. .1461-0248.1999.00083.x.
Funk, Jennifer L., Elsa E. Cleland, Katherine N. Suding, Kobelt, Manuel, and Wolfgang Nentwig. 2008. “Alien
and Erika S. Zavaleta. 2008. “Restoration through re- spider introductions to Europe Supported by global
assembly: Plant traits and invasion resistance.” Trends in trade.” Diversity and Distributions 14 (2): 273–280. https://
Ecology and Evolution 23 (12): 695–703. https://doi.org/10 doi.org/10.1111/j.
.1016/j.tree.2008.07.013. Kumschick, Sabrina, G. John Measey, Giovanni Vimer-
Grosholz, E. D., L. A. Levin, A. C. Tyler, and C. Neira. cati, F., Andre de Villiers, Mohlamatsane M. Mokhatla,
2009. “Changes in community structure and ecosys- Sarah J. Davies, Corey J. Thorp, Alexander D. Rebelo,
tem function following Spartina alterniflora invasion of Tim M. Blackburn, and Fred Kraus. 2017. “How re-
Pacific estuaries.” In Human Impacts on Salt Marshes: A Global peatable is the Environmental Impact Classification of
Perspective, ed. B. R. Silliman, M. D. Bertness, and E. D. Alien Taxa (EICAT)? Comparing independent global
Grosholz, 23–40. University of California Press. impact assessments of amphibians.” Ecology and Evolution
Havens, Kayri, Pati Vitt, Shannon Still, Andrea T. Kramer, 7 (8): 2661–2670. https://doi.org/10.1002/ece3.2877.
Jeremie B. Fant, and Katherine Schatz. 2015. “Seed Latombe, Guillaume, Petr Pyšek, Jonathan M. Jeschke,
sourcing for restoration in an era of climate change.” Tim M. Blackburn, Sven Bacher, César Capinha, Mark
Natural Areas Journal 35 (1): 122–133. J. Costello, et al. 2016. “A vision for global monitoring
Hawkins, Charlotte L., Sven Bacher, Franz Essl, Philip of biological invasions.” Biological Conservation (June).
E. Hulme, Jonathan M. Jeschke, Ingolf Kühn, Sabrina https://doi.org/10.1016/j.biocon.2016.06.013.
Kumschick, et al. 2015. “Framework and guidelines Levine, Jonathan M., and Carla M. D’Antonio. 1999. “El-
for implementing the proposed IUCN Environmental ton revisited: A review of evidence linking diversity
Impact Classification for Alien Taxa (EICAT).” Diversity and invasibility.” Oikos 87 (1): 15–26.
and Distributions 21 (11): 1360–1363. https://doi.org/10 Li, Xianping, Xuan Liu, Fred Kraus, Reid Tingley, and
.1111/ddi.12379. Yiming Li. 2016. “Risk of biological invasions is con-
Hellmann, Jessica J., James E. Byers, Britta G. Bierwa- centrated in biodiversity hotspots.” Frontiers in Ecology
gen, and Jeffrey S. Dukes. 2008. “Five potential con- and the Environment 14 (8): 411–417. https://doi.org/10
sequences of climate change for invasive species.” .1002/fee.1321.
268 W H AT DOE S THE FUTUR E HOL D?
Lu, Xinmin, Evan Siemann, Minyan He, Hui Wei, Xu Peterson, A. Townsend, Aimee Stewart, Kamal I. Mo-
Shao, and Jianqing Ding. 2015. “Climate warming in- hamed, and Miguel B. Araújo. 2008. “Shifting global
creases biological control agent impact on a non-tar- invasive potential of European plants with climate
get species.” Ecology Letters 18: 48–56. https://doi.org/ change.” PLOS One 3 (6): 1–7. https://doi.org/10.1371/
10.1111/ele.12391. journal.pone.0002441.
Lunt, Ian D., Margaret Byrne, Jessica J. Hellmann, Nicola Pyke, Christopher R., Roxanne Thomas, Read D. Porter,
J. Mitchell, Stephen T. Garnett, Matt W. Hayward, Jessica J. Hellmann, Jeffrey S. Dukes, David M. Lodge,
Tara G. Martin, Eve McDonald-Maddden, Stephen E. and Gabriela Chavarria. 2008. “Current practices and
Williams, and Kertin K. Zander. 2013. “Using assisted future opportunities for policy on climate change and
colonization to conserve biodiversity and restore eco- invasive species.” Conservation Biology 22 (3): 585–592.
system function under climate change.” Biological Con- https://doi.org/10.1111/j.1523-1739.2008.00956.x.
servation 157: 172–177. Richardson, D. M., J. J. Hellmann, J. S. McLachlan, D. F.
Mack, Michelle C., and Carla M. D’Antonio. 1998. “Im- Sax, M. W. Schwartz, P. Gonzalez, E. J. Brennan, A.
pacts of biological invasions on disturbance regimes.” Camacho, T. L. Root, O. E. Sala, S. H. Schneider, D. M.
Trends in Ecology and Evolution 13 (5): 195–198. https://doi Ashe, J. R. Clark, R. Early, J. R. Etterson, E. D. Fielder,
.org/10.1016/S0169-5347(97)01286-X. J. L. Gill, B. A. Minteer, S. Polasky, H. D. Safford, A. R.
Mainka, Susan A., and Geoffrey W. Howard. 2010. “Cli- Thompson, and M. Vellend. 2009. “Multidimensional
mate change and invasive species: Double jeopardy.” evaluation of managed relocation.” Proceedings of the Na-
Integrative Zoology 5 (2): 102–111. https://doi.org/10 tional Academy of Sciences 106: 9721–9724.
.1111/j.1749-4877.2010.00193.x. Sala, Osvaldo E., F. Stuart Chapin III, Juan J. Armesto,
Maino, James L. Paul A. Umina, Ary A. Hoffmann, and Eric Berlow, Janine Bloomfield, Rodolfo Dirzo, Elisa-
Arndt Hampe. 2017. “Climate contributes to the evo- beth Huber-Sanwald, et al. 2000. “Global biodiversity
lution of pesticide resistance.” Global Ecology and Biogeog- scenarios for the year 2100.” Science 287 (5459): 1770–
raphy 27: 223–232. 1775. https://doi.org/10.1126/science.287.5459.1770.
Millenium Ecosystem Assessment. 2005. “Ecosystems Sanders, Nathan J., Nicholas J. Gotelli, Nicole E. Heller,
and human well-being: Synthesis.” World Resources and Deborah M. Gordon. 2003. “Community disas-
Institute. sembly by an invasive species.” Proceedings of the National
Molnar, Jennifer L., Rebecca L. Gamboa, Carmen Re- Academy of Sciences 100 (5): 2474–2477. https://doi.org/
venga, and Mark D. Spalding. 2008. “Assessing the 10.1073/pnas.0437913100.
global threat of invasive species to marine biodiver- Sarma, Roshmi Rekha, Madhushree Munsi, and Ara-
sity.” Frontiers in Ecology and the Environment 6 (9): 485–492. vind Neelavara Ananthram. 2015. “Effect of climate
https://doi.org/10.1890/070064. change on invasion risk of giant African snail (Acha-
Mueller, J. M., and J. J. Hellmann. 2008. “An assessment tina fulica ferussac, 1821: Achatinidae) in India.” PLOS One
of invasion risk from assisted migration.” Conservation 10 (11): 1–16. https://doi.org/10.1371/journal.pone
Biology 22: 562–567. .0143724.
Naeem, Shahid, Johannes M. H. Knops, David Tilman, Selvaraj, S., P. Ganeshamoorthi, and T. Pandiaraj. 2013.
Katherine M. Howe, Theodore Kennedy, and Samuel “Potential impacts of recent climate change on bio-
Gale. 2000. “Plant diversity increases resistance to in- logical control agents in agro-ecosystem: A review.”
vasion in the absence of covarying extrinsic factors.” Journal of Biodiversity and Conservation 5 (12): 845–852.
Oikos 91 (1): 97–108. https://doi.org/10.1034/j.1600 https://doi.org/10.5897/IJBC2013.0551.
-0706.2000.910108.x. Smith, Laurence C., and Scott R. Stephenson. 2013.
Ogden, Nicholas H., and Milka Radojevic. 2014. “Esti- “New trans-Arctic shipping routes navigable by
mated effects of projected climate change on the ba- midcentury.” Proceedings of the National Academy of Sciences
sic reproductive number of the Lyme disease vector 110 (13): E1191–E1195. https://doi.org/10.1073/pnas
Ixodes scapularis.” Environmental Health Perspectives 122 (6): .1214212110.
631–638. Sorte, Cascade J. B., Ines Ibáñez, Dana M. Blumenthal,
Packer, Jasmin G., Laura A. Meyerson, David M. Richard- Nicole A. Molinari, Luke P. Miller, Edwin D. Gro-
son, Giuseppe Brundu, Warwick J. Allen, Ganesh P. sholz, Jeffrey M. Diez, et al. 2013. “Poised to prosper?
Bhattarai, Hans Brix, et al. 2017. “Global networks for A cross-system comparison of climate change effects
invasion science: Benefits, challenges and guidelines.” on native and non-native species performance.” Ecol-
Biological Invasions 19 (4): 1081–1096. https://doi.org/10 ogy Letters 16 (2): 261–270. https://doi.org/10.1111/ele
.1007/s10530-016-1302-3. .12017.
Parker, I. M., D. Simberloff, W. M. Lonsdale, K. Goodell, Strecker, Angela L., Philip M. Campbell, Julian D. Olden,
M. Wonham, P. M. Kareiva, M. H. Williamson, B. Angela L. Strecker, Philip M. Campbell, Julian D.
von Holle, P. B. Moyle, J. E. Byers, and L. Goldwas- Olden, and Philip M. Campbell. 2011. “The aquarium
ser. 1999. “Toward a framework for understanding trade as an invasion pathway in the Pacific North-
the ecological effects of invaders.” Biological Invasions 1: west.” Fisheries 36 (2): 74–85. https://doi.org/10.1577/
3–19. https://doi.org/10.1023/A:1010034312781. 03632415.2011.10389070.
I N VA S I V E S P E C I E S A N D C L I M AT E C H A N G E 269
Thuiller, Wilfried, D. M. Richardson, and G. F. Midg- Evolution 24 (12). https://doi.org/10.1016/j.tree.2009
ley. 2007. “Will climate change promote alien plant .06.008.
invasions?” In Biological Invasions, ed. W. Nentwig, 193, Ware, Chris, Jørgen Berge, Anders Jelmert, Steffen M.
197–211. Springer Verlag. http://www.springerlink Olsen, Loïc Pellissier, Mary Wisz, Darren Kriticos,
.com/index/G66R623223012856.pdf. Georgy Semenov, Sławomir Kwaśniewski, and Inger G.
Veken, Sebastiaan Van der, Martin Hermy, Mark Vellend, Alsos. 2016. “Biological introduction risks from ship-
Anne Knapen, Martin Hermy, Mark Veliend, Anne ping in a warming Arctic.” Journal of Applied Ecology 53:
Knapenl, and Kris Verheyen. 2008. “Garden plants get 340–349. https://doi.org/10.1111/1365-2664.12566.
a head start on climate change.” Frontiers in Ecology and Ware, Chris, Jørgen Berge, Jan H. Sundet, Jamie B. Kirk-
the Environment 6 (4): 212–216. https://doi.org/10.1890/ patrick, Ashley D. M. Coutts, Anders Jelmert, Steffen
070063. M. Olsen, Oliver Floerl, Mary S. Wisz, and Inger G.
Vitt, Pati, Pairsa N. Belmaric, Rily Book, and Melissa Alsos. 2014. “Climate change, non-indigenous species
Curran. 2016. “Assisted migration as a climate change and shipping: Assessing the risk of species introduc-
adaptation strategy: Lessons from restoration and tion to a high-Arctic archipelago.” Diversity and Distribu-
plant reintroductions.” Israel Journal of Plant Sciences 63: tions 20: 10–19. https://doi.org/10.1111/ddi.12117.
250–261. Willis, Charles G., Brad R. Ruhfel, Richard B. Primack,
Walther, Gian-reto, Alain Roques, Philip E. Hulme, Abraham J. Miller-Rushing, Jonathan B. Loso, and
Martin T. Sykes, Petr Pys, Christelle Robinet, and Vi- Charles C. Davis. 2010. “Favorable climate change
taliy Semenchenko. 2009. “Alien species in a warmer response explains non-native species’ success in Tho-
world: Risks and opportunities.” Trends in Ecology and reau’s woods.” PLOS One 5 (1): e8878.
CHAPTER TWENTY-ONE Rumination about biological implications
began soon after the realization that green-
Climate Change house gases were increasing in the atmo-
sphere; a prominent element was that disease
and Disease transmission would increase (e.g., Epstein et
al. 1998). A 1995 Intergovernmental Panel
LINDSAY P. CAMPBELL, A. TOWNSEND on Climate Change (IPCC) report (Watson
PETERSON, ABDALLAH M. SAMY, AND et al. 1996) speculated that the potential
CARLOS YAÑEZ-ARENAS geographic distribution of malaria could
increase “from approximately 45 percent of
the world population to 60 percent by the
latter half of the next century,” although
subsequent IPCC reports became more mea-
sured in their disease projections and em-
phasized human socioeconomic dimensions
more clearly (McCarthy et al. 2001).
Considering disease transmission as the
product of interactions among suites of
species and the environmental landscapes
where they are distributed (Peterson 2014),
basic implications of warming climate for
disease transmission are clear: ranges will
generally shift to higher latitudes and el-
evations, and transmission phenomena will
likely occur earlier in spring and later in
fall. There are different geographic and en-
vironmental dimensions of climate change
effects on species. Many of these expecta-
tions are already being fulfilled (see other
contributions in this volume). Still, evidence
needed to pinpoint climate change causally
is difficult to assemble, such that climate
change effects may often go unappreciated.
Here, we review models that have been
used to explore climate change effects on
disease. These distinct approaches offer
some possibility of anticipation of poten-
tial changes in disease transmission across
landscapes. We explore a case study—ma-
laria in highland East Africa—to illustrate
complexities inherent in the question.
270
CLIM ATE CH A NGE A ND DISE ASE 271
T R A NSM IS SION MODE LS cally. Martens et al. (1999) analyzed effects
of temperature and precipitation on biting
Disease transmission models were devel- rates and adult emergence in Anopheles un-
oped to understand and predict disease der scenarios derived from HadCM2 and
dynamics. Transmission models are math- HadCM3 climate models and concluded
ematical models that mimic circulation of that a 1ºC increase could increase malaria
pathogens at the population level (Keeling transmission in low-temperature regions.
and Rohani 2007). SIR models, which parti- Ogden et al. (2014) evaluated climate
tion populations into classes of susceptible change effects on R0 for Lyme disease and
(S), infected (I), and recovered (R), have projected increases of 150 percent–500 per-
seen considerable development (Hethcote cent in North America.
2000). These models are used to predict and Other transmission models have focused
explain changes in disease transmission. on climate change effects on vectorial capac-
Ronald Ross and George MacDonald ity, the integration of feeding rate, survival,
were pioneers in this field. Ross (1904) de- and incubation period. Kearney et al. (2009)
veloped a first mathematical model to de- evaluated how climate change may expand
scribe adult mosquito movements and the distributional potential of the mosquito Aedes
spatial scale of larval control that would best aegypti in Australia (Figure 21.1); Liu-Helm-
reduce mosquito populations, thus contrib- ersson et al. (2014) examined effects of cli-
uting to elimination of disease transmis- mate change on Aedes aegypti vectorial capac-
sion. Ross’s first model did not address the ity in dengue transmission. In both, mean
question of malaria transmission directly, temperature and diurnal temperature range
but it certainly earned him a place in his- were key factors. Another example com-
tory. Later, he developed malaria models pared development of Plasmodium falciparum at
incorporating differential equations. In forested and deforested sites (Afrane et al.
1950–1956, MacDonald used field-collected 2008): increased temperature and solar ra-
data to test Ross’s models on infant malaria diation at deforested sites elevates infection
rates and sporozoite prevalences; finally, in rate and vectorial capacity of the mosquito
1957, he published a book Epidemiology and Anopheles gambiae, and Plasmodium sporozoites
Control of Malaria (MacDonald 1957), sum- developed more rapidly than at forested sites,
marizing factors limiting malaria transmis- suggesting combined effects of deforestation
sion. MacDonald proposed expression of and climate change on malaria dynamics.
the basic reproductive rate (R0) for malaria Transmission models are thus power-
as the mean number of secondary infec- ful in permitting “experimentation” with
tions produced when an infected individual specific effects of particular parameters,
is introduced into a host population, as- though not without limitations. Keeling and
suming that every other individual in the Rohani (2007) noted frequent information
population is susceptible; Keeling and Ro- gaps for specific model parameter values:
hani (2007) estimated R0 for several diseases full model implementation requires esti-
and found values ranging from ~5 (polio) mates of all parameters, but population- or
to ~100 (malaria). species-specific estimates are frequently un-
Such models incorporated environmen- available or imprecise. Even when detailed
tal drivers in malaria transmission. Lindsay estimates are available, interactions among
and Birley (1996) used a simple mathemati- parameters may be poorly understood; as-
cal framework to study temperature ef- sumptions of independence can bias results
fects on vectorial capacity of Anopheles macu- substantially. Models do not necessarily
lipennis in transmitting Plasmodium vivax; they reconstruct reality; rather, they should be
found that small changes in temperature used as heuristics that instruct about the
can change malaria transmission dramati- behavior of complex systems.
272 W H AT DOE S THE FUTUR E HOL D?
Figure 21.1. Predictions of number of consecutive life- niche estimates onto future climate scenar-
cycle completions per year that could emerge from two
ios allows for estimating future potential
simulated breeding containers (rainwater tanks, buck-
ets) under different levels of shade. The model assumes distributions of disease-related phenomena.
that urbanized areas with containers that could sustain Early examples applying ecological niche
Aedes aegypti are present. (A) The historical distribution is models (ENMs) to questions of climate
imposed, (B) the present distribution is imposed, and change and disease include Rogers and Ran-
(C) and (D) the locations of recent (since 1990) dengue
dolph (2000), who assessed climate effects
outbreaks are imposed. (From Kearney et al. 2009.)
on human malaria cases and future expan-
COR R E L ATIV E MODE LS sions in some areas being offset by contrac-
tions elsewhere (Figure 21.2). Peterson and
Another approach to understanding climate Shaw (2003) focused on climate change
change effects on disease transmission has effects on distributions of Lutzomyia sandfly
been correlative ecological niche modeling species (vectors of cutaneous leishmaniasis)
(Peterson et al. 2011); these approaches are in Brazil and anticipated southward ex-
also referred to as species distribution mod- pansion of distributions. Many subsequent
els. The approach relates occurrences of a studies have used ENMs to assess potential
species or disease to raster data sets summa- effects of climate change on disease geogra-
rizing environmental variables to identify phy, and many have found significant dis-
environments under which transmission tributional implications.
occurs (or under which a species can main- Recent studies have refined techniques in
tain populations). These putatively suitable model calibration to deal with limited dis-
environments (i.e., ecological niches) can persal and uneven sampling, which affect
be used to identify regions where the spe- model results (Peterson et al. 2011). Costa
cies can maintain populations. Transferring et al. (2014) included explicit consideration
Figure 21.2. Early example of application of correlative between predicted distributions in (A) and (B), showing
ecological niche modeling to questions of disease trans- areas where malaria is predicted to disappear (i.e., prob-
mission risk and climate change. (A) Current global map ability of occurrence decreases from >0.5 to <0.5) (in
of malaria caused by Plasmodium falciparum (hatching) and medium gray) or invade (i.e., probability of occurrence
predicted distribution (light gray areas indicate no pre- increases from <0.5 to >0.5) (in dark gray) by the 2050s
diction). (B) Discriminating criteria from the current sit- in relation to the present situation. The hatching is the
uation were applied to equivalent climate surfaces from current global malaria map shown in hatching in (A).
the high scenario from HadCM2: hatching and the prob- (From Rogers and Randolph 2000.)
ability scale are the same as in (A). (C) The difference
274 W H AT DOE S THE FUTUR E HOL D?
detail to illustrate these complexities and to that significant changes in mean precipita-
emphasize the interconnected, interdepen- tion or temperature had not occurred in
dent nature of the panoply of factors influ- the region and thus could not be a causal
encing disease transmission. factor in increasing case rates. Shanks et al.
Highland malaria regions are areas near (2002) also argued that climate change was
the maximum elevational limit of malaria not a significant factor in increased malaria
transmission, where transmission is un- incidence in the 1990s because significant
stable, resulting in epidemic or episodic changes in mean ambient temperature and
outbreaks. Endemic malaria is not gener- mean precipitation were not found across
ally present above 1,500 m in East Africa, 1965–1995.
although cases have been recorded above Alternatively, Githeko (2001) investi-
2,000 m; a minimum temperature thresh- gated effects of temperature and precipi-
old of <18°C exists at which Plasmodium falci- tation anomalies on incubation periods of
parum transmission is not stable, which was Plasmodium falciparum in Anopheles gambiae s.s.,
the mean temperature at ~2,000 m in East and A. funestus, creating an epidemic pre-
Africa in 2006 (Githeko et al. 2006). Peri- diction model based on their results. Posi-
odically, human and environmental con- tive temperature anomalies from mean
ditions become suitable for malaria trans- monthly climate trends were, they found, a
mission in these regions, often following precursor to malaria epidemics when mean
anomalous weather events. monthly precipitation was above 150 mm;
Malaria transmission in the western the authors suggested that monthly temper-
highlands of Kenya can be traced to comple- ature data are too general to detect impor-
tion of a railway system in 1901 that moved tant malaria risk signals.
infected mosquitoes and laborers into the As studies became more sophisticated,
region; epidemics began around 1918, last- the debate grew. A series of analyses of ma-
ing until widespread mosquito control and laria incidence at four high-elevation sites
antimalarial programs were implemented in Kenya, Uganda, Rwanda, and Burundi
in the 1950s, particularly as regards the highlight differences in inferences that can
dominant vector species, Anopheles gambiae. In be obtained from a single climate data set.
the late 1980s, however, epidemic malaria Hay et al. (2002) investigated meteorologi-
reappeared in this region, with outbreaks cal trends, with results indicating no signif-
occurring nearly every year. Understand- icant changes in mean temperature or va-
ing how disturbances of different duration por pressure during 1911–1995 at any of the
could affect transmission equilibrium be- study locations, despite warmer and wetter
gan to define scientific research in this area: seasonal trends. A separate analysis of the
the following is a summary of the ongoing same data set (Patz et al. 2002) found that
dialogue and debate. warming trends did, in fact, correspond to
A major factor in the debate is differences increased malaria incidence at specific sites,
in how climate change was characterized in noting methods used to downscale the cli-
analyses. Although several studies analyzed mate data by the original authors as a likely
changes in mean values, others focused on driver of the incongruity in results. Addi-
variability and emphasized that climate tionally, malaria incidence corresponded
change is not restricted to simple, general more closely to climate anomalies, so the
warming trends (Patz et al. 2002). For ex- authors emphasized that mean warming
ample, Malakooti, Biomndo, and Shanks trends were not sufficient as indicators
(1998) analyzed temporal relationships be- of climate change, echoing the results of
tween malaria epidemics and temperature Githeko (2001).
and precipitation in 1990–1997; results sug- A third analysis added 5 years of daily
gested that disease incidence fluctuated but climate data to the data set (Pascual et al.
CLIM ATE CH A NGE A ND DISE ASE 277
2006). The authors used a nonparametric found statistically significant signals relat-
singular spectrum analysis and a paramet- ing climate factors and their interactions to
ric seasonal autoregressive moving average recent increases in malaria incidence in the
(SARMA) model to examine data for signif- region. These findings do not negate effects
icant warming trends. A significant warm- of population movements, drug resistance,
ing trend of 0.5°C was identified, contra- land-use change, population immunity, or
dicting Hay et al. (2002). The warming other factors, but changing climate appears
trend began in 1980, coinciding with the to make a significant contribution.
rise in malaria incidence at these locations;
although other factors (e.g., drug resistance,
land-cover change, their interactions) CONC LUSIONS
could not be excluded as important factors,
changing climate could not be excluded as Links between climate change and shifts
a contributing factor. in the distribution, timing, and intensity of
Zhou et al. (2004) analyzed climate vari- disease transmission are complex. The early
able interactions and their effects on malaria IPCC reports painted a dark picture in which
incidence at highland locations in Ethiopia, climate change would somehow “activate”
Kenya, and Uganda. Whereas only two of diseases and create a wave of new problems.
seven sites exhibited statistically significant Twenty or so years later, the picture is more
increases in mean temperature (1989–1998 nuanced: whereas some diseases are indeed
vs. 1978–1988), variance in maximum beginning to shift distributionally, those
monthly temperature increased signifi- shifts are invariably subtle and multifactorial.
cantly at five of seven sites, and all seven What is more, increases and expansions in
sites showed significant positive effects of transmission may be balanced in other re-
interactions on malaria transmission. gions, where transmission may be reduced.
Results from more recent analyses con- In sum, then, climate change will indeed
tinue to support significant climate change change disease transmission patterns, region-
impacts on malaria incidence in the region. ally and for particular disease systems, but it
Alonso, Bouma, and Pascual (2011) created may not augment or reduce the overall mag-
a coupled human-malaria model that incor- nitude of disease transmission worldwide.
porated temperature and precipitation time Two points are worth highlighting. First,
series data, again using malaria data from as diseases are the product of interactions
the Kericho tea estate, and identified a non- among biological species (e.g., pathogens,
linear positive relationship between mean vectors, hosts), their transmission areas
warmer temperatures and more cases from and intensities depend on the ecological
the 1970s to the 1990s. Pascual et al. (2006) niches of those species and their interac-
implemented a stage-structured biological tions, which will make for considerable
model for Anopheles gambiae, incorporating complexity in responses to climate change.
climate data from the 1980s to the pres- Second, is more of a comment on human
ent and found amplification of mosquito perspectives: early commentaries on effects
population dynamics of a magnitude much of climate change on disease were written
greater than changes in the environmental from a very “developed world” or “north-
variables, such that subtle climatic changes ern” perspective, such that transformation
can have major biological implications. of temperate zone climates to subtropical
Clearly, the relationship between malaria climates at the southern borders of those
and climate change in the highland East Af- regions and consequent arrival of “tropi-
rica is complex, with multiple human, bio- cal” diseases might seem a dominant ex-
logical, and environmental factors affecting pectation of warming climates. The reality,
transmission. However, several studies have of course, is that people the world over are
278 W H AT DOE S THE FUTUR E HOL D?
experiencing these changes, so a more var- tion: CONACyT for support of CYR, the
ied suite of responses will be experienced: C-CHANGE IGERT program for support of
in some areas, disease transmission will in- LPC, the Egyptian Fulbright Mission Pro-
crease, whereas in others it may decrease. gram (EFMP) for support of AMS, and a
As can be appreciated from the discus- grant from the Inter-American Institute to
sions above, however, the picture of climate Tulane University, which supported (par-
change effects on disease transmission pat- tially) ATP’s work.
terns remains far from clear, and causal
chains have not been documented defini-
tively in any disease system. In terms of
R E FE R E NCE S
tools with which to identify and anticipate
these changes, mechanistic and correlative Afrane, Y. A., T. J. Little, B. W. Lawson, A. K. Githeko, and
models offer complementary perspectives. G. Y. Yan. 2008. “Deforestation and vectorial capacity
The few direct comparisons between the of Anopheles gambiae mosquitoes in malaria transmis-
two approaches indicate that they converge sion, Kenya.” Emerging Infectious Diseases 14: 1533–1538.
https://doi.org/10.3201/eid1410.070781.
on similar projections, but clearly much
Alonso, D., M. J. Bouma, and M. Pascual. 2011. “Epi-
more experimentation, exploration, and demic malaria and warmer temperatures in recent
comparison is needed to understand the decades in an East African highland.” Proceedings of the
two approaches in greater detail. Royal Society B 278: 1661–1669.
The data on which conclusions about cli- Anyamba, A., J.-P. Chretien, J. Small, C. J. Tucker, P. B.
Formenty, J. H. Richardson, S. C. Britch, D. C. Schna-
mate change and disease are based are cru-
bel, R. L. Erickson, and K. J. Linthicum. 2009. “Pre-
cial, and yet have not been curated or con- diction of a Rift Valley fever outbreak.” Proceedings of the
served in any detail (Peterson 2008). That National Academy of Sciences 106: 955–959.
is, baseline data to which present and future Bebber, D. P., Á. Delgado Castillo, and S. J. Gurr. 2016.
transmission patterns can be compared to “Modelling coffee leaf rust risk in Colombia with cli-
mate reanalysis data.” Philosophical Transactions of the Royal
document changes are sorely and woefully
Society B 371: 20150458.
lacking. As a consequence, a crucial priority Campbell, L. P., C. Luther, D. Moo-Llanes, J. M. Ramsey,
is organization, integration, and documen- R. Danis-Lozano, and A. T. Peterson. 2015. “Climate
tation of existing primary data regarding change influences on global distributions of dengue
disease geography (i.e., data that place oc- and chikungunya virus vectors.” Philosophical Transactions
of the Royal Society B 370: 20140135.
currences of pathogens, vectors, and hosts
Cash, B. A., X. Rodo, M. Emch, M. Yunus, A. S. Faruque,
in particular places at particular times), and and M. Pascual. 2014. “Cholera and shigellosis: Dif-
making those data openly available to the ferent epidemiology but similar responses to climate
broader scientific and public health com- variability.” PLOS One 9: e107223.
munities. Such retrospective data curation Costa, J., L. L. Dornak, C. E. Almeida, and A. T. Peterson.
2014. “Distributional potential of the Triatoma brasilien-
can be paired with efficient monitoring sys-
sis species complex at present and under scenarios of
tems put in place explicitly to detect future future climate conditions.” Parasites and Vectors 7: 238.
changes: permanent monitoring networks Descamps, S., J. Aars, E. Fuglei, K. M. Kovacs, C. Lyder-
can characterize transmission patterns over sen, O. Pavlova, Å. Ø. Pedersen, V. Ravolainen, and H.
the long term, being sampled at regular in- Strøm. 2016. “Climate change impacts on wildlife in
a High Arctic archipelago—Svalbard, Norway.” Global
tervals (e.g., years, decades, centuries), such
Change Biology 23: 490–502.
that ideal data exist with which to under- Epstein, P. R., H. F. Diaz, S. Elias, G. Grabherr, N. E. Gra-
stand these complex processes. ham, W. J. M. Martens, E. Mosley-Thompson, and J.
Susskind. 1998. “Biological and physical signs of cli-
mate change: Focus on mosquito-borne diseases.” Bul-
letin of the American Meteorological Society 79 (3): 409–417.
ACK NOWLE DGME NTS
https://doi.org/10.1175/1520-0477(1998)079<0409:
Bapsoc>2.0.Co;2.
We thank various agencies for support Garamszegi, L. Z. 2011. “Climate change increases ma-
during the development of this contribu- laria risk in birds.” Global Change Biology 17: 1751–1759.
CLIM ATE CH A NGE A ND DISE ASE 279
Githeko, A. K. 2001. “Predicting malaria epidemics Malakooti, M. A., K. Biomndo, and G. D. Shanks. 1998.
in the Kenyan highlands using climate data: A tool “Reemergence of epidemic malaria in the highlands
for decision makers.” Global Change and Human Health 2: of western Kenya.” Emerging Infectious Diseases 4: 671–676.
54–63. Martens, P., R. S. Kovats, S. Nijhof, P. de Vries, M. T. J.
Githeko, A. K., J. M. Ayisi, P. K. Odada, F. K. Atieli, B. A. Livermore, D. J. Bradley, J. Cox, and A. J. McMichael.
Ndenga, J. I. Githure, and G. Yan. 2006. “Topography 1999. “Climate change and future populations at risk
and malaria transmission heterogeneity in western of malaria.” Global Environmental Change 9: S89–S107.
Kenya highlands: Prospects for focal vector control.” McCarthy, J. J., O. F. Canziani, N. A. Leary, D. J. Dokken,
Malaria Journal 5: 107. and K. S. White, eds. 2001. Climate Change 2001: Impacts,
Guillaumet, A., W. A. Kuntz, M. D. Samuel, and E. H. Adaptation, and Vulnerability; Contribution of Working Group II
Paxton. 2017. “Altitudinal migration and the future to the Third Assessment Report of the Intergovernmental Panel on
of an iconic Hawaiian honeycreeper in response to Climate Change. Cambridge University Press.
climate change and management.” Ecological Monographs. Nakazawa, Y., R. Williams, A. T. Peterson, P. Mead, E.
https://doi.org/10.1002/ecm.1253. Staples, and K. L. Gage. 2007. “Climate change effects
Harvell, C. D., C. E. Mitchell, J. R. Ward, S. Altizer, A. P. on plague and tularemia in the United States.” Vector
Dobson, R. S. Ostfeld, and M. D. Samuel. 2002. “Cli- Borne and Zoonotic Diseases 7: 529–540.
mate warming and disease risks for terrestrial and Ogden, N. H., M. Radojevic, X. Wu, V. R. Duvvuri, P.
marine biota.” Science 296: 2158–2162. A. Leighton, and J. P. Wu. 2014. “Estimated effects of
Hay, S. I., J. Cox, D. J. Rogers, S. E. Randolph, D. I. Stern, projected climate change on the basic reproductive
G. D. Shanks, M. F. Myers, and R. W. Snow. 2002. “Cli- number of the Lyme disease vector Ixodes scapularis.” En-
mate change and the resurgence of malaria in the East vironmental Health Perspectives 122: 631–638.
African highlands.” Nature 415: 905–909. Pascual, M., J. A. Ahumada, L. F. Chaves, X. Rodo, and M.
Hethcote, H. W. 2000. “The mathematics of infectious Bouma. 2006. “Malaria resurgence in the East African
diseases.” SIAM Review 42: 599–653. highlands: Temperature trends revisited.” Proceedings of
Junk, J., L. Kouadio, P. Delfosse, and M. El Jarroudi. 2016. the National Academy of Sciences 103: 5829–5834.
“Effects of regional climate change on brown rust dis- Patz, J. A., M. Hulme, C. Rosenzweig, T. D. Mitchell, R. A.
ease in winter wheat.” Climatic Change 135: 439–451. Goldberg, A. K. Githeko, S. Lele, A. J. McMichael, and
Kearney, M., W. P. Porter, C. Williams, S. Ritchie, and A. D. Le Sueur. 2002. “Climate change: Regional warm-
A. Hoffmann. 2009. “Integrating biophysical models ing and malaria resurgence.” Nature 420: 627–628.
and evolutionary theory to predict climatic impacts Peterson, A. T., and J. J. Shaw. 2003. “Lutzomyia vectors for
on species’ ranges: The dengue mosquito Aedes aegypti cutaneous leishmaniasis in southern Brazil: Ecological
in Australia.” Functional Ecology 23: 528–538. https://doi niche models, predicted geographic distributions, and
.org/10.1111/j.1365-2435.2008.01538.x. climate change effects.” International Journal of Parasitology
Keeling, M. J., and P. Rohani. 2007. Modeling Infectious Dis- 33: 919–931.
eases in Humans and Animals. Princeton University Press. Peterson, A. T. 2008. “Improving methods for reporting
Kim, K.-H., and J. Cho. 2016. “Predicting potential epi- spatial epidemiologic data.” Emerging Infectious Diseases
demics of rice diseases in Korea using multi-model 14: 1335–1336.
ensembles for assessment of climate change impacts Peterson, A. T. 2014. Mapping Disease Transmission Risk in Geo-
with uncertainty information.” Climatic Change 134: graphic and Ecological Contexts. Johns Hopkins University
327–339. Press.
LaPointe, D. A., C. T. Atkinson, and M. D. Samuel. “Ecol- Peterson, A. T., J. Soberón, R. G. Pearson, R. P. Anderson,
ogy and conservation biology of avian malaria.” Annals E. Martínez-Meyer, M. Nakamura, and M. B. Araújo.
of the New York Academy of Sciences 1249: 211–226. 2011. Ecological Niches and Geographic Distributions. Princeton
Lindsay, S. W., and M. H. Birley. 1996. “Climate change University Press.
and malaria transmission.” Annals of Tropical Medicine and Rogers, D. J., and S. E. Randolph. 2000. “The global
Parasitology 90: 573–588. spread of malaria in a future, warmer world.” Science
Lipp, E. K., A. Huq, and R. R. Colwell. 2002. “Effects 289: 1763–1766.
of global climate on infectious disease: The cholera Ross, R. 1904. “The anti-malaria experiment at Mian-
model.” Clinical Microbiology Reviews 15: 757–770. Mir.” British Medical Journal 2: 632–635.
Lips, K. R., J. Diffendorfer, J. R Mendelson III, M. W. Roy-Dufresne, E., T. Logan, J. A. Simon, G. L. Chmura,
Sears. 2008. “Riding the wave: Reconciling the roles and V. Millien. 2013. “Poleward expansion of the
of disease and climate change in amphibian declines.” white-footed mouse (Peromyscus leucopus) under climate
PLOS Biology 6: e72. change: Implications for the spread of Lyme disease.”
Liu-Helmersson, J., H. Stenlund, A. Wilder-Smith, and PLOS One 8: e80724.
J. Rocklov. 2014. “Vectorial capacity of Aedes aegypti: Ef- Siraj, A. S., M. Santos-Vega, M. J. Bouma, D. Yadeta, D. R.
fects of temperature and implications for global den- Carrascal, and M. Pascual. 2014. “Altitudinal changes
gue epidemic potential.” PLOS One 9: e89783. in malaria incidence in highlands of Ethiopia and
MacDonald, G. 1957. The Epidemiology and Control of Malaria. Colombia.” Science 343: 1154–1158. https://doi.org/10
Oxford University Press. .1126/science.1244325.
280 W H AT DOE S THE FUTUR E HOL D?
Stapp, P., M. F. Antolin, and M. Ball. 2004. “Patterns of Yates, T. L., J. N. Mills, C. A. Parmenter, T. G. Ksiazek, R.
extinction in prairie dog metapopulations: Plague R. Parmenter, J. R. Vande Castle, C. H. Calisher, S. T.
outbreaks follow El Nino events.” Frontiers in Ecology and Nichol, K. D. Abbott, J. C. Young, M. L. Morrison, B. J.
the Environment 2: 235–240. Beaty, J. L. Dunnum, R. J. Baker, J. Salazar-Bravo, and
Vezzulli, L., R. R. Colwell, and C. Pruzzo. 2013. “Ocean C. J. Peters. 2002. “The ecology and evolutionary his-
warming and spread of pathogenic vibrios in the tory of an emergent disease: Hantavirus pulmonary
aquatic environment.” Microbial Ecology 65: 817–825. syndrome.” BioScience 52: 989–998.
Watson, R. T., M. C. Zinyowera, R. H. Moss, and In- Zhou, G., N. Minakawa, A. K. Githeko, and G. Yan. 2004.
tergovernmental Panel on Climate Change, Working “Association between climate variability and malaria
Group II. 1996. Climate Change, 1995: Impacts, Adaptations, epidemics in the East African highlands.” Proceedings of
and Mitigation of Climate Change: Scientific-Technical Analyses; the National Academy of Sciences 191: 2375–2380.
Contribution of Working Group II to the Second Assessment Report
of the Intergovernmental Panel on Climate Change. Cambridge
University Press.
PA RT V
283
284 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
3. A simple theory that can effec- year (Loarie et al. 2009); that rate can be
tively accommodate the biophysi- much higher, depending on the latitude,
cal dynamics important to species altitude, and topography of a given loca-
persistence tion. The velocity of change will be lowest
in mountainous biomes, as a result of topo-
4. Innovative protected area design and graphic variability, and higher in lowlands.
management approaches that trans- Lowland species will need to travel greater
late the theory into effective practice distances and at faster rates to track their
for conservation suitable climates than will their mountain-
dwelling counterparts. This suggests that
We discuss these components below and mountainous regions may be critically im-
provide an assessment of emerging ap- portant to maintaining biodiversity in some
proaches to address the adaptation of pro- regions by facilitating species movement to
tected areas to climate change. suitable climate.
Recent studies in paleoecology suggest
that species occupying discrete microcli-
T H E M AGN IT U DE , R AT E , TIM I NG, mates played an important role in species
A N D I N T E NSIT Y OF C L IM AT E C H A NGE responses to rapid climate change dur-
A N D PROT E CT E D A R E AS ing the Last Glacial Maximum (Stewart et
al. 2010). The effective role of microrefu-
In 2018, we passed an average of 1°C de- gia, holdouts, and stepping stones in spe-
grees of warming globally, and we expect cies-range shift dynamics depends on the
to reach an average of 2°C warming by magnitude and velocity of climate change
2050. With just 1°C warming, changes in (Hannah 2014), as well as direct and in-
species are documented across the globe direct impacts of extreme climatic events.
(e.g., Parmesan 2006), including distribu- Even so, it is possible that some species will
tional shifts, changes in timing of biologi- be able thrive in protected areas if they are
cal phenomena (e.g., flowering, breeding, able to find microenvironments that will al-
migration), and decoupling of coevolved low them to adapt to changing climate con-
species interactions. Species responses have ditions, even if only temporarily (Thomas
resulted in observed range shifts both pole- and Gillingham 2015), thus buying time for
ward and upward along elevational gradi- more significant management intervention.
ents (Parmesan 2006). The possibility for microrefugia or tempo-
Predicting species- and ecosystem-level rary stepping stones within and between
responses to future warming is difficult, existing protected areas will be very im-
as warming has not been, and will not be, portant in the successful adaptation of some
uniform across the globe. As we head to a species.
2°C increase in global average temperature Just as we are beginning to fully con-
by midcentury, we expect an increase of sider the potential conservation value of
more than 4°C in some geographies and microclimates for facilitating biodiversity
less than 0.5°C in others (IPCC 2014). In persistence, we are beginning to consider
general, we expect greater warming at the the potential consequences of the less pre-
poles than at the equator. This variation in dictable extreme events for the protection
the magnitude of change will manifest in of biodiversity. As is evident by the increas-
significant variations in the rate of change. ing frequency and magnitude of severe
On average, the velocity of the climate weather events worldwide, climate change
change (i.e., the instantaneous local veloc- is characterized not simply by shifts in
ity along Earth´s surface needed to maintain mean global temperature and precipitation
a constant temperature) is 0.42 km each alone but also by changes in the pattern
PROTECTED-A R EA MANAGEMENT AND CLIMATE CH ANGE 285
of extreme climatic and climate-induced city of information about the desirability,
events. To challenge the survival of spe- feasibility, and effectiveness of adaptation
cies even further, extreme climatic events and mitigation options (Lemieux et al.
will deliver punctuated impacts in time and 2011; Hannah 2008). In addition, much of
space that are likely to magnify the influ- this research aimed to solve ecological and
ence of average climatic trends and other biological problems (e.g., loss of species’
stressors. That is, the character and severity representation) caused by climate change in
of impacts from climate extremes depend protected areas, whereas impacts on local
not only on the extremes themselves but communities or their participation in the
also on the background exposure and vul- management solution are rarely discussed.
nerability of the species. It is the internal institutional publications,
or the “gray” literature, that are beginning
to deliver an integrated vision (i.e., includ-
GA PS I N PROT E CT E D -A R E A DE SIGN ing social and political aspects) of the role
A N D M A NAGE M E N T of protected areas under climate change
(e.g., Hansen et al. 2003; Dudley et al. 2010).
Effective policy, design, and management
of protected areas in the context of climate
The Geographic Gap
change require relevant science to inform
decision making. In the past decade there There are gaps in the geographic focus of
has been a surge in the number of scientific studies on the impact, vulnerability, and
articles focusing on the effects of climate adaptation of protected areas to climate
change on protected areas, which usually change, with a significant bias toward
provide recommendations for management North America and Europe (Heller and Za-
and design of protected areas to support ad- valeta 2009). The urgency to address the
aptation (Heller and Zavaleta 2009). Most impact of climate change on biodiversity
of these studies do not address the impli- is important worldwide, but the gap in in-
cations of the full range of direct and in- formation to support action is much higher
direct climate-induced impacts; therefore, in developing countries, where species ex-
they have not fully explored the range of tinction risk rates are high and financial
protected-area design and management ap- and technical capacities can be very low
proaches to facilitate species persistence in (Hannah et al. 2002). Although some of the
the future. The scientific literature has im- knowledge that is generated in the scientific
portant gaps, which limit the applicability literature is transferable to multiple regions,
of the design and management approaches it is necessary to address the geographic gap
into different situations and regions across and focus future research on less studied
the world. These gaps include the science- and more vulnerable regions to attend their
management gap, the geographic gap, and specific contexts and needs. For example,
the ecosystem gap. the consequences of climate change for
the representation of species in protected
areas in lowland tropical rainforest are
The Science-Management Gap
poorly documented (Thomas and Gilling-
Although many studies focus on basic sci- ham 2015). Furthermore, several strategies
entific issues emphasizing theoretical and/ developed for protected areas under cli-
or general principles of climate change mate change are “data hungry,” demanding
biology, only a few studies provide prac- high-quality information about geographi-
tical guidance for protected areas design cal distribution, vulnerability, and dispersal
and management and adaptation planning capacity of the target species. Clearly, these
(Heller and Zavaleta 2009). There is a pau- strategies are very difficult to implement in
286 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
protected areas located in regions of low new assemblage maintain key ecosystem
capacity where detailed information about services? The complexity of the answers to
species is scarce and species inventories are these simple questions illustrates the vexing
incomplete. nature of the issue at hand. First, we need
to explore which theories of species persis-
tence can inform our answer. There are two
The Ecosystem Gap
with which to begin, island biogeography
Finally, there are gaps in the type of the and metapopulation theory, complemented
ecosystems studied (e.g., terrestrial, marine, by a third emerging one, graph theory or
freshwater). Most of the research on climate network theory. These theories tell us that
change effects on biodiversity and their im- the expected number of species in the pro-
plications for conservation planning has tected area will depend on the rate at which
concentrated on the terrestrial and ma- species colonize (through in-migration or
rine systems, whereas studies that consider speciation) and leave (through emigration
freshwater diversity have been limited, or extinction) the protected areas. While
despite the high vulnerability of freshwa- the former rate depends on the size of the
ter diversity to climate change (Bush et al. pool and the degree of isolation of the pro-
2014). The combined species distribution- tected area or how reachable it is, the latter
climate models assessments for freshwater depends on area, shape, topography, and
needs to be improved rapidly to adequately condition of the surrounding agricultural,
address freshwater conservation planning urban, and natural land use matrix. In par-
that incorporates both biodiversity and hu- ticular, it will depend on whether the ma-
man needs in the face of climate change. To trix will allow for the species to leave the
do this, it will be important to understand protected area, as many species are very
the effects of climate change on the unique sensitive to movement across matrix habi-
properties of freshwater systems, such as re- tats, making it likely that they will become
gime flow and intercatchment connectivity increasingly isolated within protected ar-
(Bush et al. 2014). A broader analysis of all eas. This is, by no means, the whole story
ecosystems is needed because the effects of as rates will also depend on the number
climate change and requirements for bio- of species already present in the protected
diversity conservation differ across systems. area, their interaction, and the ability of
protected areas to support viable popula-
tions for different species.
T H E ORY OF C L IM AT E C H A NGE In Figure 22.1 we show the relationship
A N D PROT E CT E D -A R E A DE SIGN between body size and home-range size for
A N D M A NAGE M E N T marine and terrestrial species and, super-
imposed over it, the median size of marine
One of the pressing challenges to protect- and terrestrial protected areas. It is striking
ing biodiversity in the future is the extent that most protected areas are smaller than
to which protected-area networks will be the area required by an average individual
able to retain their biodiversity, provide of most marine and approximately 50 per-
connectivity for migrating species, and cent of the terrestrial vertebrate species
serve as refuges for new ones. In general, considered. Further, since the median size
this amounts to providing an answer to the of minimum viable populations (MVPs) is
following questions: If a given protected in the range of 1,000 to 4,000 individuals
area possesses S0 species today, how many (Brook, Traill, and Bradshaw 2006), Figure
species will it likely have by the end of the 22.1 implies that very few protected areas
century? How different will these species will be able to support species over the long
be from the original assemblage? Will this term unless connectivity is enhanced and
PROTECTED-A R EA MANAGEMENT AND CLIMATE CH ANGE 287
Figure 22.1. Relationship between the scale of habitat use should be borne in mind that it is very dif-
per individual in marine and terrestrial species. The dot- ficult to assess the number of species that
ted lines demarcate the current median size of all marine
(MPA) and terrestrial (TPA) protected areas. (Redrawn
have currently failed to expand because of
from McCauley et al. 2015.) The median in terrestrial lack of connectivity between protected ar-
PAs was calculated by considering all PAs described in eas and surrounding habitats, and so it is
the World Protected Area Data Base (IUCN and UNEP- difficult to assess its importance in com-
WCMC, 2015) that do not include marine habitats. parative terms, and improving connectivity
may not be the best option in all places and
the matrix in between protected areas be- times. Indeed, the usual strategy of restor-
comes more biodiversity friendly. This fig- ing or generating habitat corridors within
ure highlights the deep disconnect between landscapes may have unintended negative
the scale at which species interact with their effects associated with the spread of patho-
environments and how we manage land use gens, disturbances (e.g., fire), or increasing
to enhance species persistence (McCauley et synchrony in local population fluctuations
al. 2015). Thus, everything else being equal, (Simberloff and Cox 1987), thus leading to
even if a well-connected network of pro- metapopulation extinction due to regional
tected areas will do better than a small iso- stochasticity. Investments in connectivity
lated one, the average size of protected areas should be analyzed in the context of poten-
is so small that many species risk extinction tial positive and negative impacts as well as
if the hostility of the surrounding matrix the expected benefits that may be accrued
increases as a result of climate change. by investing in alternative actions to build
Available evidence shows that lack of resilience, such as increasing the number,
connectivity in fragmented landscapes has area, habitat quality, and heterogeneity of
negative impacts upon biodiversity con- protected areas.
servation and functioning and that species When it comes to the theoretical under-
do expand their ranges out of or into pro- standing of climate change impacts on spe-
tected areas while responding to climate cies, it is essential to bear in mind that the
change (Araújo et al. 2004; Thomas et al. problem of persistence is a problem of scale
2012). This suggests that connectivity is es- commensurability. That is, between the rate
sential in ensuring resilience. However, it
288 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
at which a landscape or habitat changes and and help buffer the impact of climate fluc-
the changes that a species can tolerate or tuation (Schindler et al. 2010).
adapt to, as the saying goes, it takes two to
tango. Habitats change because of changes
in their physical and chemical properties I N NOVAT I V E A PPROAC H E S TO A DA P T
but also because of changes in the biologi- PROT E CT E D A R E AS TO C L IM AT E
cal processes and species that are found in C H A NGE
those habitats; under climate change both
rates will increase. Unfortunately, little the- Several actions can be taken to improve
ory has been developed to understand the the role of protected area networks in pro-
issue of persistence under climate change, tecting biodiversity under climate change.
or in dynamic landscapes, and how it de- These range from maintaining current con-
pends on the climate change itself but in servation successes to innovative responses
interaction with the scale at which spe- specific to climate change.
cies perceive and interact with their en-
vironments (but see Marquet and Velasco
Ensure the Permanence and Quality
Hernandez 1997). Similarly, the impact of
of Existing Protected Areas
climate change on networks of interacting
species is similarly challenging and diffi- Protected areas are an important tool to
cult to tackle, first because interactions will conserve biodiversity and adapt to climate
likely change before species go extinct, and change. The current protected-area network
second because multiple anthropogenic provides a platform from which to build.
drivers affect them at the same time, which It is critical that the existing protected area
undermines the ability to make predictions. network be maintained and strengthened as
In recent years, simple static approaches current trends and future scenarios of land
(e.g., Araújo et al. 2011) and dynamic mod- use may threaten their persistence and vi-
els have been developed to make predic- ability. Probably most important, shifting
tions on the impact of climate change on climate conditions will affect where food
networks of interacting species. However, and energy crops can be grown to meet
the complexity of the issue still defies the the demands of a growing human popula-
identification of first principles (e.g., Mar- tion, thus creating the potential for land-use
quet et al. 2014) that can be translated conflicts. Hannah et al. (2013) studied the
into simple guidelines for management. In changing geographic suitability of regions
many ways, this is an empirical problem as for wine grape production under a chang-
most networks are not monitored in time or ing climate and highlighted the potential
at different locations in space, which hin- for a shift into existing protected areas and
ders our capacity to understand how they important matrix habitat for species of con-
change and how resilient they are. The chal- cern across the globe. Predicted shifts in ag-
lenge is also theoretical, as we do not fully ricultural suitability such as these can create
understand how richness, interaction type, competition between conservation and food
and interaction strengths affect stability and production for land and will have a signifi-
resilience; nor do we know how to include cant impact on protected areas and habitat
these principles in conservation planning connectivity for wide-ranging species.
and the design of protected-area networks.
Simple theory and empirical evidence show
Expand or Modify Protected Areas
that, in general, species diversity and popu-
and Networks
lation diversity can bring, through portfolio
effects, stability to ecosystems and the ser- As species ranges shift in response to
vices they provide (e.g., Hilborn et al. 2003) changes in climatic factors, many are pro-
PROTECTED-A R EA MANAGEMENT AND CLIMATE CH ANGE 289
jected to move outside the boundaries of ex- gests that protecting the “ecological stage”
isting protected lands and waters. There will for processes as well as preserving particu-
need to be the modification of boundaries of lar “actors” (species, ecosystems) may be
existing protected areas, or establishment of important in protecting species diversity
new protected areas to cover lands and wa- over time.
ters likely to be important to these species in
the future. Even with new protected areas,
Assist Migration
it is predicted that by 2050, 20 percent of
species will have disappeared from reserves Assisted migration is a strategy designed to
(Hannah et al. 2007); thus, it is important to help species overcome dispersal barriers
keep in mind that the function of protected associated with tracking climate shifts by
areas will not continue under business-as- physically relocating them outside their his-
usual emissions trajectories (Thomas and torical range to climatic zones considered
Gillingham 2015), and more innovative and suitable. The outcome of assisted migration
aggressive management of species and pro- is not qualitatively different from enhanc-
tected areas will be necessary. ing habitat connectivity (Lawler and Olden
2011), and although this may be the only
strategy that will be effective for achieving
Use Protected Areas to Enhance
the persistence of some species, many are
Connectivity between Present
opposed to it given the poor performance of
and Future Populations
past efforts (Ricciardi and Simberloff 2009).
Regardless of how large an individual pro-
tected area is, if it is an island of habitat sur-
Enhance Resilience
rounded by inhospitable habitat, it will be
difficult or impossible for species to migrate Resilience has emerged as perhaps the
as climate shifts. Enhancing connectivity dominant paradigm for addressing climate
with special attention to microclimate step- change impacts on protected areas. Resil-
ping stones among protected areas to enable ience can refer to the ability of a system to
species movements should be employed to return to its original state following a per-
ensure resilience of protected areas over time turbation, to maintain functionality during
(e.g., Belote et al. 2016). For climate change it the course of a transition, or to self-organize
is especially important to connect present and in maintaining basic system traits and func-
future populations of species, which will not tioning (Folke et al. 2004). In the practice
often be the same as connecting protected of biodiversity conservation, the concept
areas. Although connections between pro- emphasizes the notion that robust protected-
tected areas for large mammals may provide area networks will be better able to ensure
connectivity for climate change, connecting persistence of species across the landscape in
present and future populations of plants and the context of climate change. How to do
nonvertebrates may involve different areas this requires a multifaceted approach includ-
and be less area demanding. ing abating not only climate change but also
other global change drivers and fostering
legislation and education to better preserve
Protect Enduring Geophysical Features
the seminatural matrix or landscapes where
Protecting geophysical settings will be im- human uses coexist with protected areas.
portant for protecting species diversity in
the future (Anderson and Ferree 2010; Brost
Create Stepping Stones
and Beier 2012). The relationship between
species diversity and geophysical setting Existing conservation tools for creating
(elevation, slope, substrate, and aspect) sug- protected areas like purchase fee or ease-
290 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
ments, though valuable, are static and in- Enhance Synergies between Protected-Area
sufficient to maintain biodiversity over Design and Management for the Adaptation
time. To encourage investment at the size and Resilience of Human Communities to
and rate required, we must develop con- Climate Change
servation tools that are adaptive and dy-
Some of the impacts of climate change are
namic over time, in response to the chang-
unavoidable, but protected areas may offer
ing resource. New conservation tools such
options for biodiversity and society adapta-
as habitat exchange or bird return are de-
tion to these impacts (Heller and Zavaleta
signed to achieve spatially and temporally
2009; Welch 2005). By helping maintain
significant landscape-scale conservation
natural ecosystems and enhancing their re-
through participation from private land-
silience to climate change, protected areas
owners. The Habitat Exchange allows de-
can reduce the risk of disasters such as hur-
velopers to offset their impacts on habitat
ricanes, droughts, floods, and landslides,
and species by purchasing credits gener-
which will likely increase as a consequence
ated through conservation actions of pri-
of increasing temperature and deforesta-
vate landowners. The Birds Returns pro-
tion. In the same way, the protection of
gram typically pays farmers to flood their
natural ecosystems can contribute to secur-
fields at certain levels and certain times of
ing the supply of ecosystem services to local
the year to provide habitat for birds. Both
communities and providing additional in-
of these are dynamic, incentive-based ap-
surance against the predicted instability of
proaches that provide economic opportu-
agriculture, fisheries, and water resources.
nity for private landowners while deliver-
ing scientifically robust benefits to species
and habitat.
CONSE RVATION I N A
C H A NGI NG C L IM AT E
A review of 22 years of recommendations.” Biological Marquet, Pablo A., and Jorge X. Velasco Hernandez.
Conservation 142 (1): 14–32. https://doi.org/10.1016/j 1997. “A source-sink patch occupancy metapopulation
.biocon.2008.10.006. model.” Revista chilena de historia natural 70 (3): 371–380.
Hilborn, Ray, Thomas P. Quinn, Daniel E. Schindler, and McCauley, Douglas J., Malin L. Pinsky, Stephen R.
Donald E. Rogers. 2003. “Biocomplexity and fisheries Palumbi, James A. Estes, Francis H. Joyce, and Robert
sustainability.” Proceedings of the National Academy of Sciences R. Warner. 2015. “Marine defaunation: Animal loss in
100 (11): 6564–6568. https://doi.org/10.1073/pnas the global ocean.” Science 347 (6219). https://doi.org/
.1037274100. 10.1126/science.1255641.
Hole, David G., Stephen G. Willis, Deborah J. Pain, Parmesan, Camille. 2006. “Ecological and evolutionary
Lincoln D. Fishpool, Stuart H. M. Butchart, Yvonne responses to recent climate change.” Annual Review of
C. Collingham, Carsten Rahbek, and Brian Huntley. Ecology Evolution and Systematics 37: 637–669. https://doi
2009. “Projected impacts of climate change on a .org/10.1146/annurev.ecolsys.37.091305.110100.
continent-wide protected area network.” Ecology Letters Peters, Robert L., and Joan D. S. Darling. 1985. “The
12 (5): 420–431. https://doi.org/10.1111/j.1461-0248 greenhouse effect and nature reserves.” BioScience 35
.2009.01297.x. (11): 707–717. https://doi.org/10.2307/1310052.
IPCC. 2014. Climate Change 2014: Synthesis Report. Contribution Pressey, Robert L., Mar Cabeza, Matthew E. Watts, Rich-
of Working Groups I, II and III to the Fifth Assessment Report of the ard M. Cowling, and Kerrie A. Wilson. 2007. “Conser-
Intergovernmental Panel on Climate Change, ed. Core Writing vation planning in a changing world.” Trends in Ecology
Team, R. K. Pachauri, and L. A. Meyer. IPCC. and Evolution 22 (11): 583–592. https://doi.org/10.1016/
Jackson, Stephen T., and Richard J. Hobbs. 2009. “Eco- j.tree.2007.10.001.
logical restoration in the light of ecological history.” Ricciardi, Anthony, and Daniel Simberloff. 2009. “As-
Science 325 (5940): 567–569. https://doi.org/10.1126/ sisted colonization is not a viable conservation strat-
science.1172977. egy.” Trends in Ecology and Evolution 24 (5): 248–253.
Lawler, Joshua J., and Julian D. Olden. 2011. “Reframing https://doi.org/10.1016/j.tree.2008.12.006.
the debate over assisted colonization.” Frontiers in Ecol- Rodrigues, Ana S. L., Sandy J. Andelman, Mohamed I.
ogy and the Environment 9 (10): 569–574. https://doi.org/ Bakarr, Luigi Boitani, Thomas M. Brooks, Richard
10.1890/100106. M. Cowling, Lincoln D. C. Fishpool, Gustavo A. B. da
Lawson, Callum R., Jonathan J. Bennie, Chris D. Thomas, Fonseca, Kevin J. Gaston, Michael Hoffmann, Janice
Jenny A. Hodgson, and Robert J. Wilson. 2014. “Active S. Long, Pablo A. Marquet, John D. Pilgrim, Robert
management of protected areas enhances metapopu- L. Pressey, Jan Schipper, Wes Sechrest, Simon N. Stu-
lation expansion under climate change.” Conservation art, Les G. Underhill, Robert W. Waller, Matthew E. J.
Letters 7 (2): 111–118. https://doi.org/10.1111/conl Watts, and Xie Yan. 2004. “Effectiveness of the global
.12036. protected area network in representing species diver-
Lemieux, Christopher J., Thomas J. Beechey, Daniel J. sity.” Nature 428: 640–643. https://doi.org/10.1038/
Scott, and Paul A. Gray. 2011. “The state of climate nature02422.
change adaptation in Canada’s protected areas sector.” Schindler, Daniel E., Ray Hilborn, Brandon Chasco,
Canadian Geographer/Geographe Canadien 55 (3): 301–317. Christopher P. Boatright, Thomas P. Quinn, Lauren
https://doi.org/10.1111/j.1541-0064.2010.00336.x. A. Rogers, and Michael S. Webster. 2010. “Population
Lenoir, J., J. C. Gégout, P. A. Marquet, P. de Ruffray, and diversity and the portfolio effect in an exploited spe-
H. Brisse. 2008. “A significant upward shift in plant cies.” Nature 465 (7298): 609–612. https://doi.org/10
species optimum elevation during the 20th century.” .1038/nature09060.
Science 320 (5884): 1768–1771. https://doi.org/10 Seastedt, Timothy R., Richard J. Hobbs, and Katharine N.
.1126/science.1156831. Suding. 2008. “Management of novel ecosystems: Are
Loarie, Scott R., Philip B. Duffy, Healy Hamilton, novel approaches required?” Frontiers in Ecology and the
Gregory P. Asner, Christopher B. Field, and David D. Environment 6 (10): 547–553. https://doi.org/10.1890/
Ackerly. 2009. “The velocity of climate change.” Na- 070046.
ture 462 (7276): 1052–1055. https://doi.org/10.1038/ Shaw, M. Rebecca, Kirk Klausmeyer, D. Richard Cam-
nature08649. eron, Jason Mackenzie, and Patrick Roehrdanz. 2012.
Margules, Chris R., and Robert L. Pressey. 2000. “Sys- “Economic costs of achieving current conservation
tematic conservation planning.” Nature 405: 243–253. goals in the future as climate changes.” Conservation Bi-
https://doi.org/10.1038/35012251. ology 26 (3): 385–396. https://doi.org/10.1111/j.1523
Marquet, Pablo A., Andrew P. Allen, James H. Brown, -1739.2012.01824.x.
Jennifer A. Dunne, Brian J. Enquist, James F. Gillooly, Shaw, M. Rebecca, Linwood Pendleton, D. Richard
Patricia A. Gowaty, Jessica L. Green, John Harte, Steve Cameron, Belinda Morris, Dominique Bachelet, Kirk
P. Hubbell, James O’Dwyer, Jordan G. Okie, Annette Klausmeyer, Jason MacKenzie, David R. Conklin,
Ostling, Mark Ritchie, David Storch, and Geoffrey B. Gregory N. Bratman, James Lenihan, Erik Haunre-
West. 2014. “On theory in ecology.” BioScience 64 (8): iter, Christopher Daly, and Patrick R. Roehrdanz.
701–710. https://doi.org/10.1093/biosci/biu098. 2011. “The impact of climate change on California’s
PROTECTED-A R EA MANAGEMENT AND CLIMATE CH ANGE 293
ecosystem services.” Climatic Change 109 (1): 465–484. responses of species in space and time.” Proceedings of
https://doi.org/10.1007/s10584-011-0313-4. the Royal Society B: Biological Sciences 277 (1682): 661–671.
Simberloff, Daniel, and James Cox. 1987. “Consequences https://doi.org/10.1098/rspb.2009.1272.
and costs of conservation corridors.” Conservation Biology Thomas, Chris D., and Phillipa K. Gillingham. 2015.
1 (1): 63–71. https://doi.org/10.1111/j.1523-1739.1987 “The performance of protected areas for biodiversity
.tb00010.x. under climate change.” Biological Journal of the Linnean So-
Soares-Filho, Britaldo, Paulo Moutinho, Daniel Nepstad, ciety 115: 718–730. https://doi.org/10.1111/bij.12510.
Anthony Anderson, Hermann Rodrigues, Ricardo Thomas, Chris D., Alison Cameron, Rhys E. Green, Mi-
Garcia, Laura Dietzsch, Frank Merry, Maria Bowman, chel Bakkenes, Linda J. Beaumont, Yvonne C. Collin-
Leticia Hissa, Rafaella Silvestrini, and Claudio Maretti. gham, Barend F. N. Erasmus, Marinez Ferreira de
2010. “Role of Brazilian Amazon protected areas in Siqueira, Alan Grainger, Lee Hannah, Lesley Hughes,
climate change mitigation.” Proceedings of the National Brian Huntley, Albert S. van Jaarsveld, Guy F. Midgley,
Academy of Sciences 107 (24): 10821–10826. https://doi Lera Miles, Miguel A. Ortega-Huerta, A. Townsend
.org/10.1073/pnas.0913048107. Peterson, Oliver L. Phillips, and Stephen E. Wil-
Staudinger, Michelle D., Nancy B. Grimm, Amanda liams. 2004. “Extinction risk from climate change.”
Staudt, Shawn L. Carter, F. Stuart Chapin III, Peter Nature 427 (6970): 145–148. https://doi.org/10.1038/
Kareiva, Mary Ruckelshaus, and Bruce A. Stein. 2012. nature02121.
Impacts of Climate Change on Biodiversity, Ecosystems, and Ecosys- Tognelli, Marcelo F., Pablo I. Ramirez de Arellano, and
tem Services: Technical Input to the 2013 National Climate Assess- Pablo A. Marquet. 2008. “How well do the existing and
ment. Cooperative Report of the 2013 National Climate proposed reserve networks represent vertebrate spe-
Assessment. cies in Chile?” Diversity and Distributions 14 (1): 148–158.
Stein, Bruce A., and M. Rebecca Shaw. 2013. “Biodiver- https://doi.org/10.1111/j.1472-4642.2007.00437.x.
sity conservation in a climate-altered future.” In Suc- Turner, Bryan M. 2009. “Epigenetic responses to envi-
cessful Adaptation to Climate Change: Linking Science and Practice ronmental change and their evolutionary implica-
in a Rapidly Changing World, ed. S. C. Moser and M. T. tions.” Philosophical Transactions of the Royal Society B: Biologi-
Boykoff, 50–66. Routledge. cal Sciences 364 (1534): 3403–3418. https://doi.org/10
Stewart, John R., Adrian M. Lister, Ian Barnes, and .1098/rstb.2009.0125.
Love Dalén. 2010. “Refugia revisited: Individualistic
CASE STUDY 9
Extinction risk from climate change first made a major impact on international climate
change policy dialogue in 2004, following publication of the first global estimates of pos-
sible species losses due to climate change (Thomas et al. 2004). Following the publication of
these initial estimates, extinction risk was discussed in climate change debates in the House
of Lords and was the subject of US Senate hearings. While these and other, more multilateral
debates (as in the Convention for Biological Diversity and UN Framework Convention on
Climate Change) finally began to grapple with this issue, the concept of extinction risk itself
had much deeper roots in climate change biology.
One of the most important insights from preselected to cope with cooler conditions
paleoecology is that species’ ranges shift in and climate transitions. But all these Paleo
response to climate change. Such shifts have events offer at best hints of what the future
demonstrably been the overwhelming re- may hold. There is no exact past analog for
sponse of both plant and animal species to the warming that is now unfolding.
climate change for several millions of years. Beginning to quantify the possible scale
The process can even lead to speciation as of the extinction problem associated with
ranges are fragmented (e.g., Hewitt 2000). human-caused climate change was possible
However, these range shifts can also result only with the evolution of species distribu-
in extinction when species ranges run into tion models (ecological niche models) prior
barriers or are in competition with new to the 2004 landmark paper. Large-scale
species, or they can involve catastrophic modeling of thousands of species across
loss of range. In a corollary to this, regions continental or subcontinental domains
that experienced relatively less recent pa- emerged in the late 1990s, enabling authors
leoclimate change tend to have higher bio- to assess extinction risk and entire floras
diversity (Dynesius and Janssen 2013). and faunas. A group of these researchers
The lesson from paleoecology is that pooled several large-scale modeling efforts
context matters. Major extinction events in to produce the 2004 estimates.
the past 500 million years are associated, Although the first estimates of extinction
directly or indirectly, with climate change. risk from climate change attracted much
But some major climatic events are not as- public policy attention, they also garnered
sociated with megaextinction, although substantial scientific criticism. However,
they may have driven major shifts in domi- the general magnitude of extinction risk
nant taxa. The Paleocene-Eocene Ther- in these initial estimates has now been
mal Maxima (a rapid warming period), corroborated through several independent
for instance, is associated with the rise of studies (Malcolm et al. 2006; Sinervo 2010;
primates and many modern mammals but Urban 2015) and in theory (Deutsch et al.
with few global extinctions. More recently, 2008; see Plate 4). There is now little expert
there was a wave of extinctions as the world disagreement that climate change poses a
cooled into the ice ages but few extinctions major extinction risk, perhaps to hundreds
associated with the glacial and interglacial of thousands of species or more around the
transitions themselves. An explanation for
this is that the regime shift caused extinc- Copyright © 2019 by Yale University. All rights reserved.
tions, but then the remaining species were ISBN 978-0-300-20611-1.
294
E XTINCTION R ISK FROM CLIM ATE CH A NGE 295
world, which appears to be strongly de- species extinction risk over the temperature
pendent on the rate of climate change (e.g., range of 2°C–3°C above preindustrial levels
O’Neill et al. 2017). but identified a logarithmic increase with
The Intergovernmental Panel on Climate increasing temperature, indicating acceler-
Change (IPCC) published a consensus esti- ating risk of extinction. More detailed work
mate of species extinction (Fischlin et al. on several key taxonomic groupings and us-
2007) that used several of the studies in- ing even finer-grained species and climate
cluded in the 2004 Thomas study, together data (Warren et al. 2018) supports a finding
with several studies published subsequently. of an accelerating trend of projected species
It estimated that “approximately 20 to 30 geographic range loss. They report that the
percent of plant and animal species assessed number of species projected to lose over
so far (in an unbiased sample) are likely to half their range doubles between 1.5°C and
be at increasingly high risk of extinction as 2°C above the preindustrial baseline, an in-
global mean temperatures exceed a warm- crease of more than 8 times with 4.5°C of
ing of 2°C to 3°C above pre-industrial lev- warming.
els (medium confidence)” (Climate Change Whatever the proportional loss of spe-
2007, 213). This study was an advance on cies projected, it must be multiplied by the
Thomas et al. (2004) in the sense that the number of species in the world to estimate a
estimate was made after all local studies global extinction risk estimate. This is what
had been referenced to a global tempera- Thomas et al. did in the press release for the
ture increase, allowing for impacts to be 2004 study, to translate it for the media and
directly comparable on this temperature the public. With 6 million–10 million spe-
scale. This was a significant effort with the cies in the world, conservatively, 30 percent
tools of the day. IPCC authors deliberately risk translates to between 1.8 million and 3
steered away from the phrase “committed million species. The majority of these are
to extinction,” and the species area curve insects, so what happens to insects is criti-
technique, using rather the raw projections cal to these estimates. Few modeling stud-
of species losses due to projected total range ies have addressed large numbers of insects,
loss. The phrase “at increasingly high risk of but we know that most insect diversity is
extinction” was used to indicate a signifi- narrowly distributed in tropical forests,
cant threat to biodiversity. It was possible to where recent upticks in drought and fire as-
focus on a policy relevant global warming sociate with or are compounded by climate
range of 1.5°C–2.5°C above a 1980–1999 change, are of great concern.
average baseline (2°C–3°C above preindus- This risk of extinction is further intensi-
trial average baseline) temperatures because fied because of extensive human modifica-
of a concentration of studies in that range, tion of landscapes. The combination of spe-
allowing the authors to average a number cies ranges moving in response to climate
of studies. A simple linear regression on the change and human-dominated landscapes
results revealed a possible linear response full of hard edges is what the late Stephen
of extinction risk with warming, a trend Schneider referred to as “a no-brainer for an
that was described in a subsequent CBD re- extinction spasm.” In its most recent report,
port as a “roughly 10 percent increase with IPCC (Settele et al. 2015) concluded that “a
every degree C warming” (Convention on large fraction of species faces increased ex-
Biological Diversity 2009). tinction risk due to climate change during
Urban (2015) published a revised es- and beyond the 21st century, especially as
timate of extinction risk that used more climate change interacts with other pres-
studies and employed a superior weighting sures . . . (high confidence)” (Climate
system based on numbers of species per Change 2014, 275). This study explicitly
study. This study reduced the estimate of quantified how most plant species would
296 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
297
Box 23.1 Glossary
Figure 23.1. Classification tree for 2009 crop yields in For example, the higher-yield “8/10” leaf (with dashed
Europe as explained by growing season weather and black border at bottom) indicates that eight Western
SOC. The tree represents the partitioning of 2009 crop- Europe yield observations with higher yields did not
specific growing season weather and SOC across Europe experience very extreme growing season weather and
that best explains a sample of observed 2009 yields. Ob- had an SOC of 25 g kg-1 or more. In contrast, only two
served yields are either placed in the lower-yield bin (a lower Western European yields had similar weather and
bottom 50th percentile yield observation for the given soil conditions. In other words, observations on this
crop) or the higher-yield bin (a top 50th percentile yield branch are predicted to have a top half yield. The other
observation for the given crop). At each node a yes to the SOC node, SOC ≥ 23, indicates that observations with
weather or soil characteristic means a move to the left that branch’s growing season weather profile are much
best fits the data and a no means a move to the right best more likely to have a top half yield if SOC < 23 g kg-1.
fits the data (Loh 2011; Varian 2014). A yes on the East- In other words, in this leaf’s particular growing season
ern Europe nodes indicates that the observation is from weather profile, too much SOC is associated with lower
Eastern Europe. Eastern European farms tend to use less yields. All in all, SOC impacts yield at the margin while
chemical inputs and are less capitalized than their West- growing season weather and agricultural investment
ern European counterparts. At the end of each branch, (crudely represented by Eastern vs. Western European
called a leaf, the first number gives the count of observa- observations) are the main drivers of observed yields.
tions on that branch that are in the given yield bin and See http://www.bowdoin.edu/faculty/e/enelson/ for
the second is the number of observations on that branch. more details on this analysis.
2014). Economists have demonstrated that is projected to become drier while southern
small payments encourage farmers to pro- China gets wetter (Piao et al. 2010). Such
vide pollinator habitat when accompanied changes will force northern farmers to
by larger fines for any subsequent habitat adopt more drought-tolerant crops or, less
destruction (Cong et al. 2014). likely, devise ways to transport water from
As global demand for food grows and south to north (Piao et al. 2010). In other
rainfall patterns change, farmers will in- places, climate change may cause increased
creasingly look to irrigation as an adapta- precipitation in the winter and decreased
tion measure. For example, northern China precipitation during the growing season.
ECOSYSTEM-BASE D A DA P TATION 301
Storing non-growing-season precipitation cities, sea-level rise (SLR) and storm surge
in networks of constructed retention ponds are direct consequences of climate change
and restored wetlands could ensure water that require management. (We address SLR
availability during increasingly dry grow- management specifically in the coastal-zone
ing seasons (Baker et al. 2012) while also section.) Green infrastructure and urban
reducing flood risk and providing wildlife greenspace are two related EbA approaches
habitat. that can be used to address many of these is-
Farmers of agroecosystems, subsistence sues related to heat and water management.
farmers, and small-scale farmers generally Green infrastructure alternatives to the
tend to rely much more heavily on ecosys- hard or “gray” infrastructure typically used
tem services to manage uncertainty and for stormwater management (e.g., drains,
environmental variability than do large- pipes) include green roofs, bioswales, rain
scale production farmers (Tengö and Bel- gardens, and constructed retention ponds
frage 2004). Grazing animals on crop fields and wetlands. The vegetation and soils as-
and including grasses in crop rotations are sociated with green infrastructure intercept
two tactics still widely used by subsistence precipitation and reduce the rate and vol-
farmers to maintain SOC and biodiversity. ume of runoff (Gill et al. 2007; Pappalardo
Intercropping and maintenance of pollina- et al. 2017). For example, a project in Seattle,
tor and natural pest habitat also help main- Washington, that included bioswales, re-
tain acceptable yields. Although these agri- tention ponds, and a series of stepped pools
cultural systems are much more sustainable retained 99 percent of wet-season runoff
than the high-yield systems described (Horner et al. 2004). Similarly, a review of
above, their low productivity means that green roofs in German cities showed that
they will not contribute significantly to intensive green roofs (with substrate >150
global food supplies. Instead, if the global mm) could retain 75 percent of annual run-
agriculture system is to become more re- off (Mentens et al. 2006). In other words,
silient to climate change, the more input- green infrastructure can mitigate urban
intensive farmers will have to adopt agro- flooding by storing some of the excess
ecosystem techniques that are compatible water created by a storm. Further, any ur-
with high-yield farming. ban drought after an extreme precipitation
event can be alleviated by the slow release
of this excess water.
M A K I NG U R BA N L I V I NG MOR E Ground and rooftop plantings in lieu of
SUSTA I NA BL E conventional dark materials and pavement
reduces the UHI in general and may reduce
According to the United Nations (2014), mortality impact of a heat wave specifically
66 percent of the world’s population will (Li et al. 2014). Alternatively, “white” and
live in urban areas by 2050, which means “cool” roofs, or even lighter-colored pave-
that the majority of the world’s people will ments that have higher albedo, can reduce
directly experience climate change and at- the UHI and the effects of heat waves, al-
tempt to adapt to it in the urban environ- though these approaches do not moderate
ment. Warmer temperatures will exacer- stormwater runoff. Co-benefits of green in-
bate the urban heat island (UHI) effect and frastructure include water-quality improve-
the frequencies of extreme weather events ments due to filtration of natural pollutants
are expected to increase in urban areas. Ex- and a reduction of water temperatures, aes-
treme heat and precipitation events in cities thetic values of plantings, the potential for
increase human mortality rates and ham- wildlife and pollinator habitat provision-
per the ability of infrastructure to perform ing, and opportunities for small-scale urban
adequately (IPCC 2014). In many coastal agriculture.
302 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
Expanding and conserving existing ur- (IPCC 2014). Conserving existing ecosys-
ban greenspaces in and around cities— tems, restoring degraded ones, and pursu-
from pocket parks to vegetated corridors ing integrated management to reduce the
to protected forests—provide myriad eco- cumulative risks from local stressors all
system services and inherently entail EbA. have the potential to enhance and main-
Forested lands and street trees increase tain the functioning of coastal and marine
evaporative cooling and shade pavement, ecosystems and the benefits they provide
which help reduce the magnitude of the (Ruckelshaus et al. 2014; Morris et al. 2018).
UHI, the impacts of heat waves, and energy Below we discuss these EbA strategies in
consumption (e.g., for air conditioning; light of two important services: coastal
Parmova et al. 2012). For example, adding storm protection and fisheries production.
just 10 percent green cover to urbanized Rising seas and potential increases in the
parts of Greater Manchester, in the United intensity and frequency of storms pose risks
Kingdom, is projected to keep maximum to the 200 million people living in coastal
summer surface temperatures at 29°C in regions worldwide (IPCC 2014). By attenu-
the 2080s, compared to 32°C with current ating waves and storm surge, coastal and
cover and 35°C with a 10 percent loss of marine ecosystems such as wetlands, coral
green cover (Gill et al. 2007). reefs, and coastal forests can help reduce
Restoring native species or planting the impacts of such hazards (Shepard et al.
drought-resistant species or hybrids can 2011; Arkema et al. 2013). Conserving ex-
help ensure urban tree health while also isting habitats and restoring degraded ones
promoting biodiversity in plantings, which in regions with low-lying sandy and muddy
helps prevent outbreaks of species-specific coastlines (e.g., the US east and gulf coasts)
diseases and pests (Alvey 2006; Paap et al. may effectively halve the number of people
2017), particularly if urban trees are al- at high risk under climate change (Arkema
ready experiencing climate-related stress. et al. 2013) (Figure 23.2). Funded by the US
Co-benefits of increased greenspaces, espe- federal government in the wake of the 2012
cially urban forests, include carbon seques- storm Hurricane Sandy, several innovative
tration, improved air quality, psychological projects involve building reefs to serve as
benefits, and opportunities for recreation, “natural breakwaters” to attenuate waves
in addition to other benefits enumerated and reduce erosion while providing habi-
above and in the forest discussion below. tat for fish, shellfish, and lobsters.2 Leverag-
ing ecosystems for coastal defense is often
less costly to implement and maintain than
COASTA L STOR M PROT E CTION hard infrastructure approaches including
A N D F ISH E R IE S PRODUCTION seawalls and levees (Morris et al. 2018).
Increasing ocean temperatures and
Coastal and marine ecosystems provide changing water chemistry are disrupting
a diversity of benefits. Fish and shellfish the delivery of marine food and the live-
are important sources of sustenance and lihoods that facilitate this service (Pinsky
protein, and many coastal habitats protect and Mantua 2014; Weatherdon et al. 2016).
infrastructure from storm surge and offer For example, in the northwestern Atlantic,
opportunities for recreation and aesthetic 24 out of 36 commercially exploited fish
enjoyment. However, ocean acidification showed significant range (latitudinal and
and warming are leading to shifts in the depth) shifts between 1968 and 2007 due to
distribution of economically and ecologi- warming water (Nye et al. 2009). Changes
cally important ecosystems and species, in water chemistry affect the calcification
while more frequent and intense storms rates of marine organisms (IPCC 2014),
increase the threat to coastal infrastructure many of which are an important food
ECOSYSTEM-BASE D A DA P TATION 303
Figure 23.2. Exposure of the US coastline and coastal and marine protected areas (MPAs) to safe-
populations to sea-level rise in 2100 (A2 scenario) and guard biodiverse sites and sustain fish stocks
storms. Darker colors indicate regions with more ex-
posure to coastal hazards (index >3.36). The bar graph
to ensure food security in coastal areas.3
shows the population living in areas most exposed to Although alternative approaches to marine
hazards (darkest 1 km2 coastal segments in the map) food production such as aquaculture may
with protection provided by habitats (dotted bars) and offset some climate impacts on natural sys-
the increase in population exposed to hazards if habitats tems, these alternatives come with risks of
were lost owing to climate change or human impacts
(white bars). Letters on the x-axis represent US state ab-
disease, habitat destruction, and pollution
breviations. This figure first appeared in the journal Na- (Ruckelshaus et al. 2014).
ture Climate Change (Arkema et al. 2013). Conservation, restoration, and integrated
management of coastal and marine ecosys-
source (e.g., oysters) or provide important tems have numerous co-benefits. Seagrasses
nursery and adult habitats for fishes (e.g., and mangroves sequester carbon, such that
corals). Coral conservation is particularly degradation and conversion of these sys-
challenging in tropical systems where pol- tems globally release 0.15–1.02 Pg CO2 an-
lution, sedimentation, and unsustainable nually and result in damages of US$6–$42
fishing stressors have not been addressed as billion annually (Pendleton et al. 2012).
successfully as in other parts of the world Healthy coastal and marine ecosystems also
(Ruckelshaus et al. 2014). Efforts such as the provide tourism opportunities and support
multilateral Coral Triangle Initiative draw livelihoods in this sector. Given these addi-
on ecosystem-based fisheries management tional co-benefits, EbA approaches are often
304 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
superior to other options, but they do have bucco et al. 2008). These approaches also
their challenges. In the case of coastal pro- increase soil moisture and groundwater
tection services, perceptions of risk can be recharge, thereby providing direct benefits
higher with green infrastructure than with for agriculture, facilitating downstream
built systems, and thus challenging to im- groundwater withdrawals, and reducing
plement. Likewise, fisheries management fire risk (Ellison et al. 2017), among other
can require coordination among many en- forest-related co-benefits identified in the
tities, making it particularly challenging. forest discussion below.
It is likely that climate change will fur-
ther reduce water availability in areas cur-
M A I N TA I N I NG F R E SH WAT E R rently suffering from water scarcity (Gos-
R E SOU RC E S ling and Arnell 2016). Indeed, 1.3 billion
people already live in water-scarce regions,
Climate change is projected to substantially and global warming of 2°C, 3.5°C, and
and nonlinearly affect surface water and 5°C are projected to expose an additional 8
groundwater resources. These impacts are percent, 11 percent, and 13 percent of the
likely to progress more quickly in heavily world population to greater water scarcity,
populated areas (Gerten et al. 2013). Geo- respectively (Gerten et al. 2013). Beyond
graphic and temporal shifts in precipitation the problem of limited water for direct
regimes will interact with changing tem- human consumption and agricultural pur-
peratures to create more prolonged drought poses, water shortages may result in habi-
or, conversely, flooding associated with ex- tat loss for pollinators and other species of
treme weather events (IPCC 2014), impacts economic importance, increased threat of
that will be felt most acutely in heavily pop- forest fires, and saltwater intrusion into
ulated areas (Gerten et al. 2013). overdrawn aquifers. Substantial efficiencies
Traditional flood management strategies, can be achieved in water use by adopting
such as straightening river channels and practices such as upgrading leaky water-
building dikes and levees, are vulnerable delivery systems, recycling wastewater, and
to failure and largely pass potential flood- implementing more efficient agricultural
ing problems downstream. EbA strategies practices (IPCC 2014).
for reducing the impact of floods enable Some of these solutions, however, may
excess water to spread into side channels be prohibitively expensive and will of-
and beyond riverbanks, where it slows and fer little benefit if natural ecosystems and
infiltrates soils, thereby reconnecting rivers processes that augment or ensure water
with floodplains, and holding excess water provisioning are not protected. As noted in
in natural areas above population centers the forest discussion, forest ecosystems are
(Palmer et al. 2009). Stream and riparian critical for the sustainable water delivery
restoration that increases large woody de- to over a third of the world’s largest cities
bris, structural habitat, and channel com- (Dudley and Stolten 2003). The protection
plexity encourages incised channels to ag- of high-elevation wetlands and peatlands—
grade (Palmer et al. 2009) and improves such as the Andean bofedales—may be criti-
overall ecosystem health (Thompson et al. cal for the persistence of pastoral native
2018). Beyond the stream channel, affor- communities in otherwise inhospitable
estation and reforestation increase evapo- environments; these fragile ecosystems
transpiration and can prolong snow cover, are extremely sensitive to climatic changes
thereby reducing downstream flooding by (Squeo et al. 2006). Elsewhere, natural
up to 54 percent in drier areas and 15 per- and constructed wetlands have been effec-
cent in more humid areas, as demonstrated tively used to retain surface water, recharge
in four South American case studies (Tra- groundwater, and filter out pollutants; these
ECOSYSTEM-BASE D A DA P TATION 305
systems can be more cost-effective and per- stream large woody debris, riparian forests
manent than treatment facilities (Jones et maintain thermal refugia for temperature-
al. 2012). Natural ecosystem engineers like sensitive species, such as spawning salmon
beavers (Castor canadensis) have been used to (Palmer et al. 2009). Although conservation
increase water retention and hydrologic and restoration of these natural coastal and
stability through their creation of wetland inland buffering systems can require mul-
complexes where wetlands would other- tiscale and cross-sector coordination for
wise not exist (Dittbrenner et al. 2018). Al- successful maintenance, the co-benefits and
though these EbA strategies are promising, cost savings compared to dams, levees, and
realization of large-scale adaptation goals other shoreline defense approaches can be
will require watershed assessments, long- considerable (Jones et al. 2012).
term planning, synchronization of multiple As described above, changes in precipi-
cross-watershed entities, and close coor- tation regimes may necessitate greater wa-
dination with local populations to ensure ter storage capacity and filtration—services
stakeholder buy-in. that forest vegetation and soils afford. Al-
ready many municipalities have realized
significant cost savings from investing in
M A I N TA I N I NG T H E BE N E F ITS T H AT the forested watershed conservation (e.g.,
F OR E STS PROV I DE the Catskill-Delaware watershed north of
New York City) instead of water purification
Humans derive many benefits from forests, infrastructure, which requires initial capi-
from local to global scales. For example, tal and continued maintenance. Elsewhere,
one-third of the world’s largest cities obtain tropical montane cloud forests are prime
a significant proportion of their drinking candidates for continued conservation in a
water directly from protected forests (Dud- changing climate, as they play an important
ley and Stolten 2003). Conserving the bio- role in water supplies: water vapor con-
diversity and processes of forest ecosystems denses on foliage and flows into streams,
and restoring the integrity and resilience significantly augmenting water availability
of degraded ones are primary strategies from rainfall in drier, low-elevation areas
for leveraging forests for climate change (Ellison et al. 2017). By contrast, deforesta-
adaptation. tion throughout Amazonia increases runoff
Restoration of coastal forests can mini- and water discharge at local scales and—as
mize inland flooding and coastal storm- demonstrated with simulations—affects
surge events, which are projected to the water balance, hydrology, and surface
become more frequent and greater in mag- temperatures across the entire Amazon Ba-
nitude (IPCC 2014). From riparian forests sin and likely globally (Foley et al. 2007).
to coastal mangroves, vegetation structure Along with deforestation and degrada-
physically slows water flow, attenuates wave tion, a major threat to ecosystem services
and tidal energy, and stores water through from forests is forest fires. From Australia
plant uptake, thus minimizing the threat to North America, synergistic effects of
of flooding. Indeed, fewer lives were lost climate change–related drivers are fueling
in coastal communities with healthy man- more catastrophic forest fires. For example,
grove forests than those without during earlier snowmelt and drought conditions
the 2004 Indian Ocean tsunami (Das and in the western United States are interact-
Vincent 2009). Mangroves additionally cap- ing with increasingly widespread mountain
ture nutrient-rich sediments in their root pine beetle (Dendroctonus ponderosae) outbreaks
structures and maintain important habitat as the species’ range expands (Loehman et
for birds, fish, and other marine species. al. 2016). Combined with a management
Similarly, through shading and inputs of in- legacy of fire suppression and selective
306 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
landscape-scale management pay?” Ecological Economics T. Bilir, M. Chatterjee, K. Ebi, Y. Estrada, R. Genova, B.
99: 53–62. Girma, E. Kissel, A. Levy, S. MacCracken, P. Mastran-
Das, S., and J. R. Vincent. 2009. “Mangroves protected drea, and L. White. Cambridge University Press.
villages and reduced death toll during Indian super Jones, H. P., D. G. Hole, and E. S. Zavaleta. 2012.
cyclone.” Proceedings of the National Academy of Sciences 106: “Harnessing nature to help people adapt to climate
7357–7360. change.” Nature Climate Change 2: 504–509.
de Vries, F. T., E. Thébault, M. Liiri, K. Birkhofer, M. A. Klein, A. M., B. E. Vaissière, J. H. Cane, I. Steffan-Dewen-
Tsiafouli, L. Bjørnlund, H. B. Jørgensen, et al. 2013. ter, S. A. Cunningham, C. Kremen, and T. Tscharntke.
“Soil food web properties explain ecosystem services 2007. “Importance of pollinators in changing land-
across European land use systems.” Proceedings of the Na- scapes for world crops.” Proceedings of the Royal Society B:
tional Academy of Sciences 110: 14296–14301. Biological Sciences 274: 303–313.
Dittbrenner, B. J., M. M. Pollock, J. W. Schilling, J. D. Li, D., E. Bou-Zeid, and M. Oppenheimer. 2014. “The
Olden, J. J. Lawler, and C. E. Torgersen. 2018. “Mod- effectiveness of cool and green roofs as urban heat is-
eling intrinsic potential for beaver (Castor canadensis) land mitigation strategies.” Environmental Research Letters
habitat to inform restoration and climate change ad- 9 (5): 055002.
aptation.” PLOS One 13 (2): e0192538. Loehman, R. A., R. E. Keane, L. M. Holsinger, and Z.
Dudley, N., and S. Stolton. 2003. Running Pure: The Impor- Wu. 2017. “Interactions of landscape disturbances and
tance of Forest Protected Areas to Drinking Water. World Bank climate change dictate ecological pattern and process:
and World Wildlife Fund Alliance for Forest Conser- Spatial modeling of wildfire, insect, and disease dy-
vation and Sustainable Use. namics under future climates.” Landscape Ecology 32 (7):
Foley, J. A., G. P. Asner, M. H. Costa, M. T. Coe, R. De- 1447–1459.
Fries, H. K. Gibbs, E. A. Howard, S. Olson, J. Patz, and Loh, W. 2011. “Classification and regression trees.” Wiley
N. Ramankutty. 2007. “Amazonia revealed: Forest Interdisciplinary Reviews: Data Mining and Knowledge Discovery 1:
degradation and loss of ecosystem goods and services 14–23.
in the Amazon Basin.” Frontiers in Ecology and the Environ- Mentens, J., D. Raes, and M. Hermy. 2006. “Green roofs
ment 5: 25–32. as a tool for solving the rainwater runoff problem in
Gerten, D., W. Lucht, S. Ostberg, J. Heinke, M. Kowarsch, the urbanized 21st century?” Landscape and Urban Planning
H. Kreft, Z. W. Kundzewicz, J. Rastgooy, R. Warren, 77: 217–226.
and H. J. Schellnhuber. 2013. “Asynchronous expo- Millennium Ecosystem Assessment. 2003. Ecosystems and
sure to global warming: Freshwater resources and Human Well-Being: A Framework for Assessment. Island Press.
terrestrial ecosystems.” Environmental Research Letters 8: Morris, R. L., T. M. Konlechner, M. Ghisalberti, and S.
34032. E. Swearer. 2018. “From grey to green: Efficacy of
Gill, S. E., J. F. Handley, A. R. Ennos, and S. Pauleit. 2007. eco-engineering solutions for nature-based coastal
“Adapting cities for climate change: The role of the defense.” Global Change Biology 24 (5): 1827–1842.
green infrastructure.” Built Environment 33: 115–133. Nye, J. A., J. S. Link, J. A. Hare, and W. J. Overholtz.
Gosling, S. N., and N. W. Arnell. 2016. “A global assess- 2009. “Changing spatial distribution of fish stocks in
ment of the impact of climate change on water scar- relation to climate and population size on the North-
city.” Climatic Change 134 (3): 371–385. east United States continental shelf.” Marine Ecology Prog-
Hannah, L., M. Steele, E. Fung, P. Imbach, L. Flint, and A. ress Series 393: 111–129.
Flint. 2017. “Climate change influences on pollinator, Paap, T., T. I. Burgess, and M. J. Wingfield. 2017. “Ur-
forest, and farm interactions across a climate gradi- ban trees: Bridge-heads for forest pest invasions and
ent.” Climatic Change 141 (1): 63–75. sentinels for early detection.” Biological Invasions 19 (12):
Herrera, D., A. Ellis, B. Fisher, C. D. Golden, K. Johnson, 3515–3526.
M. Mulligan, A. Pfaff, T. Treuer, and T. H. Ricketts. Palmer, M. A., D. P. Lettenmaier, N. L. Poff, S. L. Postel,
2017. “Upstream watershed condition predicts rural B. Richter, and R. Warner. 2009. “Climate change and
children’s health across 35 developing countries.” Na- river ecosystems: Protection and adaptation options.”
ture Communications 8: 811. Environmental Management 44: 1053–1068.
Horner, R. R., H. Lim, and S. J. Burges. 2004. Hydrologi- Pan, Y., R. A. Birdsey, J. Fang, R. Houghton, P. E. Kauppi,
cal Monitoring of the Seattle Ultra-Urban Stormwater Management W. A. Kurz, O. L. Phillips, A. Shvidenko, S. L. Lewis,
Projects: Summary of the 2000–2003 Water Years. Water Re- J. G. Canadell, P. Ciais, R. B. Jackson, S. W. Pacala, A.
sources Series Technical Report No. 181. Department D. McGuire, S. Piao, A. Rautiainen, S. Sitch, and D.
of Civil and Environmental Engineering, University of Hayes. 2011. “A large and persistent carbon sink in the
Washington. world’s forests.” Science 333: 988–993.
Intergovernmental Panel on Climate Change (IPCC). Pappalardo, V., D. La Rosa, A. Campisano, and P. La
2014. Climate Change 2014: Impacts, Adaptation, and Vulner- Greca. 2017. “The potential of green infrastructure
ability. Contribution of Working Group II to the Fifth Assessment application in urban runoff control for land use plan-
Report of the Intergovernmental Panel on Climate Change, ed. C. ning: A preliminary evaluation from a southern Italy
Field, V. Barros, D. Dokken, K. Mach, M. Mastrandrea, case study.” Ecosystem Services 26: 345–354.
ECOSYSTEM-BASE D A DA P TATION 309
Parker, G. 2013. Global Crisis: War, Climate Change and Catastro- Tengö, M., and K. Belfrage. 2004. “Local management
phe in the Seventeenth Century. Yale University Press. practices for dealing with change and uncertainty: A
Parmova, E., B. Locatelli, H. Djoudi, and O. A. Somo- cross-scale comparison of cases in Sweden and Tanza-
rin. 2012. “Forests and trees for social adaptation to nia.” Ecology and Society 9: 4.
climate variability and change.” Wiley Interdisciplinary Thompson, M. S., S. J. Brooks, C. D. Sayer, G. Wood-
Reviews: Climate Change 3: 581–596. ward, J. C. Axmacher, D. M. Perkins, and C. Gray.
Pendleton, L., D. C. Donato, B. C. Murray, S. Crooks, W. 2018. “Large woody debris ‘rewilding’ rapidly re-
A. Jenkins, S. Sifleet, C. Craft, et al. 2012. “Estimating stores biodiversity in riverine food webs.” Journal of
global ‘blue carbon’ emissions from conversion and Applied Ecology 55 (2): 895–904.
degradation of vegetated coastal ecosystems.” PLOS One Trabucco, A., R. J. Zomer, D. A. Bossio, O. van Straaten,
7 (9): e43542. and L. V. Verchot. 2008. “Climate change mitigation
Piao, S., P. Ciais, Y. Huang, Z. Shen, S. Peng, J. Li, L. through afforestation/reforestation: A global analysis
Zhou, et al. 2010. “The impacts of climate change on of hydrologic impacts with four case studies.” Agricul-
water resources and agriculture in China.” Nature 467: ture Ecosystems and Environment 126: 81–97.
43–51. UN Department of Economic and Social Affairs, Popula-
Pinsky, M. L., and N. J. Mantua. 2014. “Emerging adapta- tion Division. 2014. World Urbanization Prospects: The 2014
tion approaches for climate-ready fisheries manage- Revision (ST/ESA/SER.A/352). United Nations.
ment.” Oceanography 27 (4): 146–159. Varian, H. R. 2014. “Big data: New tricks for economet-
Rosenzweig, C., J. Elliott, D. Deryng, A. C. Ruane, C. rics.” Journal of Economic Perspectives 28: 3–27.
Müller, A. Arneth, K. J. Boote, C. Folberth, M. Glot- Vittor, A. Y., R. H. Gilman, J. Tielsch, G. Glass, T. Shields,
ter, N. Khabarov, et al. 2014. “Assessing agricultural W. S. Lozano, V. Pinedo-Cancino, and J. A. Patz. 2006.
risks of climate change in the 21st century in a global “The effect of deforestation on the human-biting rate
gridded crop model intercomparison.” Proceedings of the of Anopheles darlingi, the primary vector of falciparum
National Academy of Sciences 111 (9): 3268–3273. malaria in the Peruvian Amazon.” American Journal of
Ruckelshaus, M., P. Kareiva, and L. Crowder. 2014. “The Tropical Medicine and Hygiene 1 (74): 3–11.
future of marine conservation.” In Marine Community Weatherdon, L. V., A. K. Magnan, A. D. Rogers, U. R. Su-
Ecology and Conservation, ed. M. Bertness and B. Silliman. maila, and W. W. Cheung. 2016. “Observed and pro-
Sinauer Associates. jected impacts of climate change on marine fisheries,
Shepard, C. C., C. M. Crain, and M. W. Beck. 2011. “The aquaculture, coastal tourism, and human health: An
protective role of coastal marshes: A systematic review update.” Frontiers in Marine Science 3: 48.
and meta-analysis.” PLOS One 6 (11): e27374. Westerling, A. L., H. G. Hidalgo, D. R. Cayan, and T.
Squeo, F. A., B. G. Warner, R. Aravena, and D. Espinoza. W. Swetnam. 2006. “Warming and earlier spring in-
2006. “Bofedales: High altitude peatlands of the cen- crease western US forest wildfire activity.” Science 313
tral Andes.” Revista chilena de historia natural 79: 245–255. (5789): 940–943.
Tallis, H., P. Kareiva, M. Marvier, and A. Chang. 2008. Wheeler, T., and J. von Braun. 2013. “Climate change
“An ecosystem services framework to support both impacts on global food security.” Science 341: 508–513.
practical conservation and economic development.”
Proceedings of the National Academy of Sciences 105 (28):
9457–9464.
CHAPTER TWENTY-FOUR I N T RODUCTION
310
CLIM ATE CH A NGE MITIGATION USING TER R E ST R IA L ECOSYSTEMS 311
of GHGs emitted depends on the original developing countries, and some funders are
and the “new” use of the land as well as making resources available to support this.
on longer-term land management. Some Such incentives would cover five activities,
ecosystems have particularly high potential collectively referred to as REDD+: (1) re-
to contribute to climate change mitigation. ducing emissions from deforestation and
These include forests, wetlands, coastal (2) from forest degradation, plus (3) con-
ecosystems, and drylands. Management of servation of forest carbon stocks, (4) sus-
agricultural ecosystems offers great poten- tainable management of forest, and (5) en-
tial for mitigation. hancement of forest carbon stocks.
REDD+ has the potential to contribute
substantially to climate change mitigation
Mitigation Measures Targeting
and thus to achievement of the goals in
Forest: REDD+
the Paris Agreement and to help meet the
The largest emissions from land-use change objectives of the New York Declaration on
result from conversion of mature tropi- Forests (halving natural forest loss by 2020,
cal forest to agricultural land to produce and striving to end it by 2030), endorsed by
both food and bioenergy crops, and from world leaders at the UN Secretary-General’s
the burning and drainage of tropical peat- Climate Summit in 2014. Although the five
swamp forests and soils (van der Werf et REDD+ activities are not further defined,
al. 2009). In 2015, about 30.6 percent of they are widely interpreted as applying to
the world’s terrestrial surface (just under (1) forests under direct deforestation pres-
4 billion ha) was covered by forests, stor- sure, (2) forests under direct degradation
ing about 296 gigatons of biomass carbon pressure, (3) forests not under direct pres-
(Food and Agriculture Organization of the sure but acting as important carbon sinks,
United Nations [FAO] 2016). The global rate (4) forests used for their resources, and
of net annual forest loss more than halved (5) areas that provide opportunities for for-
from 7.2 million ha in the 1990s, and the est restoration (including degraded forest),
current net annual decrease in forest area reforestation, or afforestation. REDD+ thus
is still very large: about 3.3 million ha, av- addresses the different sources of GHG emis-
eraged over the period 2010–2015 (FAO sions, and opportunities for sequestration,
2016). In addition, roughly 20 percent of the in the forest sector and from deforestation.
world’s remaining forests are degraded (Po- Ecosystems and biodiversity play a role in
tapov, Laestadius, and Minnemeyer 2011), each of the REDD+ activities. Intact natural
often through selective timber harvesting forests are more resilient to climate change
and damage caused to surrounding vegeta- than degraded, often less biodiverse forests
tion (Pearson, Brown, and Casarim 2014). (Miles et al. 2010). Conserving these areas
Over the decade 2007–2016, CO2 emissions thus helps maintain not only important for-
from land use change (including deforesta- est carbon reservoirs but also the ecosys-
tion, afforestation, logging and associated tem’s natural resilience and biodiversity.
forest degradation, shifting cultivation, and
regrowth of forests following wood harvest
or abandonment of agriculture) accounted
W IDE R E COSYST E M-BASE D
for about 12 percent of the world’s total
MITIGATION
carbon emissions (Le Quéré et al. 2018; see
also Figure 24.1).
Management of Non-Forest Ecosystems
The UN Framework Convention on Cli-
mate Change (UNFCCC) has agreed on key Ecosystems such as peatlands, littoral zones,
components to provide financial incentives and dryland ecosystems have great potential
to reduce forest-related GHG emissions in to contribute to climate change mitigation.
312 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
Figure 24.1. Forest cover loss in Paraguay between 2000 much larger than that in its aboveground
and 2012 (map) and emissions/removals of CO2 equiva- biomass (Scharlemann et al. 2014). Peat
lents associated with net forest change over the same pe-
riod of time (graph). Sources: Walcott et al. 2015 (using data
soils and coastal sediments are particularly
from PNC-ONU REDD+ Paraguay 2011, The Global Land rich in carbon given their density, depth,
Cover Facility 2006) by permission of UNEP-WCMC; and anaerobic conditions, so CO2 emissions
Hansen et al. 2013 (map), and FAOSTAT 2014 (graph). from soil and sediment are of particular
concern in these ecosystems. The world’s
Whereas their aboveground biomass may be wetlands are also the largest natural source
lower than that of forests, the carbon stored of methane (Turetsky et al. 2014), a GHG
in their belowground biomass (roots, soils, that has a much stronger global warming
and sediment) can be substantial. Overall, potential than CO2, at least for the first six
the carbon stored in the world’s soils is or seven decades after its emission.
CLIM ATE CH A NGE MITIGATION USING TER R E ST R IA L ECOSYSTEMS 313
meet demand for wood pellets that are re- in Southeast Asia and Africa. Models indi-
placing coal for electric power generation cate that between 1999 and 2015 half the
has caused significant deforestation. population of the Bornean orangutan (Pongo
In some tropical areas of the world, such pygmaeus) was affected by logging, deforesta-
as Brazil and Indonesia, biofuel expansion tion, or industrialized plantations (Voigt et
is a major emerging threat to biodiversity al. 2018). More than half the oil palm con-
(Stromberg et al. 2010). Biodiversity im- cessions for which data were available in
pacts are feedstock specific. For example, Africa overlap with great ape distributions,
oil palm (mainly Elaeis guineensis but also Elaeis and about 42 percent of the geographic dis-
oleifera and Attalea maripa) cultivation often tribution of great apes overlaps with areas
entails conversion of virgin forest, whereas suitable for oil palm production (Wich et
cassava (Manihot esculenta) and Jatropha (Jatro- al. 2014). Studies have found that orang-
pha curcas) can be cultivated on grassland or utans disperse into mature oil palm planta-
land already used for agriculture. A com- tions only where natural forest patches are
parison of studies from Indonesia suggests nearby (Ancrenaz et al. 2014). Large-scale
that total vertebrate species richness in oil conversion for monoculture oil palm plan-
palm plantations is less than half that of tations both in Southeast Asia and Africa,
tropical forests, with great differences in which is likely to be increased by demand
community composition (Figure 24.2). for biofuel, could severely affect the distri-
Some of the main components of tropical bution and the conservation status of great
forests, such as forest trees, lianas, epiphytic apes.
orchids, and indigenous palms, are absent, Although difficult to trace, it can be as-
so the flora in plantations is impoverished sumed that indirect land-use change from
(Danielsen et al. 2009). biofuel production occurs at substantial ex-
There is increasing concern about the tent. For example, in Brazil, area for both
impact of oil palm cultivation on great apes biofuel production and cattle ranches has
expanded over recent years. In the Amazon
region more than 90 percent of soybean
Figure 24.2. Impact on fauna of replacing forest with oil
palms. Mean number of invertebrate species (patterned plantations planted after 2006 replaced cat-
bar) and mammal, bird, and reptile species (solid bar) tle ranches (Lapola et al. 2010), which in
recorded in oil palm (including new colonists of oil turn were displaced into areas of forest and
palm) as a proportion of those recorded in forest. Meta- cerrado (tropical savanna ecoregion of Bra-
analysis sample sizes are provided in parentheses; error
zil). Land-use change scenarios for Europe
bars are 95% confidence intervals. (Redrawn from Dan-
ielsen et al. 2009.) for the period 2000–2030 have shown that
CLIM ATE CH A NGE MITIGATION USING TER R E ST R IA L ECOSYSTEMS 317
indirect effects of the European Union’s ergy, and wind, is continuously increasing,
biofuel directive on land use and biodiver- spurred by government policies and subsi-
sity in Europe are much larger than the di- dies, the decreasing cost of renewable en-
rect effects (Hellmann and Verburg 2010). ergy technologies, and other factors (IPCC
Ways to reduce the impact of biofuel 2011). Such options presumably can have
production on ecosystems and biodiversity large impacts on ecosystems and biodiver-
are being explored. For example, growing sity, although knowledge of these impacts is
low-input, high-diversity mixtures of na- limited for some options.
tive grassland perennials on degraded ag- In general, impacts fall into five main
ricultural land could provide more usable groups: (1) direct environmental foot-
energy, greater GHG reductions, and less print of the hard structures (including
agrichemical pollution than conventional habitat alteration, fragmentation, isolation
biofuel crop cultivation without displacing of protected areas, and loss); (2) noise of
food production or harming biodiversity construction, decommissioning, and dis-
(Tilman, Hill, and Lehman 2006; Verdade, turbance by human presence; (3) collision
Piña, and Rosalino 2015). Agricultural in- with operating infrastructure; (4) pollution
tensification has been suggested as another and disturbance from operating infrastruc-
alternative to continued growth in land use ture; and (5) provision of new substrate for
by the biofuels sector. establishment of new communities, includ-
The known risks to biodiversity of bio- ing invasive species (amended from Bertzky
fuel production have triggered the devel- et al. 2010; Hoetker, Thomsen, and Jeromin
opment of safeguards and standards for 2006).
this sector. The EU Renewable Energy Di- Wind farms can harm biodiversity
rective (RED), for example, includes com- through enhanced collision risks for birds
prehensive sustainability criteria that aim and bats as well as disturbance leading to
to prevent the conversion of areas of high displacement, for example, of waders nest-
biodiversity and high carbon stock for the ing on open ground. However, impacts are
production of biofuel feedstocks. Biofuels species specific, and some species seem to
must comply with these criteria if they are benefit from wind farms, possibly because
to be counted toward the 10 percent renew- the agricultural use of land in the immedi-
able energy target for the transport sector ate vicinity often is suspended. Limitations
or to receive state financial support. Given to fishing in the vicinity of offshore wind
the limited contribution of the EU to global farms can have similar positive impacts
biofuel consumption, only a small share of (Bergström et al. 2014). The species affected
global biofuel production is directly subject depend on the location, such as near to
to these targets and standards. Nonethe- shore, on mountain ridges, or close to for-
less, concerns about biofuels impacts have est (Hoetker, Thomsen, and Jeromin 2006).
prompted development of relevant volun- Appropriate siting and the use of blinking
tary safeguards and standards initiatives lights have reduced collisions of birds with
that may be more widely applied, including wind towers.
the Roundtable on Sustainable Biomaterials, Hydropower is a major renewable energy
the Roundtable on Sustainable Palm Oil, source for a number of countries world-
and the Better Sugarcane Initiative. wide. Environmental impacts of dam con-
struction include the destruction of habitat
and altered water-flow patterns, sedimen-
Other Renewable Energy Options
tation, discharge change, and heavy metal
The use of non-ecosystem-based renew- pollution. It is anticipated that the hundreds
able energy options, such as direct solar en- of new dams planned for construction in
ergy, geothermal, hydropower, ocean en- the major tropical rivers will disrupt fish
318 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
(3): 34012. https://doi.org/10.1088/1748-9326/9/3/ Harvey, Celia A., Barney Dickson, and Cyril Kormos.
034012. 2010. “Opportunities for achieving biodiversity con-
Bertzky, Monika, Barney Dickson, Russell Galt, Eleanor servation through REDD.” Conservation Letters 3 (1): 53–
Glen, Mike Harley, Nikki Hodgson, Igor Lysenko, et 61. https://doi.org/10.1111/j.1755-263X.2009.00086.x.
al. 2010. Impacts of Climate Change and Selected Renewable En- Hellmann, Fritz, and Peter H. Verburg. 2010. “Impact
ergy Infrastructures on EU Biodiversity and the Natura 2000 Net- assessment of the European biofuel directive on land
work: Summary Report. European Commission and Inter- use and biodiversity.” Journal of Environmental Manage-
national Union for Conservation of Nature. ment 91 (6): 1389–1396. https://doi.org/10.1016/j
Brack, Duncan. 2017. The Impacts of the Demand for Woody Bio- .jenvman.2010.02.022.
mass for Power and Heat on Climate and Forests. Royal Institute Hernandez, Rebecca R., Shane B. Easter, Michelle L.
of International Affairs, Chatham House. Murphy-Mariscal, Fernando T. Maestre, M. Tavassoli,
Chen, Shaoqing, Bin Chen, and Brian D. Fath. 2015. “As- Edith B. Allen, Cameron W. Barrows, et al. 2014. “En-
sessing the cumulative environmental impact of hydro- vironmental impacts of utility-scale solar energy.” Re-
power construction on river systems based on energy newable and Sustainable Energy Reviews 29: 766–779. https://
network model.” Renewable and Sustainable Energy Reviews 42: doi.org/10.1016/j.rser.2013.08.041.
78–92. https://doi.org/10.1016/j.rser.2014.10.017. Hoetker, Hermann, Kai-Michael Thomsen, and Heike
Cusack, Daniela F., Jonn Axsen, Rachael Shwom, Lau- Jeromin. 2006. Impacts on Biodiversity of Exploitation of Renew-
ren Hartzell-Nichols, Sam White, and Katherine R. able Energy Sources: The Example of Birds and Bats—Facts, Gaps in
M. Mackey. 2014. “An interdisciplinary assessment of Knowledge, Demands for Further Research, and Ornithological Guide-
climate engineering strategies.” Frontiers in Ecology and the lines for the Development of Renewable Energy Exploitation. Mi-
Environment 12 (5): 280–287. chael-Otto-Institut of Naturschutzbund Deutschland.
Danielsen, Finn, Hendrien Beukema, Neil D. Burgess, IPCC. 2011. Special Report on Renewable Energy Sources and Climate
Faizal Parish, Carsten A. Brühl, Paul F. Donald, Daniel Change Mitigation. Edited by Ottmar Edenhofer, Rafael
Murdiyarso, et al. 2009. “Biofuel plantations on for- Pichs-Madruga, Youba Sokona, Kristin Seyboth, Pat-
ested lands: Double jeopardy for biodiversity and cli- rick Matschoss, Susanne Kadner, Timm Zwickel, et al.
mate.” Conservation Biology 23 (2): 348–358. https://doi Cambridge University Press.
.org/10.1111/j.1523-1739.2008.01096.x. Joosten, Hans. 2015a. “Current soil carbon loss and land
FAO. 2013. FAO Statistical Yearbook 2013. Food and Agricul- degradation globally—Where are the hotspots and
ture Organization of the United Nations. why there?” In Soil Carbon: Science, Management and Policy
FAO. 2016. Global Forest Resources Assessment 2015. FAO Forestry. for Multiple Benefits, ed. Steven A. Banwart, Elke Noelle-
Food and Agriculture Organization of the United Na- meyer, and Elinor Milne, 224–234. CAB International.
tions. https://doi.org/10.1002/2014GB005021. Joosten, Hans. 2015b. Peatlands, Climate Change Mitigation
FAOSTAT. 2014. “Emissions/removals of Carbon Diox- and Biodiversity Conservation: An Issue Brief on the Importance of
ide Associated with Forest and Net Forest Conversion. Peatlands for Carbon and Biodiversity Conservation and the Role of
Paraguay.” Food and Agriculture Organization of the Drained Peatlands as Greenhouse Gas Emission Hotspots. Nordic
United Nations Statistics Division. http://faostat3.fao Council of Ministers. http://norden.diva-portal.org/
.org/browse/G2/GF/E. smash/get/diva2:806688/FULLTEXT01.pdf.
Gardner, Toby A., Neil D. Burgess, Naikoa Aguilar-Amu- Lamb, David, Peter D. Erskine, and John A. Parrotta.
chastegui, Jos Barlow, Erika Berenguer, Tom Clem- 2005. “Restoration of degraded tropical forest land-
ents, Finn Danielsen, et al. 2012. “A framework for in- scapes.” Science 310 (5754): 1628–1632. https://doi
tegrating biodiversity concerns into national REDD+ .org/10.1126/science.1111773.
programmes.” Biological Conservation 154: 61–71. https:// Lambin, Eric F., and Patrick Meyfroidt. 2011. “Global
doi.org/10.1016/j.biocon.2011.11.018. land use change, economic globalization, and the
Global Land Cover Facility. 2006. Forest Cover Change looming land scarcity.” Proceedings of the National Acad-
in Paraguay, Version 1.0. University of Maryland In- emy of Sciences 108 (9): 3465–3472. https://doi.org/10
stitute for Advanced Computer Studies. College Park, .1073/pnas.1100480108.
MD, 1990–2000. Lapola, David M., Ruediger Schaldach, Joseph Alcamo,
Griscom, Bronson W., Justin Adams, Peter W. Ellis, Alberte Bondeau, Jennifer Koch, Christina Koelking,
Richard A. Houghton, Guy Lomax, Daniela A. Miteva, and Joerg A. Priess. 2010. “Indirect land-use changes
William H. Schlesinger, et al. 2017. “Natural climate can overcome carbon savings from biofuels in Brazil.”
solutions.” Proceedings of the National Academy of Sciences Proceedings of the National Academy of Sciences 107 (8): 3388–
114 (44): 11645–11650. https://doi.org/10.1073/pnas 3393. https://doi.org/10.1073/pnas.0907318107.
.1710465114. Le Quéré, Corinne, Robbie M. Andrew, Pierre Friedling-
Hansen, M. C., P. V. Potapov, R. Moore, M. Hancher, S. stein, Stephen Sitch, Julia Pongratz, Dan Zhun Zhu
A. Turubanova, A. Tyukavina, D. Thau, et al. 2013. Korsbakken, Glen P. Peters, Josep G. Canadell, Robert
“High-resolution global maps of 21st-century forest B. Jackson, Thomas A. Boden, et al. 2018. “The global
cover change.” Science 342 (6160): 850–853. http://sci carbon budget 2017.” Earth System Science Data Discussions
ence.sciencemag.org/content/342/6160/850.abstract. 405–448. https://doi.org/10.5194/essdd-2017-123.
CLIM ATE CH A NGE MITIGATION USING TER R E ST R IA L ECOSYSTEMS 321
Mcleod, Elizabeth, Gail L. Chmura, Steven Bouillon, Safriel, Uriel, Zafar Adeel, David Niemeijer, Juan Puig-
Rodney Salm, Mats Björk, Carlos M. Duarte, Cathe- defabregas, Robin White, Rattan Lal, Mark Winslow,
rine E. Lovelock, William H. Schlesinger, and Brian et al. 2005. “Dryland systems.” In Ecosystems and Human
R. Silliman. 2011. “A blueprint for blue carbon: To- Well Being: Current State and Trends, ed. Rashid Hassan,
ward an improved understanding of the role of veg- Robert Scholes, and Neville J. Ash, 623–662. Island
etated coastal habitats in sequestering CO2.” Frontiers in Press.
Ecology and the Environment 9 (10): 552–560. https://doi Scharlemann, Jörn P. W., Edmund V. J. Tanner, Roland
.org/10.1890/110004. Hiederer, and Valerie Kapos. 2014. “Global soil car-
Meyfroidt, Patrick, and Eric F. Lambin. 2009. “Forest bon: Understanding and managing the largest ter-
transition in Vietnam and displacement of deforesta- restrial carbon pool.” Carbon Management 5 (1): 81–91.
tion abroad.” Proceedings of the National Academy of Sciences https://doi.org/10.4155/cmt.13.77.
106 (38): 16139–16144. https://doi.org/10.1073/pnas Smith, Pete. 2012. “Agricultural greenhouse gas miti-
.0904942106. gation potential globally, in Europe and in the
Miles, Lera, Emily Dunning, Nathalie Doswald, and UK: What have we learnt in the last 20 years?”
Matea Osti. 2010. “A safer bet for REDD+: Review of Global Change Biology 18 (1): 35–43. https://doi
the evidence on the relationship between biodiversity .org/10.1111/j.1365-2486.2011.02517.x.
and the resilience of forest carbon stocks.” Working Stromberg, Per M., Alexandros Gasparatos, Janice S. H.
Paper v2, Multiple Benefits Series 10. UNEP World Lee, John Garcia-Ulloa, Lian Pin Koh, and Kazuhiko
Conservation Monitoring Centre. Takeuchi. 2010. Impacts of Liquid Biofuels on Ecosystem Services
Miles, Lera, and Valerie Kapos. 2008. “Reducing green- and Biodiversity. United Nations University Institute of
house gas emissions from deforestation and forest Advanced Studies.
degradation: Global land-use implications.” Science 320 Tilman, David, Jason Hill, and Clarence Lehman. 2006.
(5882): 1454–1455. https://doi.org/10.1126/science “Carbon-negative biofuels from low-input high-di-
.1155358. versity grassland biomass.” Science 314 (5805): 1598–
Millennium Ecosystem Assessment. 2005. Ecosystems and 1600. https://doi.org/10.1126/science.1133306.
Human Well-Being: Desertification Synthesis. World Resources Trisos, Christopher H., Giuseppe Amatulli, Jessica Gure-
Institute. vitch, Alan Robock, Lili Xia, and Brian Zambri. 2018.
Pearson, Timothy R. H., Sandra Brown, and Felipe “Potentially dangerous consequences for biodiversity
M. Casarim. 2014. “Carbon emissions from tropi- of solar geoengineering implementation and termina-
cal forest degradation caused by logging.” Envi- tion.” Nature Ecology & Evolution 2 (3): 475–482. https://
ronmental Research Letters 9 (3): 34017. https://doi doi.org/10.1038/s41559-017-0431-0.
.org/10.1088/1748-9326/9/3/034017. Turetsky, Merritt R., Agnieszka Kotowska, Jill Bubier,
Peh, Kelvin S. H., Andrew Balmford, Rob H. Field, An- Nancy B. Dise, Patrick Crill, Ed R. C. Hornibrook,
thony Lamb, Jennifer C. Birch, Richard B. Bradbury, Kari Minkkinen, et al. 2014. “A synthesis of methane
Claire Brown, et al. 2014. “Benefits and costs of eco- emissions from 71 northern, temperate, and subtropi-
logical restoration: Rapid assessment of changing eco- cal wetlands.” Global Change Biology 20 (7): 2183–2197.
system service values at a U.K. Wetland.” Ecology and https://doi.org/10.1111/gcb.12580.
Evolution 4 (20): 3875–3886. https://doi.org/10.1002/ van der Werf, Guido R., Douglas C. Morton, Ruth S.
ece3.1248. DeFries, Jos G. J. Olivier, Prasad S. Kasibhatla, Robert
PNC ONU-REDD+ Paraguay. 2011. Mapa de bosque/no B. Jackson, G. Jim Collatz, and James T. Randerson.
bosque. Inventario Forestal Nacional. Asunción, Para- 2009. “CO2 emissions from forest loss.” Nature Geoscience
guay: PNC ONU-REDD+. 2: 737–738.
Potapov, Peter, Lars Laestadius, and Susan Minnemeyer. Verdade, Luciano M., Carlos I. Piña, and Luís Miguel Ro-
2011. Global Map of Forest Landscape Restoration Opportunities. salino. 2015. “Biofuels and biodiversity: Challenges
World Resources Institute. http://www.wri.org/forest and opportunities.” Environmental Development 15: 64–78.
-restoration-atlas. https://doi.org/10.1016/j.envdev.2015.05.003.
Ren, Yanjiao, Yihe Lü, Bojie Fu, and Kun Zhang. 2017. Voigt, Maria, Serge A. Wich, Marc Ancrenaz, Erik Mei-
“Biodiversity and ecosystem functional enhancement jaard, Nicola Abram, Graham L. Banes, Gail Camp-
by forest restoration: A meta-analysis in China.” Land bell-Smith, et al. 2018. “Global demand for natural
Degradation & Development 28 (7): 2062–2073. https://doi resources eliminated more than 100,000 Bornean
.org/10.1002/ldr.2728. orangutans.” Current Biology 28 (5): 761–769. https://doi
Russell, Lynn M., Philip J. Rasch, Georgina M. Mace, .org/10.1016/J.CUB.2018.01.053.
Robert B. Jackson, John Shepherd, Peter Liss, Margaret Walcott, Judith, Julia Thorley, Valerie Kapos, Lera Miles,
Leinen, et al. 2012. “Ecosystem impacts of geoengi- Stephen Woroniecki, and Ralph Blaney. 2015. Mapping
neering: A review for developing a science plan.” AM- Multiple Benefits of REDD+ in Paraguay: Using Spatial Information
BIO 41 (4): 350–369. https://doi.org/10.1007/s13280 to Support Land-Use Planning. UNEP-WCMC. http://www
-012-0258-5. .un-redd.org/tabid/5954/Default.aspx.
322 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
Wich, Serge A., John Garcia-Ulloa, Hjalmar S. Kühl, Tat- Witt, Matthew J., Emma V. Sheehan, Stuart Bearhop, An-
anya Humle, Janice S. H. Lee, and Lian Pin Koh. 2014. nette C. Broderick, Daniel C. Conley, Stephen P. Cot-
“Will oil palm’s homecoming spell doom for Africa’s terell, E. Crow, et al. 2012. “Assessing wave energy
great apes?” Current Biology 24 (14): 1659–1663. https:// effects on biodiversity: The wave hub experience.”
doi.org/10.1016/j.cub.2014.05.077. Philosophical Transactions Series A: Mathematical, Physical, and
Winemiller, K. O., P. B. McIntyre, L. Castello, E. Fluet- Engineering Sciences 370 (1959): 502–529. https://doi.org/
Chouinard, T. Giarrizzo, S. Nam, I. G. Baird, et al. 10.1098/rsta.2011.0265.
2016. “Balancing hydropower and biodiversity in
the Amazon, Congo, and Mekong.” Science 351 (6269):
128–129. https://doi.org/10.1126/science.aac7082.
CASE STUDY 10
The accelerated pace of both land use and climate change is closing the window of op-
portunity to design and manage landscapes for biodiversity. The paradigm of connectivity
conservation emphasizes that biodiversity and ecosystem services can be maintained by
optimizing the spatial composition and configuration of ecosystem and habitat types in the
landscape. Hundreds of habitat-network projects that implement this paradigm are being
established around the world, but the application of connectivity concepts for biodiversity
in changing landscapes remains a challenge.
323
324 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
Figure CS10.1. The six sequential and parallel steps (struc- the period 2041–2070 (type 1, decrease; 2, no change;
tured as a loop) of the general framework to identify 3, increase; gray layer over map = climatically suitable
spatial conservation priorities: 1) In this example, we se- areas); model changes in regional climate suitability. 5)
lected five focal species based on their dispersal ability— Spatial prioritization: We used the software Zonation
as a proxy for vulnerability to fragmentation—among v3.1 to produce a spatial ranking of forest conservation
birds (Seiurus aurocapilla, Strix varia), amphibians (Plethodon priorities (grayscale with darkest tones = highest prior-
cinereus), and mammals (Odocoileus virginianus, Martes ameri- ity) based on habitat quality and complementary mea-
cana). 2) Assess species-specific habitat quality and resis- sures of short- and long-range connectivity. 6) Land-
tance layers (quality shown by tone of gray). 3) Habitat use change simulations: We simulated land-use change
network: Each species network is composed of nodes over the period 2000–2050 with a model combining a
and links with higher to lower (grayscale) probabilities top-down demand for new urban and agricultural ar-
of being used. Habitat nodes are connected from edge- eas, bottom-up constraints (e.g., slope, soil quality),
to-edge via least-cost links into a stepping-stone struc- and spatial processes (e.g., diffusion). Land-use change
ture. 4) Climate suitability was derived for each species simulations involved probabilistic transitions from for-
from an ensemble forecasting approach. Models were est to agricultural lands or urban land cover. (For further
calibrated for the period 1971–2000 and projected for methodological details, see Albert et al. 2017.)
short to long range, such as well-connected was derived from connectivity and habitat-
large patches, matrix areas that channel quality maps in the years 2000, 2025, and
flow from one forest patch to another, and 2050 under land-use and climate change
strategically positioned patches that form a scenarios.
stepping-stone pathway between the Lau- Highest conservation priority is given to
rentian and Appalachian Mountains. The the series of stepping-stone patches to the
spatial ranking of conservation priorities north that promotes traversability between
CONNECTIVITY BY DESIGN 325
the mountain ranges (Figure CS10.1). Many the investment required to acquire, restore,
small forest fragments in the agriculturally and protect habitat within a network, it is
intensive zone are also of high conservation important for the network to be robust to as
value, providing connectivity within the many likely scenarios of change as possible.
St. Lawrence lowlands, particularly in the The strong stakeholder relationships we
southwestern and northeastern portions. developed with government ministries,
municipalities, nongovernmental organiza-
tions, and landowners mean that our results
CONC LUSIONS are being used in the design of Montreal’s
greenbelt project; measures for the imple-
We quantified and ranked the contribution mentation of the habitat network include
of all forest fragments to the region’s forest new protected areas, large-scale tree plant-
network. Our approach to network design ing, and conservation easements for land-
stresses the value of different dimensions owners. Ongoing work involves strength-
of habitat connectivity needed to mitigate ening the economic argument for investing
the impacts of land use and climate change in the greenbelt for the range of ecosystem
on the region’s biodiversity. Networks op- services the forest delivers to the people
timized for multiple movement objectives that inhabit the region.
can allow a diverse set of species to move
among habitat patches in the short term and
adjust their distributions in the longer term R E FE R E NCE S
as climate and land use change. We recom-
mend the use of realistic scenario-based Albert, C., B. Rayfield, M. Dumitru, and A. Gonzalez.
2017. “Applying network theory to prioritize multi-
projections of climate and land-use change
species habitat networks that are robust to climate and
so as to future-proof the structure of the land-use change.” Conservation Biology 31: 1383–1396.
habitat network being prioritized. Given
CHAPTER TWENTY-FIVE It is clear from the preceding chapters that
the biosphere and biological diversity are
Regreening the more sensitive to climate change than had
previously been imagined, and that it is im-
Emerald Planet:The portant to try to limit climate change and
impacts to a manageable level. With the
Role of Ecosystem 1.0°C increase in global temperature, there
already are some ecosystems beyond the
Restoration in point of adapting naturally (Box 25.1).
That conclusion leads quite naturally
Reducing Climate to the question of whether there are ways
to reduce the amount of impending cli-
Change mate change. There are two ways to do it:
increase the albedo of the planet so that
THOMAS E. LOVEJOY more incoming radiation is reflected back
to space, or reduce the atmospheric CO2
concentrations so that less radiant heat is
trapped to warm the planet.
Reducing albedo can include modest
changes to the built environment such as
painting roofs white as well as schemes,
often called geoengineering. An example
would be the introduction of sulfates into
the atmosphere to reflect incoming radia-
tion, mimicking the known effects from
major volcanic eruptions such as that of Mt.
Pinatubo in 1991. Those kinds of schemes
have the serious disadvantage that they ad-
dress temperature (the symptom) but not
elevated greenhouse gas levels (the cause).
Also, the moment such an intervention
ceases, planetary temperature jumps back
to what it would have been without the in-
tervention. In addition, they do nothing to
address ocean acidification.
Because most of such geoengineer-
ing schemes are planetary in scale, what-
ever downsides they have will similarly be
planetary in scale. In addition, the National
Academy of Sciences deemed “geoengi-
neering” an inappropriate term because it
is only possible to engineer systems that are
326
REGREENING THE EMERALD PLANET 327
For a long time much of the international sure that food production is not
negotiation and discussion has been threatened, and to enable economic
around not exceeding a target of 2°C of development to proceed in a sus-
warming. This was largely based on the tainable manner.”
perception that it was achievable from The impacts already observed of
an energy perspective rather than hav- coral bleaching events and pervasive
ing intrinsic merit. At the Paris COP 21, mortality of coniferous forests (be-
1.5°C, long championed by James Han- cause the balance has been tipped
sen (Hansen et al. 2013), emerged as an in favor of pine bark beetles) cannot
important topic largely at the behest of be viewed as “adapting naturally.”
indigenous peoples and small island de- In both cases what is occurring was
veloping states. not predictable with climate mod-
That 2°C is too much should have els and vegetation models because
emerged far earlier for the simple reason both come down to specific bio-
that the last time the world was two de- logical relationships: between the
grees warmer, the oceans were four to coral animals and algae species in
six meters higher (Kopp et al. 2009). That the case of the reefs, and the trees
would imply the elimination of low-lying and native beetles in the second.
nations and huge coastal areas involving It can be argued that some spe-
the considerable human populations clus- cies of corals are more resilient in
tered on them. In a sense the planet has the face of higher temperatures and
already done “the experiment,” and the that even the sensitive ones may
only questions are about rates of rise, not switch to a more resistant symbiont.
about the endpoint. Nonetheless bleaching events will
Impacts on biodiversity also point to occur with greater and greater fre-
1.5° being far preferable to 2°C (Midgley quency before such adaptations can
2018; Warren et al. 2018). As this book take place. There is the additional
reveals, the impacts of climate change on factor of increasing acidity that will
biodiversity are essentially ubiquitous. The surely play a stressful role.
actual language of the convention reads: These are effects observed at
0.9°C, and it is logical to assume
The ultimate objective of the Con- there will be many more as the
vention is to stabilize greenhouse planet warms to 1.5°C. The conclu-
gas concentrations “at a level that sion is the biology of the planet is
would prevent dangerous anthro- very sensitive to climate change.
pogenic (human induced) interfer- The more that climate change
ence with the climate system.” It can be limited, the easier it will
states that “such a level should be be to manage the consequences,
achieved within a time-frame suf- and beyond 1.5°C the biology of
ficient to allow ecosystems to adapt the planet basically will become
naturally to climate change, to en- unmanageable.
understood (National Academy of Sciences lem directly. CO2 injection into oil wells is
2015). a time-tested technique of the oil industry
To the extent that it is possible, reduc- that extends the productive lifetime of the
ing atmospheric CO2 can reduce the prob- wells. Carbon capture and storage are still
328 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
in early experimental phases. Localized quent warming, because it takes time for
iron fertilization of the ocean replicates a the CO2 to trap the corresponding amount
natural phenomenon that stimulates phy- of radiant heat. Reducing the atmospheric
toplankton blooms; what is still unclear is CO2 concentration before that heat is
whether the CO2 remains sequestered for trapped can therefore reduce the amount of
long periods of time. global warming.
Largely overlooked until recent decades Reducing the atmospheric CO2 burden by
is the substantial amount of CO2 in the at- 1 ppm is equivalent to removing 7.7 Pg of
mosphere from degradation and destruc- carbon. To lower global CO2 concentrations
tion of modern ecosystems, mostly in the to close to 350 ppm from the current level
past two centuries. Although there has cor- of 400 ppm would mean removal of 388.5
rectly been a strong effort to reduce emis- Pg of carbon (Shukla, personal communica-
sions from fossil fuel use, limiting emis- tion). That is at the same order of magni-
sions especially from deforestation has also tude as the estimated loss (450+ Pg) from
been a justified part of the global environ- ecosystem destruction and degradation
mental (carbon cycle management) agenda. The magnitude of the potential to reduce
Indeed, the gross figure for deforestation is CO2 concentrations and associated global
about 30 percent of all emissions (bigger warming focusing on tropical forests alone
than the transportation sector) (Houghton is considerable. Elimination of current de-
et al. 2015), a hidden number when the net forestation and degradation of forests and
emissions (about 10 percent) are reported, reforesting 500 million ha could together
as is usually the case. reduce annual emissions by 4.4 Pg/yr
The amount of carbon lost to the atmo- (Houghton et al. 2015). A further complica-
sphere from terrestrial ecosystems has long tion is that removing one gigaton does not
been estimated to be 200–300 Pg (1 pet- lower CO2 concentration by exactly 1 ppm
agram = 1 trillion kilograms = 1 gigaton) because of a very slow release back from the
(Houghton 2012). A more conservative es- oceans, which of course produces a small
timate is about 150 Pg (Shevliakova et al. reduction of acidity.
2009). Part of the range of uncertainty re- This kind of carbon dynamic is not new
flects inadequate knowledge of soil carbon in the history of life on Earth. Twice there
(belowground carbon). Soil carbon is prob- have been extremely high atmospheric con-
ably distributed patchily in any given soil centrations of CO2. In both instances they
type, and knowledge about that distribution were brought down to preindustrial levels
is sketchy. Even with this imprecision it has by natural processes (Figure 25.1) (Royer
been clear for some time that the number is 2006; Beerling 2008). In the Cretaceous the
large (Lal et al. 2012). Recently Sanderman et arrival of plants on land and a major op-
al. (2017) have estimated the amount of soil portunity for additional photosynthesis re-
carbon lost to the atmosphere from human sulted in a major CO2 sequestration. There
landuse to be on the order of 133 Pg C itself. was also a significant CO2 drawdown con-
The most recent estimate of the loss of tribution from soil creation. Although soil
CO2 from ecosystem destruction and degra- formation is essentially an abiotic process,
dation is that only 450 Pg remain in extant soil biota also contribute significantly. My-
ecosystems out of the 913 Pg prior to human corrhizal processes played a magnifying
impact (Erb et al. 2018). In other words the role.
amount of carbon in the atmosphere from The second major drawdown occurred
destroyed and degraded ecosystems is es- in the Paleogene and the Neogene because
sentially equal to what remains. of major engagement of modern flowering
Reaching a given atmospheric CO2 con- plants. In both cases the process took tens
centration does not result in instant conse- of millions of years—a luxury we do not
REGREENING THE EMERALD PLANET 329
tion. There are obviously places where the
potential is limited (e.g., urban centers), but
there is an enormous area of degraded land
where ecosystem restoration would be of
value in itself.
Reforestation has an important role to
play because forests hold so much carbon
per unit area, but there are other benefits
as well, including biodiversity conservation
and watershed functions. Reforestation in
the Amazon would secure the important
hydrological cycle that maintains that forest
and provides moisture south of the Amazon
Figure 25.1. During prior geological periods there were
two times when very high CO2 concentrations were
even as far as northern Argentina (Martinez
brought down to preindustrial levels by natural pro- and Dominguez 2014; Lovejoy and Nobre
cesses (mostly terrestrial photosynthesis and soil forma- 2018). Worldwide, reforestation is impor-
tion, both physical and biological). The first reduction tant in restoration of watershed function.
coincided with the arrival of higher plants on land and These forests need not all be natural, but
the second with the arrival of modern flowering plants.
(From Royer 2006.)
they could include production forests of
various kinds as well. The carbon and eco-
system service values are not substantially
have. We can, however, proactively restore affected by harvest of minor forest products
ecosystems (Lal et al. 2012; Lovejoy 2014; (e.g., rattan), which can provide additional
Griscom et al. 2017) as part of a global ef- economic return.
fort to recarbonize the biosphere by pro- Degraded grasslands and grazing lands
moting ecological restoration. represent considerable potential for carbon
A particularly important source of an- sequestration, not just in the plant systems
thropogenic carbon emissions from natural but also in the soils. Beyond the carbon
ecosystems is from peatlands. When peat- value, restoration would result in better
lands are drained, the peat oxidizes and grazing or other uses of grasslands, simul-
burns readily. In Southeast Asia there are taneously improving productivity.
extensive peat forests, which if deforested Degraded lands encompass abandoned
render the underlying peat deposits very agricultural lands and grazing lands that
vulnerable to burning, especially in the dry were originally grasslands or former for-
El Niño years. During those years, fires have ested lands. Globally degraded lands are
been so extensive in Indonesia and Borneo between 1 billion and 6 billion ha in ex-
as to cause health-threatening air pollution tent because of considerable uncertainty in
even in adjacent mainland (South) Asia. many instances (Gibbs and Salmon 2014).
There should be a concerted effort glob- Whatever the number, the potential to se-
ally to maintain all peatland as wet environ- quester carbon and simultaneously restore
ments. Carbon accretes slowly in peatlands, productivity is substantial.
so they do not have a major role to play in Agroecosystems also have tremendous
carbon sequestration through restoration, potential to sequester carbon instead of leak-
but it is important that their integrity as ing carbon, as is so often the case in current
carbon stocks be protected. agricultural practice. An important benefit
The Woods Hole Research Center, a of sequestering carbon is improvement in
world authority on the role of forests in soil fertility. Interestingly, even though in-
the carbon cycle, is currently preparing a dustrial agriculture practices would seem
global map of potential ecosystem restora- counter to that, there are places in the
330 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
American Midwest where no-till and low- lost if climate change causes Amazon dieback
till practices have taken hold spontaneously. (transformation to a savanna vegetation) in
An important aspect is to restore riparian the eastern and southeastern Amazon (No-
vegetation that improves water quality and bre et al. 2016). Southern edges of North-
prevents soil erosion. That in turn adds nat- ern Hemisphere boreal forest could also die
ural connectivity in the landscape—so im- back, although they might be offset by new
portant for normal dispersal of plants and boreal forests on the northern edge of the
animals, and even more so under climate current distribution. With more warming,
change, which will drive many organisms there could be massive release of methane as
to move and track their climatic envelope. permafrost melts. Basically, the ability to use
Only relatively recently has it been rec- ecosystem restoration to sequester carbon
ognized that coastal wetlands, which to- depends on not straying too far from cur-
tal 49 million ha worldwide, have an im- rent climatic regimes, or at least understand-
portant role to play in the carbon cycle ing the dynamics of possible climatic change
(Pendleton et al. 2012). Mangroves, coastal and integrating it into planning.
marshes, and seagrasses have experienced The success of ecosystem restoration for
considerable destruction and degradation climate management will require a more
but are being recognized as important buf- integrated and sustainable approach to de-
fers against storm surge. The Mississippi velopment, especially in the face of addi-
Delta, where such ecosystems have been de- tional billions of people and the need to
graded and destroyed, is undergoing prior- feed and provide for them adequately. That
ity restoration for those reasons alone, but can occur only with a more coordinated
there are significant carbon sequestration approach to development (see Chapter 28):
benefits as well. one that avoids isolated decision making
Ecosystem restoration has yet to emerge and puts a premium on sustainable infra-
as the global priority it deserves. Still, the structure and protection or conservation of
2015 New Climate Economy report pro- natural ecosystems. It is a vision congruent
duced for the United Nations concluded with Edward O. Wilson’s (2016) Half-Earth,
there can be healthy economic growth which argues for half of the planet to be set
while rapidly reducing emissions, if there is aside primarily for nature.
a strong land use component and all three At a finer and more local scale, manage-
“Rio conventions” (i.e., UN Framework ment of ecosystems can have important car-
Convention of Climate Change, Conven- bon value implications. Changes in trophic
tion on Biological Diversity, and UN Con- structure can directly affect carbon values,
vention to Combat Desertification) include as in the example of old fields and different
restoration as part of their core agendas. dominant species of predators (Case Study
Many of the independently determined na- 11). In blue-carbon ecosystems, predators
tional commitments made at the 2015 Paris are important for maintaining high carbon
Climate Change Conference of the Par- stocks (Atwood et al. 2015).
ties focused not only on forest protection The term “Anthropocene” was coined
but also on forest restoration. The Nature to recognize a new geological age in which
Conservancy has developed a major port- human impacts became planetary and geo-
folio around land management for climate logical in scale, like the giant meteor that
change mitigation. ushered out the dinosaurs. It could also
Extensive restoration is not simple to come to mean a new age in which we learn
achieve on a planet already experiencing cli- to manage ourselves for better outcomes for
mate change and a large, still-growing hu- humans and other forms of life.
man population. The large amount of carbon When ecosystem restoration and man-
sequestered in the Amazon forest could be agement attain their rightful place in the
REGREENING THE EMERALD PLANET 331
climate change agenda, it will transform tems and the Global Carbon Cycle, ed. R. Lal, K. Lorenz, R.
public perception of the planet on which F. Hüttl, B. U. Schneider, and J. von Braun, 59–82.
Springer. https://doi.org/10.1007/978-94-007-4159-1
we live. It will require recognition that the _4.
Earth functions not just as a physical sys- Houghton, R. A., B. Byers, and A. A. Nassikas. 2015. “A
tem but also as a combined biological and role for tropical forests in stabilizing atmospheric
physical system—and that we in fact live CO2.” Nature Climate Change 5: 1022–1023.
on, and are part of, a living planet. Such a Kopp, R. E., F. J. Simons, J. X. Mitrovica, A. C. Maloof,
and M. Oppenheimer. 2009. “Probabilistic assessment
view would profoundly alter our sense of of sea level during the last interglacial stage.” Nature
our place in nature and what is appropriate 462: 863–867.
behavior. Lal, R., K. Lorenz, R. F. Huttle, B. U. Schneider, and J.
Similarly, it can empower individuals von Braun, eds. 2012. Recarbonization of the Biosphere: Eco-
who previously could see no way in which systems and the Global Carbon Cycle. Springer. https://doi
.org/10.1007/978-94-007-4159-1_4.
they individually can make a difference Lovejoy. T. E. 2014. “A natural proposal for address-
about climate change because it is such a ing climate change.” Ethics and International Affairs 28:
gigantic problem. In a simplistic way every 359–363.
individual can plant a tree or help restore Lovejoy, T. E., and C. Nobre. 2018. “Amazon tipping
a coastal wetland, much as people of all point.” Science Advances 4: eaat2340.
Martinez, J., and F. Dominguez. 2014. “Sources of atmo-
walks of life were widely involved in vic- spheric moisture for the La Plata River basin.” Journal of
tory gardens during World War II. Climate 27: 6737–6753.
Midgley, G. 2018. “Narrowing pathways to a sustain-
able future.” Science 360 (6390): 714–715. https://doi
R E FE R E NCE S .org/10.1126/science.aat6671.
National Academy of Science. 2015. Climate Intervention:
Carbon Dioxide Removal and Reliable Sequestration and Climate
Atwood, T. B., R. M. Connolly, E. G., Ritchie, C. E.
Intervention: Reflecting Sunlight to Cool Earth. National Acad-
Lovelock, M. R. Heithaus, G. C. Hays, J. W. Fourqurean,
emies Press. https://doi.org/10.17226/18988.
and P. I. Macreadie. 2015. “Predators help protect car-
Nobre, C. A., G. Sampaio, L. Borma, J. C. Castilla-Rubio,
bon stocks in blue carbon ecosystems.” Nature Climate
J. S. Silva, and M. Cardoso. 2016. “The fate of the Ama-
Change 5: 1038–1045.
zon forests: Land-use and climate change risks and the
Beerling, D. J. 2008. The Emerald Planet. Oxford University
need of a novel sustainable development paradigm.”
Press.
Proceedings of the National Academy of Sciences 113 (39):
Erb, K. H., T. Kastner, C. Plutzar, A. S. Bais, N. Carvalhais,
10759–10768.
T. Fetzel, S. Gingrich, H. Haberl, C. Lauk, M. Neiderts-
Pendleton, L., et al. 2012. “Estimating global ‘blue car-
cheider, J. Pongratz, M. Turner, and S. Luyssaert. 2018.
bon’ emissions from conversion and degradation of
“Unexpectedly large impact of forest management
vegetated coastal ecosystems.” PLOS One 7: 1–7.
and grazing on global vegetation biomass.” Nature 553:
Royer, D. A. 2006. “CO2-forced climate thresholds dur-
73–76.
ing the Phanerozoic.” Geochimica 70: 5665–5675.
Gibbs, H. K., and J. M. Salmon. 2014. “Mapping the
Sanderman, J., T. Hengl, and G. J. Fiske. 2017. “Soil car-
world’s degraded lands.” Applied Geography 57: 12–21.
bon debt of 12,000 years of human land use.” Proceed-
Griscom, B. W., J. Adams, P. W. Ellis, R. A. Houghton,
ings of the National Academy of Sciences 114: 9575–9580.
G. Lomax, D. A. Miteva, W. H. Schlesinger, et al. 2017.
Shevliakova, E., et al. 2009. “Carbon cycling under 300
“Natural climate solutions.” Proceedings of the National
years of land use change: Importance of the sec-
Academy of Sciences 114 (44): 11645–11650.
ondary vegetation sink.” Global Biogeochemical Cycles 23
Hansen, James, et al. 2013. “Assessing ‘dangerous cli-
(GB2022): 1–16.
mate change’: Required reduction of carbon emissions
Warren, R., J. Price, E. Graham, N. Forstenhaeusler, and
to protect young people, future generations and na-
J. VanDerWal. 2018. “The projected effect on insects,
ture.” PLOS One 8: 81648.
vertebrates, and plants of limiting global warming to
Houghton, R. A. 2012. “Historic changes in terrestrial
1.5°C rather than 2°C.” Science 360 (6390): 791–795.
carbon storage.” In Recarbonization of the Biosphere: Ecosys-
https://doi.org/10.1126/science.aar3646.
CASE STUDY 11
Efforts to mitigate climate change have tended to focus on managing human activities like
fossil-fuel burning and land clearing, and on innovating technologically, to reduce CO2
emissions to the atmosphere (Pacala and Socolow 2004). But this particular focus on con-
trolling atmospheric inputs represents only one side of the equation for balancing the carbon
budget. A potentially important yet largely untapped opportunity is to actively enlist natural
ecological processes to recapture carbon within terrestrial and aquatic ecosystems, thereby
offsetting atmospheric buildup. Indeed, ecosystems globally are removing as much as half
of the CO2 emitted to the atmosphere each year from anthropogenic activities (Schmitz
et al. 2014). But it may be possible to do more through strategic management of the func-
tional composition of living organisms—especially animals—and their interactions within
ecosystems, thereby enhancing the nature and magnitude of carbon exchange and storage
(Schmitz et al. 2014). Recent estimates suggest that the magnitudes of animal effects can be
profound, holding much promise in protecting and managing biodiversity as a way to help
manage the global carbon budget.
Animals can determine the fate of car- cies composition of the plant community as
bon in ecosystems through trophic interac- well as the amount of carbon in senescent
tions that directly and indirectly influence plant matter entering the soil storage pool
the abundance and chemical composition (Schmitz et al. 2017). Whether predator ef-
of plant and animal biomass and its fate as fects on herbivores are largely consumptive
soil organic matter. Top predators especially or nonconsumptive generally depends on
can trigger knock-on effects that modulate their hunting mode: active-hunting preda-
the amount of carbon exchanged between tors generally have consumptive effects,
ecosystem reservoirs (e.g., soils and plants) whereas sit-and-wait predators have non-
and the atmosphere (Schmitz et al. 2014). consumptive effects.
Predators do this in at least two ways. They Demonstration of these two predation
may have consumptive effects that reduce effects at play come from empirical analy-
herbivore density (and hence herbivory), in ses examining the link among spider pred-
turn enabling more plant-based carbon to ators, grasshopper prey, and grasses and
enter soil organic matter storage pools as herbs among old-field ecosystems in Con-
uneaten senescent plant biomass, relative necticut, United States. The important play-
to conditions where predators are absent ers in these ecosystems are three functional
(Schmitz et al. 2014). Their presence within groups of plants represented by the herb
ecosystems may also cause nonconsumptive Solidago rugosa (hereafter Solidago), a variety of
fear effects that cause herbivores to shift other old-field herb species, and grasses; a
their foraging preferences from nitrogen- dominant generalist grasshopper herbivore
rich plants that support growth and high Melanoplus femurrubrum; and the sit-and-wait
reproduction to plants with higher con- spider predator Pisaurina mira and active-
tent of soluble carbon to support height- hunting predator Phidippus clarus. In these sys-
ened metabolic costs caused by chronic
stress from fear of predation (Schmitz et al. Copyright © 2019 by Yale University. All rights reserved.
2017). Such a diet shift can alter the spe- ISBN 978-0-300-20611-1.
332
ENLISTING ECOLOGICAL INTER ACTIONS AMONG ANIMALS 333
tems, the plant functional groups compete tion. This relationship improves when ac-
asymmetrically (see Figure CS11.1), with counting for the strength of the ecological
Solidago competitively dominating the other interactions of predators on the plant com-
two functional groups of plants (Schmitz et munity (Schmitz et al. 2017). Most striking
al. 2017). is that shifting from a dominance of sit-
Solidago competitive dominance is altered, and-wait predators to a dominance of ac-
however, by hunting-mode-dependent ef- tive-hunting predators can result in a dou-
fects of predators on grasshoppers, which bling of soil carbon retention. This range in
in turn shape the composition of the plant per unit area of carbon retained in soils ri-
community predominantly through chang- vals or exceeds the range achieved through
ing effects on Solidago and grasses (Figure conventional management of soil carbon
CS11.1). In the absence of predators, grass- and land use (Post and Kwan 2000; Guo
hoppers prefer nitrogen-rich grasses to and Gifford 2002). More important, several
maximize survival and reproduction. Sit- other candidate factors, such as live plant
and-wait spiders are persistently present biomass, insect diversity, soil arthropod
and thus trigger grasshoppers to modify decomposers, degree of land-use develop-
foraging in favor of Solidago because it is a ment around the fields, field age, and soil
key source of dietary soluble carbohydrate texture, did not significantly explain soil
to fuel heightened grasshopper respiration carbon retention among the fields (Schmitz
caused by chronic stress from perceived et al. 2017).
predation risk (Schmitz et al. 2017). The These ecological principles are also scal-
potential for predator-prey encounter is less able. The loss of important predators—from
frequent with widely roaming active-hunt- wolves in boreal forests to sharks in seagrass
ing spiders. Consequently risk of predation meadows—can lead to growing popula-
is episodic for any individual grasshop- tions of terrestrial and marine herbivores,
per, making it energetically inefficient for whose widespread grazing can and does re-
grasshoppers facing active-hunting spiders duce the ability of ecosystems to absorb and
to be chronically stressed. Hence, grasshop- protect hundreds of years of built-up soil
pers facing active-hunting predators favor and sediment carbon (Atwood et al. 2015;
grasses over Solidago (Figure CS11.1). Wilmers and Schmitz 2016). These encour-
As a consequence, when active-hunting aging new insights support expanding con-
spiders dominate, Solidago, along with its sideration of biodiversity management to
relatively higher carbon content, flour- balance carbon budgets. Humans already
ishes and dominates the plant community. manage populations of many animals for
When sit-and-wait predators dominate, other purposes. So, while representing a
Solidago is suppressed, favoring a more eq- shift in management goals, this would not
uitable representation of the three func- involve a radical shift in management ap-
tional plant groups in the plant community proach. But individual animal species tend
with less total carbon. Intermediate com- to occur regionally, not globally. This re-
binations of the two spiders should result quires shying away from finding the single,
in intermediate levels of plant functional global-scale, home-run solution to seek
group abundances. Sampling among 15 old more regionally nuanced alternatives. Ani-
fields across Connecticut revealed that the mal management has the advantage that it
amount of carbon retained in soils among can take into account the values and prefer-
these old fields is related to the relative ences of local societies and would allow for
abundance of active-hunting and sit-and- regional players to reconcile their particular
wait predators within the fields, mediated concerns and values with broader climate
via changes in plant community composi- solutions. The many local and regional
ENLISTING ECOLOGICAL INTER ACTIONS AMONG ANIMALS 335
animal management strategies could then Pacala, S., and R. Socolow. 2004. “Stabilization wedges:
add up globally to create a portfolio of so- Solving the climate problem for the next 50 years with
current technologies.” Science 305: 968–972.
lutions that, together with other mitiga- Post, W. M., and K. C. Kwan. 2000. “Soil carbon seques-
tion efforts (Pacala and Socolow 2004), can tration and land use change: Processes and potential.”
meaningfully help slow climate change. Global Change Biology 6: 317–327.
Schmitz, O. J., R. W. Buchkowski, J. R. Smith, M. Telt-
horst, and A. E. Rosenblatt. 2017. “Predator com-
munity composition is linked to soil carbon reten-
R E FE R E NCE S tion across a human land use gradient.” Ecology 98:
1256–1265.
Atwood, T. B., R. M. Connolly, E. G. Ritchie, C. Schmitz, O. J., P. A. Raymond, J. A. Estes, W. A. Kurz,
E. Lovelock, M. R. Heithaus, G. C. Hays, J. W. G. W. Holtgrieve, M. E. Ritchie, D. E. Schindler, et
Fourqurean, and P. I. Macreadie. 2015. “Predators help al. 2014. “Animating the carbon cycle.” Ecosystems 7:
protect carbon stocks in blue carbon ecosystems.” Na- 344–359.
ture Geoscience 5: 1038–1045. Wilmers, C. C., and O. J. Schmitz. 2016. “Effects of
Guo, L. B., and R. M. Gifford. 2002. “Soil carbon stocks wolf-induced trophic cascades on ecosystem carbon
and land use change: A meta-analysis.” Global Change cycling.” Ecosphere 7: e01501.
Biology 8: 345–360.
Figure CS11.1. Case example of how spider predators with dago, thereby suppressing Solidago competitive dominance
different hunting modes influence food-web interactions (negative indirect effect of effect of predators on Solidago).
in old-field ecosystems, with attendant effects on plant Plant community composition: When sit-and-wait spiders domi-
community composition and soil carbon retention. Trophic nate, the plant community is equitably comprised of the
interactions: Depiction of direct (solid arrows) and indirect three plant functional groups. When active-hunting spi-
(dashed arrows) interactions among hunting spiders, ders dominate, Solidago dominates the plant community.
grasshopper herbivores, and three functional groups of Mixed communities of active-hunting and sit-and-wait
plants (left to right: carbon-rich Solidago, less carbon-rich spiders result in intermediate levels of the three plant
herbs and grasses). The strength of grasshopper direct ef- functional groups. Hence sit-and-wait spider dominance
fects on a plant functional group, via diet selection (de- leads to lower biomass density of carbon in the plant
picted as arrow thickness), depends on which hunting community than when active-hunting spiders domi-
mode of spider predator dominates and thus mediates nate. Soil carbon retention: Sampling of soil carbon among 15
competition among the plant functional groups. Domi- old-field ecosystems reveals that the knock-on effects of
nance of active-hunting predators like Phidippus clarus leads spiders on carbon density of the plant community carry
to grasshopper preference for grasses, thereby abetting further to determine the amount of carbon retained in
Solidago competitive dominance (positive indirect effect of soils, which is related to the relative abundance of active-
predators on Solidago). Dominance of sit-and-wait predators hunting and sit-and-wait predators within an ecosystem.
like Pisaurina mira leads to grasshopper preference for Soli- (Courtesy of Oswald Schmitz, Yale University.)
CHAPTER TWENTY-SIX Everything should be made as simple as possible,
but no simpler.
Increasing Public —Quote commonly attributed to Albert Einstein
336
I N C R E A S I N G P U B L I C AWA R E N E S S A N D FAC I L I TAT I N G B E H AV I O R C H A N G E 337
range of science and science-based institu- everything we can to make the behaviors
tions that strive to share current scientific we are promoting easy, fun, and popular. I
insights about the physical world with deci- refer to each of these as “heuristics,” in the
sion makers and the public (e.g., National sense that they organize a relatively large
Academy of Science, National Science Foun- amount of prescriptive information into a
dation, professional societies, science muse- relatively easy-to-use method or process.
ums, media producers). In the remainder of this chapter, I un-
My answer also aims to be useful regard- pack these two heuristics, with the aim of
less of which category of decision makers making them practical for all readers. I as-
is most relevant in a given situation—com- sume that most readers of this chapter will
munity leaders, national leaders, business be scientists and allied professionals—and
leaders, people in a specific profession (e.g., I therefore tailor my comments to them—
building contractors), or even individu- but the recommendations are equally rel-
als and families who are trying to manage evant to anyone seeking to improve climate
their own lives in the best possible manner. change and biodiversity outcomes in ways
All of these people have important climate- that are grounded in scientific evidence.
and biodiversity-related decisions to make,
whether they currently know it or not. Indi-
vidual scientists and the scientific commu- SH A R I NG W H AT W E K NOW
nity can be of value in helping these people
make wise decisions and take wise actions. Scientists are highly trained to share what
The question posed above includes two they know, but primarily to colleagues in
related yet distinct challenges. The first their own scientific discipline. Typically,
challenge is helping decision makers make this process begins with our research,
wise decisions; the second challenge is where we develop and test ideas. If an idea
helping them take wise actions. To help proves to have merit, we take steps to share
people make wise decisions, we must effec- it with our colleagues—at professional
tively bring the issue to their attention, sug- meetings, in journal articles, and in books.
gest the need to make decisions, clarify the Perhaps if we are really excited by the idea,
nature of the problems and opportunities, we might make extra efforts to share it more
and make available the best science-based broadly—possibly with scientists in other
information—in appropriate formats—for disciplines (e.g., by giving talks at their
decision makers to consider. In short, we meetings) or with the general public (e.g.,
must effectively share what we know. by working with our institution’s press of-
Helping people take wise actions in- fice to issue a press release), although these
volves a different set of activities. If it were efforts tend to be fleeting. These approaches
easy for people to convert their good deci- to sharing what we know work reasonably
sions (i.e., their good intentions) into effec- well with colleagues in our own discipline,
tive actions, the proverb “the road to hell is less well with scientists in other disciplines,
paved with good intentions” would never and not well enough with policy makers,
have arisen. Fortunately, steps can be taken business leaders, and members of the pub-
to help people convert their good inten- lic. Metaphorically, these approaches are
tions into effective actions. akin to tossing what we know “over the
My answer to the question, therefore, transom” or out the window of our lab, and
has two parts. To effectively share what we into the outside world, expecting relevant
know, we need simple clear messages, re- people to pick up our knowledge, consider
peated often, by a variety of trusted sources. it, and know what to do with it.
To help people convert their good inten- There is a better way: simple clear mes-
tions into effective actions, we need to do sages, repeated often, by a variety of trusted
338 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
sources. This is not a “magic bullet,” and it manner, we stand a much better chance of
will not solve our biodiversity and climate designing messages that illuminate, rather
challenges overnight (or anytime soon), than messages that alienate. Admittedly,
but the approach is evidence based (based conducting audience research is not always
in the science of science communication), feasible (perhaps because of lack of time,
reasonably easy (once you understand it, funds, or expertise), but that is not an ex-
it is no harder than what you are already cuse for failure to seriously consider how
doing), and ethical (it involves providing members of the target audience are likely
people with information they are likely to to understand the information we wish to
find helpful). The heuristic itself has three share with them. Making the effort to con-
elements—simple messages, message rep- sider our audience members’ views—and
etition, and trusted messengers—each of values—is helpful in that it forces us as
which is based on empirical evidence and communicators to look beyond ourselves
offers practical guidance. and to think carefully about both what is
most worth saying and how best to say it.
An approach that all science communi-
The Importance of Message Simplicity
cators can use to improve their messages in-
Most people, in most situations, do not deal volves anticipating the questions that people
well with complex information; complex are likely to ask, and drafting messages that
information is cognitively taxing, and most attempt to proactively answer those ques-
people (even very bright people), in most tions. For example, when the issue pertains
situations, are unwilling to make the effort. to a risk, people are likely to ask some varia-
Instead, people typically use mental short- tion of the following questions: What is the
cuts to avoid cognitively difficult tasks, and problem? How will it affect me (and my
when they do, they often end up reaching people)? How serious is it? Who or what is
conclusions that differ from those intended causing the problem? What are the options
by the information provider (Kahneman for dealing with it? What, if anything, can
2011). The risk communication expert Ba- I do about it?
ruch Fischhoff (1989) trenchantly summa- To test your success in drafting simple,
rized the situation—and his prescription clear messages in response to likely audi-
about how best to manage it—in the fol- ence questions, share the messages with a
lowing manner: “People simplify.” Our job few members of your target audience, one
(as risk and science communicators) is to person at a time. After each person has had
help people simplify appropriately. a chance to consider your messages, ask
So, what can we—as communicators— each one to explain—in his or her own
do to help people simplify appropriately? words—what the messages mean and how
We can develop “messages” about the in- he or she would explain them to a friend.
formation we wish to share that are specifi- Also ask, “What questions, if any, do you
cally intended to help people simplify com- have for me about this information?” When
plex information appropriately. (For readers members of your audience can adequately
who disfavor the term “messages”—equat- explain your messages in their own words,
ing it to persuasive intent—the term “brief and when your messages help them ask
statements” is an acceptable synonym.) Au- good questions, you have succeeded in
dience research is a powerful tool for de- writing simple, clear messages.
veloping such messages. Through audience The Third U.S. National Climate Assess-
research, we can systematically collect data ment (NCA3) provides a useful example of
to assess audience members’ preexisting simple clear messages (Melillo et al. 2014).
knowledge, attitudes, and values, and test Although the full report is more than 1,000
their responses to draft messages. In this pages, for the NCA3 each set of chapter au-
I N C R E A S I N G P U B L I C AWA R E N E S S A N D FAC I L I TAT I N G B E H AV I O R C H A N G E 339
thors was asked to identify key messages ness cognitively (by increasing salience
for their chapter. In addition, the federal and availability of the information) and
advisory committee that oversaw the de- affectively (by increasing positive feelings
velopment of NCA3 developed key mes- about the message) (Batra and Ray 1986;
sages for the report as a whole. Authors and Pechman and Stewart 1988; Chong and
advisory committee members were well Druckman 2013). Although truthfulness is
aware of audience research showing that of paramount importance in science com-
most Americans understand that climate munication, truthful messages amount to
change is happening but see it as a distant little without adequate message repetition,
threat—distant in space (i.e., the problems a point nicely illustrated by this quote from
will primarily manifest elsewhere, not in a political consultant: “You take the truth,
the United States), distant in time (i.e., the and I’ll take repetition; I’ll beat you every
problems will start in the future and are time” (Castellanos, personal communica-
not happening yet), and distant from hu- tion, 2010).
mans (i.e., the problems will be primarily The importance of message repetition is
felt by plants, penguins, and polar bears— something that every politician learns in
not people; Leiserowitz 2005; Leiserowitz her first political campaign, and every busi-
et al. 2017). As a result, NCA3 authors de- ness executive learns in his first marketing
veloped key messages intended to correct course, but it is a lesson rarely taught to sci-
the misperception of climate change as a entists. Admittedly, repetition is boring to
distant threat. The opening words on the most communicators, especially scientists.
NCA3 report website are the following: As scientists, novelty, innovation, and con-
“The National Climate Assessment provides troversy are what excite us—and what we
an in-depth look at climate change impacts like to talk about—but what excites us is
on the U.S. It details the multitude of ways not a relevant consideration in determin-
climate change is already affecting and will ing how best to share what we know with
increasingly affect the lives of Americans. decision makers. Moreover, like all people,
Explore how climate change affects you scientists suffer the “curse of knowledge”
and your family” (Melillo et al. 2014). This (Heath and Heath 2007); we forget that
theme is also clearly in evidence in the re- most people do not know what we know,
port’s introduction. These opening words, and as a result we end up making assump-
key messages, and even chapter names (e.g., tions in our communication that inad-
extreme weather, human health, water, ag- vertently excludes the very people we are
riculture, oceans) were all intended to help seeking to share our knowledge with. The
all readers—even those giving the informa- discipline of message repetition—repeating
tion only a quick glance—to understand the the messages that we have designed for the
most important, overarching, finding of the express purpose of helping audience mem-
assessment: climate change is happening bers simplify appropriately—forces us to
here, now, in every region of America. stay true to our plan for sharing the infor-
mation that is most helpful to members of
our target audience (rather than sharing the
The Importance of Message Repetition
information that most interests us).
One of the most robust findings to have Fortunately, message repetition is not
ever emerged from communication re- the sole burden of any one individual or
search is that “repetition is the mother of organization; message repetition works best
all learning” (Lang 2013)—an expression when many different messengers repeat
that comes to us originally from an ancient the same set of messages, consistently, over
Latin proverb (repetitio est mater studiorum). time. Individuals and organizations work-
Repetition increases message persuasive- ing on climate change and biodiversity
340 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
issues should develop the discipline to work Climate change and biodiversity com-
together to design, use, and repeat—at ev- municators can earn the trust of their
ery communication opportunity—a shared target, and leverage the impact of their
set of messages specifically intended to help communication, by recruiting additional
audience members reach appropriate con- trusted voices—people who are known by
clusions about the complex problems they the target audience, personally or by repu-
are urging them to engage with. tation—to embrace, repeat, and thereby
Reach—that is, reaching members of validate their simple clear messages. These
your intended target—is an important con- additional trusted voices need not necessar-
sideration, too. Messages that are repeated ily be from the science community. Indeed,
often but fail to reach their intended audi- we can and should cultivate communica-
ence will have no benefit for that audience. tion partnerships with individuals and or-
Consumer brands typically strive to achieve ganizations outside the realm of climate
both message reach and frequency with a change and biodiversity who are highly
combination of paid advertising, earned trusted by members of our audience—for
media (i.e., outreach to news media and example, leaders in the faith community
bloggers), social media, endorsements, or in the business community—because
paid placements, and other means. Climate doing so is a way of demonstrating one’s
change and biodiversity professionals and trustworthiness, and of maximizing one’s
organizations rarely have the opportunity message reach and frequency.
to achieve reach through paid placements The most effective endorsements come
(e.g., advertising), but through the kind of from people that our target audience trusts
collaboration suggested in the paragraph the most, regardless of their level of ex-
above, they can strive to maximize both pertise. On the issue of climate change, for
message reach and frequency (i.e., repeti- example, people typically trust most the
tion), especially to the extent that they suc- people they know the best—their family
ceed in bringing other trusted messengers members, friends, and coworkers (Figure
into the communication mix. For further 26.1; Leiserowitz et al. 2009). Scientists are
elaboration of this important idea, read on. highly trusted, too, but it is the rare indi-
vidual who places greater trust in a scientist
(whom he or she has never met) than in
The Importance of Trusted Messengers
his or her own family and friends. This is
Quite simply, where there is no trust, there precisely why the best test of the simplic-
can be no learning. As a group, scientists ity and clarity of a science-based message
are highly trusted. For example, scientists is whether members of the target audience
are trusted “a lot” by two-thirds of Ameri- are willing and able to convey the message
can adults—tied with medical doctors and to their family, friends, and coworkers.
second only to members of the military and Ultimately, that should be the aim of our
teachers (Pew Research Center 2013). How- communication—to motivate and enable
ever, when target audience members do not members of our target audience to share
know the specific scientist who is attempt- our messages with one another.
ing to communicate with them—person- Feeling overwhelmed at the prospect
ally or by reputation—their trust in that of designing and communicating simple,
messenger is likely to be superficial, provi- clear messages, repeated often, by a variety
sional, and vulnerable, and communication of trusted sources? Don’t be. In his article
mistakes (e.g., unclear messages, seemingly “Communicating about Matters of Greatest
evasive answers, lack of empathy) can rap- Urgency: Climate Change,” Baruch Fischhoff
idly undermine trust (Maibach and Covello (2007) made a strong case for improving the
2016). effectiveness of science communication by
I N C R E A S I N G P U B L I C AWA R E N E S S A N D FAC I L I TAT I N G B E H AV I O R C H A N G E 341
Figure 26.1. Americans’ trust in various possible sources By way of example, I (as a social scientist)
of information about global warming. Data from Yale/ helped organize a team of climate scien-
George Mason University Climate Change in the Ameri-
can Mind surveys conducted between 2008 and 2012
tists, social scientists, and communications
(Leiserowitz et al. 2012). practitioners at various universities (George
Mason and Yale), nonprofit organizations
approaching it as a team sport rather than (Climate Central, American Meteorologi-
as solo sport in which every scientist is ex- cal Society), and government agencies (Na-
pected to be master of the art and science of tional Oceanic and Atmospheric Admin-
communication. Specifically, Fischhoff en- istration, National Aeronautics and Space
courages the development of science com- Adminstration) to develop and distribute to
munication teams that include three distinct TV weathercasters broadcast-quality mate-
types of expertise, which can be provided by rials to help them report on the local im-
a minimum of three people: a content scien- pacts of climate change in their area. Called
tist (i.e., a person with expertise on the risk Climate Matters, the collaboration started with
or the issue), a social scientist (i.e., a person a successful pilot test at a single TV sta-
with expertise on how people interpret in- tion in Columbia, South Carolina (Zhao et
formation), and a communications practitio- al. 2014), and has expanded to a national
ner (i.e., a person with expertise in creating network with more than 500 participating
communications opportunities). These three weathercasters, and growing (Placky et al.
types of professionals each bring unique 2016; Maibach et al. 2016).
knowledge and skills to the process of devel- In addition to sustained collaborations,
oping simple, clear messages and in working as described in the Climate Matters example,
to ensure that those messages are conveyed this team-based approach to science com-
often, by a variety of trusted sources. munication is also practical for ad hoc
342 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
communications opportunities. For exam- (e.g., “I’ll have the cheeseburger and fries,
ple, climate scientists—who will soon be please”). One of the most effective means
publishing findings with important impli- of reducing this attitude-behavior gap is to
cations for decision makers—can ask a so- make the recommended behavior easier to
cial science colleague and a member of the perform (e.g., “Would you like carrot sticks,
media relations team at their institution to an apple, or fries with your burger?”).
help them craft messages and develop and People are likely to perform easy behav-
implement a communication plan through iors that they believe to be in their best in-
which to communicate the messages. terest, but they often defer—and never get
around to—behaviors they find more dif-
ficult. To save money on utility bills, for ex-
I N F LU E NC I NG BE H AV IOR ample, a homeowner may switch her lights
over time from incandescent bulbs to LEDs
Effective communication is important, al- (because doing so is relatively easy), but she
though it is often not sufficient to change may not take steps to weatherize her home
people’s behavior (Hornik 2002; McKen- (because doing so is harder), despite the
zie-Mohr 2011). Even after people decide fact that the cost savings from the latter are
to take action, many will not, or they will considerably larger.
not persevere long enough to succeed. Con- Many important actions are not easy to
sider, for example, your most recent New perform. Steps can be taken, however, to
Year’s resolution. make them easier to perform. In his excel-
Social marketing—the use of marketing lent book Fostering Sustainable Behavior, Doug
methods to promote behaviors that benefit McKenzie-Mohr (2011) recommends taking
society—is a method developed specifically an engineering-like approach to the task of
to help address this problem (Maibach et making behaviors easier. The first step in
al. 2002). Many excellent texts lay bare the the process is to conduct audience research
principles of social marketing, including for the purpose of identifying the barriers
two that specifically explore its application that impede people’s performance of a be-
to environmental challenges (McKenzie- havior of whose value they are already con-
Mohr 2011; McKenzie-Mohr et al. 2012). vinced. These barriers might include a lack
I particularly encourage readers of this of knowledge about how to perform the ac-
chapter, however, to watch a TED Talk by tion (e.g., “I can’t remember which kinds of
Bill Smith (2011)—one of social market- fish are sustainable”), a lack of skills neces-
ing’s pioneers. In his talk, Smith lays out a sary to perform the behavior well (e.g., “I
simple heuristic to guide the implementa- don’t know how to cook that kind of fish”),
tion of social marketing programs: make a lack of necessary resources (e.g., “Sustain-
the behavior you are promoting easy, fun, ably caught fish is too expensive”), concern
and popular. Although it sounds cheeky, the about the negative consequences of per-
heuristic is based on a large body of empiri- forming the behavior incorrectly (e.g., “My
cal research, and it offers important, practi- kids won’t eat it if they don’t like it”), and
cal guidance. so on.
The next step in McKenzie-Mohr’s ap-
proach is to develop and pilot test ways of
The Importance of Making
reducing—or ideally eliminating—the bar-
the Behavior Easy
riers found to be particularly common. The
Social scientists have long known that there Seafood Watch app developed by the Mon-
is often a large gap between people’s at- terey Bay Aquarium is a good example of
titudes toward a behavior (e.g., vegetables a program intended to reduce at least one
are very good for you) and their behavior barrier to purchasing sustainably caught
I N C R E A S I N G P U B L I C AWA R E N E S S A N D FAC I L I TAT I N G B E H AV I O R C H A N G E 343
fish—not knowing which fish are sus- climate change impacts stories, including
tainably caught. If the pilot-test results are lack of time to prepare stories, lack of ac-
promising, efforts can be made to encour- cess to data on local impacts, and lack of
age widespread adoption of the approach. access to appropriate graphics and visuals
The Marine Stewardship Council’s Certified to support their reporting. In response, to
Sustainable Seafood program is an example make the recommended behavior easier
of a program that has achieved consider- for weathercasters, each week our team
able success through adoption by large food produces and distributes broadcast-quality
companies and retailers. graphics, customized to each participating
Another important way to make behav- weathercasters’ media market, which often
iors easier is to have members of the tar- feature data on the local impacts of climate
get audience demonstrate to other audience change in their area; see Figure 26.2 for
members how they perform the behavior, an example. (All current and past Climate
live or on video (Bandura 2004). Modeling Matters materials are available online at the
demonstrations of this type are particularly website http://www.climatecentral.org/
effective when the models make explicit climate-matters.) To model use of these
the necessary steps to perform the behavior materials, and thereby increase participat-
successfully, the pitfalls to avoid, and the ing weathercasters’ sense of self-efficacy in
benefits of performing the behavior. Both using the materials on-air, the Climate Matters
of these approaches—reducing barriers and Facebook page includes examples of how
modeling the behavior—will increase peo- other weathercasters have used the materi-
ple’s sense of self-efficacy (self-confidence) als (see http://www.facebook.com/climate
to perform the recommended action, which .matters/videos).
increases the odds that people will try, per-
severe, and eventually succeed in perform-
The Importance of Making
ing the behavior (Bandura 2004).
the Behavior Fun
In their terrific book Switch, Chip and
Dan Health (2010) lay out a host of practical Climate change and biodiversity experts
ways to make behavior change easier and recommend behaviors not because the be-
to make behavior change programs more haviors are fun, but because they offer im-
successful. Drawing on the metaphor of a portant benefits. Regardless, experts should
rider (to represent people’s thoughts), an not lose sight of two important facts: people
elephant (to represent people’s emotions), are more likely to perform behaviors that
and their path (to represent the social and are fun than behaviors that are not; and
physical environment in which people are receiving benefits is fun, while incurring
operating), the Heaths recommend setting costs is not.
a clear (i.e., unambiguous) goal, charting People are often willing to incur costs to
milestones so that progress made toward secure benefits that they value (including
the goal will be positively reinforcing, and but certainly not limited to fun). The most
“tweaking the environment” (i.e., modify- attractive offers, however, are those that de-
ing or removing personal, social, or envi- liver valued benefits to people at the same
ronmental barriers to performance of the time as—or even before—they are required
behavior). to incur the costs (e.g., “No money down
The Climate Matters program provides an . . . take the car today”). Conversely, the
example of how my colleagues and I have least attractive offers are those that require
sought to make behavior change easier for incurring costs up front and receiving the
TV weathercasters. Our audience research benefits only much later (Rothschild 1999).
with weathercasters identified several key Many actions recommended to prevent cli-
barriers to their ability to report on local mate change and species loss are seen as
344 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
Figure 26.2. An example of broadcast-ready Climate Mat- by current practices. This iconic species
ters graphics that are produced by Climate Central and becomes the mascot of their campaign, a
distributed to TV weathercasters. (Courtesy of Climate
Central.)
campaign that embraces and embodies fun,
and offers people immediate benefits in the
requiring costs up front while delivering form of positive reinforcement for partici-
benefits only in the future, possibly the dis- pating and the satisfaction of contributing
tant future. to something of value to entire community.
To enhance the odds that people will
adopt behaviors they have already decided
The Importance of Making the Behavior (at
are in their best interest, climate change
Least Seem) Popular
and biodiversity professionals should con-
sider two important questions: What can I People are highly sensitive to social norms.
do to make the behavior easier to perform? The more common (or normative) a behav-
And what can I do to help decision makers ior is perceived to be, the more likely people
get immediate benefits from the behavior, are to perform it (Cialdini 2006). There are
especially benefits they care most about? two distinct types of social norms, both of
Rare, a biodiversity conservation and be- which exert subtle but powerful influences
havior change organization—based in the on people’s behavior. Descriptive norms
United States but working worldwide—pro- are people’s perceptions of how common
vides a great example. Rare has developed a a behavior (or attitude) is among people
highly successful model for enhancing fun, like themselves (e.g., friends in their social
making recommended behaviors easier, and network, members of their “tribe,” and/
delivering valued benefits immediately— or citizens of their community). Injunc-
called “pride campaigns” (Boss 2008; Butler tive norms, conversely, are people’s percep-
et al. 2013). Pride can be a powerful motiva- tions of the degree to which other relevant
tor (Patrick et al. 2009). Rare’s approach cen- people (friends, “tribal” members, citizens)
ters on cultivating local pride in a commu- approve or disapprove approve of the be-
nity or in a nation—in their land, in their havior or attitude.
culture, and in their willingness to rally be- The most useful way to harness the in-
hind a local iconic species that is threatened fluence of social norms depends, in part, on
I N C R E A S I N G P U B L I C AWA R E N E S S A N D FAC I L I TAT I N G B E H AV I O R C H A N G E 345
the degree to which a behavior being pro- average households subsequently decreased
moted is currently normative. When seek- their energy use, and below-average house-
ing to promote a behavior that is currently holds subsequently increased their energy
uncommon among members of the target use, thereby demonstrating the power of
audience, one can draw attention to specific descriptive norms. The boomerang effect
notable people who are already performing among below-average households was pre-
the behavior, to their reasons for perform- vented, however, by adding a smiley-face
ing the behavior, and to the benefits they image on the utility bill to signal the util-
are enjoying as a result. Shining a light on ity company’s approval of energy conser-
these behavioral models makes the behavior vation—thereby demonstrating the power
appear more descriptively normative than of injunctive norms. This simple insight
it might otherwise seem, and as described about the power of social norms has led to
above, it can also highlight the behavior’s a thriving business that is helping utility
benefits and promote self-efficacy among companies in a half dozen nations reduce
decision makers who see the modeling. their need to generate electricity.
Uncommon behaviors can quickly be- In conclusion, communications efforts
come popular when opinion leaders within that use simple, clear messages, repeated of-
a target audience embrace and endorse the ten by a variety of trusted sources, and be-
behavior, thereby exerting their powerful havior change efforts that strive to make the
social influence through injunctive norms. behavior you are promoting easy, fun, and
Sustainability professionals can seek out popular hold considerable promise in help-
and recruit opinion leaders in their target ing translate the insights of environmental
population as a strategy for accelerating up- science into more sustainable civilizations
take of behaviors they are recommending across the globe.
(Valente 2012).
If the recommended behavior is gaining
in popularity but is not yet normative, ef- ACK NOWLE DGME NTS
forts can be made to highlight its growing
popularity—in the news, in entertainment This material is based on work supported
media, and in social media—as a means of by the National Science Foundation under
reinforcing the growing norm. Such efforts Grant Numbers DRL-1422431 and DRL-
are particularly likely to be effective when 1713450. Any opinions, findings, and con-
they highlight notable respected individu- clusions or recommendations expressed in
als who are embracing the behavior (e.g., this material are those of the author(s) and
Warren Buffett), especially if most people do not necessarily reflect the views of the
would not expect those individuals to em- National Science Foundation.
brace the behavior.
Opower provides an excellent ex-
ample. Opower is an American corpora-
R E FE R E NCE S
tion that harnesses social science research
on the power of social norms to help Bandura, Albert. 2004. “Health promotion by social cog-
utility companies reduce consumer de- nitive means.” Health Education and Behavior 31: 143–164.
mand for electricity (http://opower.com/ Batra, Rajeev, and Michael Ray. 1986. “Situational effects
designprinciples). A study by some of of advertising repetition: The moderating influence of
the company’s behavioral science advis- motivation, ability and opportunity to respond.” Jour-
nal of Consumer Research 12: 432–445.
ers (Schultz et al. 2007) found that when Boss, Suzie. 2008. “The cultural touch: Environmental
shown on their monthly utility bill their non-profit RARE tailors its programs to local cultures
own energy use data relative to the neigh- and needs.” Stanford Social Innovation Review. https://ssir
borhood’s average energy use, above- .org/articles/entry/the _cultural _touch.
346 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
Butler, Paul, Kevin Green, and Dale Galvin. 2013. The Health Education, 3rd ed., ed. K. Glanz, Barbara Rimer,
Principles of Pride: The Science Behind the Mascots. Rare. http:// and Fran Marcus Lewis, 437–461. Jossey-Bass.
rare.org/publications. Maibach, Edward, Lorien Abroms, and Mark Maros-
Chong, Dennis, and James Druckman. 2013. “Counter- its. 2007. “Communication and marketing as tool to
framing effects.” Journal of Politics 75: 1–16. cultivate the public’s health: A proposed ‘people and
Cialdini, Robert. 2006. Influence: The Psychology of Persuasion. places’ framework.” BMC Public Health 7: 88.
William Morrow. Marshall, G. 2014. Don’t Even Think about It: Why Our Brains
Crow, Desiree, and Maxwell Boykoff. 2014. Culture, Poli- Are Wired to Ignore Climate Change. Bloomsbury.
tics and Climate Change: How Information Shapes Our Common McKenzie-Mohr, Doug. 2011. Fostering Sustainable Behavior.
Future. Routledge. 3rd. ed. New Society Publishers.
Fischhoff, Baruch. 1989. “Risk: A guide to controversy.” McKenzie-Mohr, Doug, Nancy Lee, Wesley Schultz, and
In Improving Risk Communication, ed. Institute of Medi- Phillip Kotler. 2012. Social Marketing to Protect the Environ-
cine, 211–319. National Academies Press. ment. SAGE Publications.
Fischhoff, Baruch. 2007. “Non-persuasive commu- Moser, Suzanne, and Lisa Dilling. 2007. Creating a Climate
nication about matters of the greatest urgency: Cli- for Change. Cambridge University Press.
mate change.” Environmental Science and Technology 41: Patrick, Vanessa, HaeEun Chun, and Deborah MacInnis.
7204–7208. 2009. “Affective forecasting and self-control: When
Heath, Chip, and Dan Heath. 2010. Switch: How to Change anticipating pride wins over anticipating shame in a
Things When Change Is Hard. Broadway Books. self-regulation context.” Journal of Consumer Psychology 19:
Holmes, John, and Rebecca Clark. 2008. “Enhancing the 537–545.
use of science in environmental policy-making and Pechman, Cornelia, and David Stewart. 1988. “Advertis-
regulation.” Environmental Science and Policy 11: 702–711. ing repetition: A critical review of wear-in and wear-
Hornik, Robert. 2002. Public Health Communication: Evidence out.” Critical Issues and Research in Advertising 11: 285–329.
for Behavior Change. Lawrence Erlbaum Associates. Pew Research Center. Public Esteem for Military Still High.
Lang, Annie. 2013. “Discipline in crisis? The shifting http:// www .pewforum .org/ 2013/ 07/ 11/ public
paradigm of mass communication research.” Commu- -esteem-for-military-still-high/.
nication Theory 23: 10–24. Placky, Bernadette, Edward Maibach, Joe Witte, Bud
Leiserowitz, Anthony. 2005. “American risk percep- Ward, Keith Seitter, Ned Garniner, David Herring,
tions: Is climate change dangerous?” Risk Analysis 23: and Heidi Cullen. 2015. “Climate Matters: A compre-
1433–1442. hensive educational resource for broadcast meteorolo-
Leiserowitz, Anthony, Edward Maibach, and Con- gists.” Bulletin of the American Meteorological Society. http://
nie Roser-Renouf. 2009. Climate Change in the Ameri- dx.doi.org/10.1175/BAMS-D-14-00235.1.
can Mind: American’s Climate Change Beliefs, Attitudes, Policy Rothschild, Michael. 1999. “Carrots, sticks and prom-
Preferences, and Actions. Yale Project on Climate Change ises: A conceptual framework for the management of
Communication. public health and social issue behaviors.” Journal of Mar-
Leiserowitz, Anthony, Edward Maibach, Connie Roser- keting 63: 24–37.
Renouf, Seth Rosenthal, Matt Cutler, and John Kotcher. Ryder, Dennis, Moya Tomlinson, Ben Gawne, and Gene
2017. “Climate change in the American mind: Octo- Likens. 2010. “Defining and using ‘best available sci-
ber, 2017.” Yale Project on Climate Change Commu- ence’: A policy conundrum for the management of
nication. https://www.climatechangecommunication aquatic ecosystems.” Marine and Freshwater Research 61:
.org/wp -content/uploads/2017/11/Climate -Change 821–828.
-American-Mind-October-2017-min.pdf. Schultz, Wesley, Jessica Nolan, Robert Cialdini, Noah
Maibach, Edward, Bernadette Placky, Joe Witte, Keith Goldstein, and Vladas Griskevicius. 2007. “The con-
Seitter, Ned Gardiner, Teresa Myers, Sean Sublette, structive, destructive and reconstructive power of so-
and Heidi Cullen. 2016. “TV meteorologists as local cial norms.” Psychological Science 18: 429–434.
climate educators.” Oxford Research Encyclopedia, Climate Sci- Smith, William. 2011. Reinventing Social Marketing. TedX
ence. https://doi.org/10.1093/acrefore/9780190228620 Penn Quarter. https://www.youtube.com/watch?v=
.013.505. IECY9LJvTf4.
Maibach, Edward, and Vincent Covello. 2016. “Commu- Valente, Tom. 2012. “Network interventions.” Science 337:
nicating environmental health.” In Environmental Health: 49–53.
From Global to Local, 3rd ed., ed. Howard Frumkin, 769– Whitmarsh, Lorraine, Sophie O’Neill, and Irene Lo-
791. Jossey-Bass. renzoni. 2011. Engaging the Public with Climate Change.
Maibach, Edward, Michael Rothschild, and William No- Earthscan.
velli. 2002. “Social marketing.” In Health Behavior and
CHAPTER TWENTY-SEVEN I N T RODUCTION
347
348 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
the world. There is evidence that agricul- ture crop yield: statistical models relying on
tural production is already being substan- historical yield and weather data, and bio-
tially and negatively affected by changes in physical or process-based models relying on
climate, and projections based on climate data from experimental crop trials (Porter
models indicate that later in this century et al. 2014; Lobell and Burke 2010). A recent
many farming areas will have climates not global meta-level study that synthesized
experienced during the entire history of ag- projections from more than 1,700 pub-
riculture. We are seeing an increasing mis- lished simulations to evaluate yield impacts
match between our crops and the human- of climate change concluded: “There is a
altered environment (Carroll et al. 2014). majority consensus that yield changes will
One in eight persons on Earth today, or be negative from the 2030s onwards. More
more than 795 million in total, are already than 70 percent of the projections indicate
considered food insecure (FAO, IFAD, and yield decreases for the 2040s and 2050s, and
WFP 2016). Most undernourished people more than 45 percent of all projections for
live in South and East Asia and Sub-Saharan the second half of the century indicate yield
Africa. Although the challenge of adapting decreases greater than 10 percent” (Challi-
agriculture to new environmental circum- nor et al. 2014, 289). The picture emerging
stances will be global, posing an obstacle to from these models is dire. At a global aggre-
development goals at all levels, it will take gate level, yields of maize and wheat begin
on a particularly urgent humanitarian as- to decline with 1°C–2°C of warming in the
pect where the poor and food insecure are tropics, while temperate maize and tropi-
concentrated, where the impacts of climate cal rice yields are significantly affected with
change on food production tragically will warming of 3°C–5°C (Porter et al. 2014).
be most severe. These global estimates mask considerable
regional differences, however. Studies pro-
jecting positive effects on crop production
IM PACTS in temperate regions buffer some of the an-
ticipated decreases in yields in tropical re-
Crop yield is the most studied aspect of the gions for which there is strong consensus
impacts of climate change on food security. (Challinor et al. 2014). But the latest report
A large number of studies document that ef- from the International Panel of Climate
fects are already evident in several regions Change (IPCC) (Porter et al. 2014) says it is
and project increasingly negative effects on now less likely that moderate warming will
yield. Climate extremes including unusually raise crop yields at mid- to high latitudes
hot night- and daytime temperatures as well than projected in their previous assessment.
as natural disasters attributed to anthropo- Instead, the IPPC projects that there will be
genic activity have already negatively affected more yield decreases than increases even
overall crop production (Porter et al. 2014). A under moderate warming.
study at the International Rice Research Insti- Sub-Saharan Africa is emerging as a re-
tute (IRRI) found that a 1°C increase in min- gion that will be particularly hard hit by
imum temperature during the dry growing climate change. On the basis of national
season had the effect of reducing rice yields historical climate and crop production data,
by 10 percent (Peng et al. 2004). Lobell et Schlenker and Lobell (2010) projected ag-
al. (2011) found that global maize production gregate production losses of 17 percent for
declined 3.8 percent and wheat declined 5.5 sorghum and 22 percent for maize in the re-
percent between 1980 and 2008 compared gion by midcentury (Schlenker and Lobell
to a counterfactual without climate change. 2010). Lobell et al. (2011) analyzed experi-
There are two principal types of studies mental yield data from 20,000 maize trials
estimating effects of climate change on fu- across South and East Africa and projected
CLIM ATE CH A NGE, FOOD, A ND BIODIV ER SIT Y 349
that 1°C warming will lead to yield losses precedented climate territory. New cli-
in 65 percent–100 percent of the regions’ mates will generate combinations of con-
maize growing areas. They found signifi- ditions and biological assemblages with
cant varietal differences depending on the no current analogues (Burke, Lobell, and
maturity period of the varieties and on Guarino 2009). Changes in climate ex-
whether the variety is hybrid or open pol- tremes can have more adverse impacts on
linated (Lobell et al. 2011). At a local level crop production than changes in means.
in this region, Thornton et al. (2009) used Moreover, today’s extremes might become
a biophysical crop model to estimate effects tomorrow’s normal—today’s hot growing
in grids corresponding to ~18 km by 18 km seasons, which often result in large de-
in the landscape and found that between 9 clines in food production, could become
percent and 33 percent of the assessed grids the cool and “best” growing seasons of
across East Africa are likely to see a potential the future (Figure 27.1). The damage done
yield loss of more than 20 percent for maize by short-run events could become long-
by 2050. term trends without sufficient invest-
Impact studies generally agree that the ments in adaptation (Battisti and Naylor
effect of temperature increase will be more 2009). A recent statistical study of global
important than changes in precipitation crop production and extreme weather
and that the benefits of increased CO2 in events found that droughts and extreme
the atmosphere for photosynthesis will be heat have reduced national cereal produc-
outweighed by the negative effects of tem- tion by 9 percent–10 percent and 8 per-
perature increase (Lobell and Gourdji 2012; cent–11 percent more damage in devel-
Thornton et al. 2009). Excessive heat can oped countries than in developing ones
harm virtually all plant parts and affect plant (Lesk, Rowhani, and Ramankutty 2016).
growth at every stage in the life cycle. The This is dire evidence in light of the esti-
life cycle and growth period are shortened. mate that by 2050, the majority of African
Leaves curl and wither, reducing light inter- countries will experience climates over at
ception. Partial closure of stomata to reduce least half their current crop area that lie
transpiration increases canopy temperature outside the range currently experienced
and reduces photosynthesis effectiveness. within the country (Burke, Lobell, and
The quantity, quality, and timing of growth Guarino 2009).
in reproductive tissue are reduced (Cairns et Evidence presented in previous chapters
al. 2013; Cossani and Reynolds 2012). Thus, of this book suggests substantial risks to
negative effects of heat both cause and com- wild plant biodiversity due to range shifts
pound already existing moisture deficien- and contraction in habitats with suitable
cies. Most yield projections omit consid- climates. In a seminal study, Thomas et al.
eration of pest, weed, and disease impacts (2004) predicted that 15 percent–37 per-
on crop yields, but several studies conclude cent of wild plant biodiversity is threatened
that changes in climate and CO2 concentra- with extinction due to climate change by
tion will alter the distribution and probably 2050. Similar projections have been made
increase the competitiveness of nematodes, for crop wild relatives. Jarvis et al. (2008)
insects, fungal diseases, and weeds in many estimates that up to 61 percent of wild pea-
areas (Ziska et al. 2011; Porter et al. 2014). nut (Arachis) species, 12 percent of potato
Effects on other biotic factors important for (Solanum) species, and 8 percent of cowpea
crop production such as on soil communi- (Vigna) species could become extinct within
ties are underresearched but also represent 50 years due to climate change. Other eco-
potential threats to agricultural production. logical niche modeling studies have pro-
At 2°C of global warming, many crops jected large contractions in distribution
will begin to enter a new, historically un- areas for wild relatives in the Cucurbitacea
350 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
Number of Summers
2080-2100 2080-2100
30 30
20 20
10 10
0 0
-5 0 5 10 -5 0 5 10
Temperature Anomaly (ºC) Temperature Anomaly (ºC)
Figure 27.1. Historic and future average growing condi- wheat and maize caused by already evident
tions in Peru and South Africa. (Data from Battisti and climate change between 1980 and 2008 re-
Naylor 2009.)
sulted in an estimated average commodity
price increase of 6.9 percent compared to
family (to which pumpkins and squashes a counterfactual with no climate change in
belong) in Mexico (Lira, Téllez, and Dávila the period (Lobell, Schlenker, and Costa-
2009), on wild relatives of maize (Zea) in Roberts 2011). Projecting this trend into the
Mexico (Ureta et al. 2012), and wild Arabica future, the IPCC finds that changes in tem-
coffee (Coffea arabica) in Ethiopia (Davis et al. perature and precipitation, without con-
2012). Like wild plant biodiversity, crop sidering effects of CO2, will contribute to
wild relatives will face many compounding increased global food prices in the range of
influences such as habitat degradation due 3 percent–84 percent by 2050 (Porter et al.
to agriculture and human infrastructure, 2014).
and shifts in pests and disease stress. Periods of abnormally hot weather in
the past provide an imperfect but important
glimpse of how we might expect agricul-
C L I M AT E C H A NGE A N D F O OD tural and social systems to react in the fu-
SE CU R IT Y ture unless food production keeps pace with
the growing demand. Significant climate
Future food security will be affected not events due primarily, it is argued, to their
only by crop production levels but also by a impact on food production have been posi-
host of socioeconomic factors that make it tively correlated with increased incidents
difficult to link climate change unambigu- of war and civil strife over the past 2,000
ously to impacts on food security (Lobell years (Büntgen et al. 2011; Hsiang, Burke,
and Gourdji 2012; HLPE 2012). Neverthe- and Miguel 2013). As for food security, the
less, the IPCC highlights that climate change effect of climate change on human security
is likely to affect all aspects of food security, will heavily depend on political and eco-
not only food availability, but also food ac- nomic pathways taken (Theisen, Gleditsch,
cess, utilization, and price stability (Porter and Buhaug 2013), but the link between cli-
et al. 2014). Nutritional quality of food and mate-caused food scarcity and social insta-
fodder (important for the utilization aspect bility is arguably an underlying trend in hu-
of food security) is negatively affected by man history. Analyzing new conflict event
elevated CO2 (Myers et al. 2014; Porter et data covering Asia and Africa in the period
al. 2014). Climate extremes have, together 1989–2014, von Uexkull et al. (2016) found
with other factors, played a role in peri- that for groups depending on agriculture
ods of rapid food and cereal price increases living in poor countries and experiencing
(important for access and stability aspects of political exclusion, droughts significantly
food security) in recent years. Yield losses in increased the likelihood of violence.
CLIM ATE CH A NGE, FOOD, A ND BIODIV ER SIT Y 351
PR E PA R I NG F OR C L IM AT E C H A NGE aptation as they are “based on long experi-
T H ROUGH C ROP A DA P TATION ence in dealing with a highly variable en-
vironment” (Vermeulen et al. 2013), while
As the twenty-first century progresses, a those involving improved varieties are
central question becomes whether, to what classified either as incremental or systemic
extent, and through which means agri- change, depending on whether the variet-
culture will successfully adapt to climate ies are available off the shelves or are devel-
change. This question is critical to food oped through targeted breeding programs.
security, to global peace, and to the envi- The cardinal question is whether farmers
ronment, because agricultural systems that and breeders will be able to adapt based
function poorly are a threat to all. on the genetic variability in traditional and
Challinor et al. (2014) assess several improved varieties. Early research into this
common adaptation strategies in agricul- question indicates that although there is
ture: cultivar adjustment, planting date ad- great genetic potential in available genetic
justment, and irrigation and fertilizer opti- resources, there are challenges regarding
mization. Cultivar adjustment is estimated access to the appropriate genetic resources,
to be the most effective measure with a me- both for farmers and for breeders (Burke,
dian potential to increase yield by 23 per- Lobell, and Guarino 2009; Bellon, Hod-
cent compared with situations where farm- son, and Hellin 2011; Bjørnstad, Tekle, and
ers do not change cultivars (Figure 27.2). Göransson 2013).
Cultivar adjustment refers to switching Adaptation of crops to warmer new cli-
to a presumably better adapted variety, be mates will not be reliably—and certainly not
it a local or an improved variety and can
involve both formal and informal seed sup- Figure 27.2. The percentage benefit (yield difference be-
tween cases with and without the adaptation) for differ-
ply systems (Westengen and Brysting 2014). ent crop management adaptations: cultivar adjustment
Recent typologies of adaptation distinguish (CA); planting date adjustment (PDA); adjusting plant-
between incremental, systemic, and trans- ing date in combination with cultivar adjustment (PDA,
formative adaptation options (Vermeulen CA); irrigation optimization (IO); fertilizer optimiza-
et al. 2013; Porter et al. 2014; Rippke et al. tion (FO); other management adaptations (Other). The
simulated median benefit is marked with a solid, dark
2016) In this typology, cultivar adjustment gray, horizontal line and the whiskers indicate the 25th
involving local varieties and informal seed and 75th percentile. (Based on data from Challinor et al.
supply systems exemplifies incremental ad- 2014; Porter et al. 2014.)
352 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
avoid significant increases in hunger, mal- Challinor, Andrew J., A.-K. Koehler, Julian Ramirez-Vil-
nutrition, political instability, and environ- legas, S. Whitfield, and B. Das. 2016. “Current warm-
ing will reduce yields unless maize breeding and seed
mental destruction. Additional investments systems adapt immediately.” Nature Climate Change 6
will need to be made in crop diversity col- (10): 954–958.
lection and conservation, in the screening Cossani, C. Mariano, and Matthew P. Reynolds. 2012.
of collections for useful traits, and particu- “Physiological traits for improving heat tolerance in
larly in plant-breeding programs, including wheat.” Plant Physiology 160 (4): 1710–1718.
Davis, Aaron P., Tadesse Woldemariam Gole, Susana
for those crops for which there is little or Baena, and Justin Moat. 2012. “The impact of climate
no effort under way presently. Such efforts change on indigenous arabica coffee (Coffea arabica):
are a basic prerequisite for crop adaptation. Predicting future trends and identifying priorities.”
If crops do not successfully adapt to climate PLOS One 7 (11): e47981.
change, neither will agriculture and neither FAO. 2010. The Second Report on the State of the World’s Plant Ge-
netic Resources. FAO.
will we. FAO. 2014. “FAOSTAT.” http://faostat3.fao.org/home/
index.html.
FAO, IFAD, and WFP. 2016. The State of Food Insecurity in the
R E FE R E NCE S World: Meeting the 2015 International Hunger Targets: Taking
Stock of Uneven Progress. Food and Agriculture Organi-
Battisti, David S., and Rosamond L. Naylor. 2009. “Histor- zation, International Fund for Agricultural Develop-
ical warnings of future food insecurity with unprec- ment, World Food Program.
edented seasonal heat.” Science 323 (5911): 240–244. Fowler, C., and T. Hodgkin. 2004. “Plant genetic re-
Bellon, M. R., D. Hodson, and J. Hellin. 2011. “Assess- sources for food and agriculture: Assessing global
ing the vulnerability of traditional maize seed systems availability.” Annual Review of Environment and Resources 29:
in Mexico to climate change.” Proceedings of the National 143–179. https://doi.org/10.1146/annurev.energy.29
Academy of Sciences 108 (33): 13432–13437. .062403.102203.
Bjørnstad, Åsmund, Selamawit Tekle, and Magnus Fowler, Cary. 1994. Unnatural Selection: Technology, Politics, and
Göransson. 2013. “‘Facilitated access’ to plant genetic Plant Evolution. Vol. 6. Gordon and Breach Yverdon.
resources: Does it work?” Genetic Resources and Crop Evolu- Fowler, Cary. 2008. “The Svalbard seed vault and crop
tion 60 (7): 1959–1965. security.” BioScience 58 (3): 190–191.
Burke, M. B., D. B. Lobell, and L. Guarino. 2009. “Shifts High Level Panel of Experts. 2012. Food Security and Climate
in African crop climates by 2050, and the implications Change: A Report by the High Level Panel of Experts on Food Secu-
for crop improvement and genetic resources conser- rity and Nutrition of the Committee on World Food Security. FAO.
vation.” Global Environmental Change: Human and Policy Di- Hsiang, Solomon M., Marshall Burke, and Edward
mensions 19 (3): 317–325. https://doi.org/10.1016/j Miguel. 2013. “Quantifying the influence of climate
.gloenvcha.2009.04.003. on human conflict.” Science 341 (6151): 1235367.
Büntgen, Ulf, Willy Tegel, Kurt Nicolussi, Michael Mc- Jarvis, Andy, Annie Lane, and Robert J. Hijmans. 2008.
Cormick, David Frank, Valerie Trouet, Jed O. Kaplan, “The effect of climate change on crop wild relatives.”
Franz Herzig, Karl-Uwe Heussner, and Heinz Wanner. Agriculture Ecosystems and Environment 126 (1–2): 13–23.
2011. “2500 years of European climate variability and https://doi.org/10.1016/j.agee,2008.01.013.
human susceptibility.” Science 331 (6017): 578–582. Khoury, Colin K., Harold A. Achicanoy, Anne D. Bjork-
Cairns, Jill E., Jose Crossa, P. H. Zaidi, Pichet Grudloyma, man, Carlos Navarro-Racines, Luigi Guarino, Ximena
Ciro Sanchez, Jose Luis Araus, Suriphat Thaitad, Dan Flores-Palacios, Johannes M. M. Engels, John H. Wi-
Makumbi, Cosmos Magorokosho, and Marianne Bän- ersema, Hannes Dempewolf, and Steven Sotelo. 2016.
ziger. 2013. “Identification of drought, heat, and com- “Origins of food crops connect countries worldwide.”
bined drought and heat tolerant donors in maize.” Crop Proceedings of the Royal Society B 283: art. 20160792.
Science 53 (4): 1335–1346. Larson, Greger, Dolores R. Piperno, Robin G. Allaby, Mi-
Carroll, Scott P., Peter Sogaard Jorgensen, Michael T. chael D. Purugganan, Leif Andersson, Manuel Arroyo-
Kinnison, Carl T. Bergstrom, R. Ford Denison, Peter Kalin, Loukas Barton, Cynthia Climer Vigueira, Tim
Gluckman, Thomas B. Smith, Sharon Y. Strauss, and Denham, and Keith Dobney. 2014. “Current perspec-
Bruce E. Tabashnik. 2014. “Applying evolutionary bi- tives and the future of domestication studies.” Proceed-
ology to address global challenges.” Science 346 (6207): ings of the National Academy of Sciences 111 (17): 6139–6146.
art. 1245993. Lesk, Corey, Pedram Rowhani, and Navin Ramankutty.
Challinor, A. J., J. Watson, D. B. Lobell, S. M. Howden, 2016. “Influence of extreme weather disasters on
D. R. Smith, and N. Chhetri. 2014. “A meta-analysis global crop production.” Nature 529 (7584): 84–87.
of crop yield under climate change and adaptation.” Lira, Rafael, Oswaldo Téllez, and Patricia Dávila. 2009.
Nature Climate Change 4 (4): 287–291. “The effects of climate change on the geographic dis-
CLIM ATE CH A NGE, FOOD, A ND BIODIV ER SIT Y 355
tribution of Mexican wild relatives of domesticated Rippke, U., J. Ramirez-Villegas, A. Jarvis, S. J. Vermeu-
Cucurbitaceae.” Genetic Resources and Crop Evolution 56 (5): len, L. Parker, F. Mer, B. Diekkrüger, A. J. Challinor,
691–703. and M. Howden. 2016. “Timescales of transforma-
Lobell, D. B., and M. B. Burke. 2010. “On the use of sta- tional climate change adaptation in sub-Saharan Afri-
tistical models to predict crop yield responses to cli- can agriculture.” Nature Climate Change 6: 605–609.
mate change.” Agricultural and Forest Meteorology 150 (11): Schlenker, W., and D. B. Lobell. 2010. “Robust negative
1443–1452. impacts of climate change on African agriculture.” En-
Lobell, D. B., M. Bänziger, C. Magorokosho, and B. Vi- vironmental Research Letters 5 (1): 014010. https://doi.org/
vek. 2011. “Nonlinear heat effects on African maize 10.1088/1748–9326/5/1/014010.
as evidenced by historical yield trials.” Nature Climate Theisen, Ole Magnus, Nils Petter Gleditsch, and Halvard
Change 1 (1): 42–45. Buhaug. 2013. “Is climate change a driver of armed
Lobell, David B., and Sharon M. Gourdji. 2012. “The in- conflict?” Climatic Change 117 (3): 613–625.
fluence of climate change on global crop productiv- Thomas, Chris D., Alison Cameron, Rhys E. Green, Mi-
ity.” Plant Physiology 160 (4): 1686–1697. chel Bakkenes, Linda J. Beaumont, Yvonne C. Collin-
Lobell, David B., Wolfram Schlenker, and Justin Costa- gham, Barend F. N. Erasmus, Marinez Ferreira De
Roberts. 2011. “Climate trends and global crop pro- Siqueira, Alan Grainger, and Lee Hannah. 2004. “Ex-
duction since 1980.” Science 333 (6042): 616–620. tinction risk from climate change.” Nature 427 (6970):
McCouch, Susan, Gregory J. Baute, James Bradeen, Paula 145–148.
Bramel, Peter K. Bretting, Edward Buckler, John M. Thornton, P. K., P. G. Jones, G. Alagarswamy, and J. An-
Burke, David Charest, Sylvie Cloutier, and Glenn Cole. dresen. 2009. “Spatial variation of crop yield response
2013. “Agriculture: Feeding the future.” Nature 499 to climate change in East Africa.” Global Environmental
(7456): 23–24. Change: Human and Policy Dimensions 19 (1): 54–65. https://
McGuire, Shawn, and Louise Sperling. 2013. “Making doi.org/10.1016/j.gloenvcha.2008.08.005.
seed systems more resilient to stress.” Global Environmen- Ureta, Carolina, Enrique Martínez-Meyer, Hugo R. Pera-
tal Change: Human and Policy Dimensions 23 (3): 644–653. les, and Elena R. Álvarez-Buylla. 2012. “Projecting the
Myers, Samuel S., Antonella Zanobetti, Itai Kloog, Pe- effects of climate change on the distribution of maize
ter Huybers, Andrew D. B. Leakey, Arnold J. Bloom, races and their wild relatives in Mexico.” Global Change
Eli Carlisle, Lee H. Dietterich, Glenn Fitzgerald, and Biology 18 (3): 1073–1082.
Toshihiro Hasegawa. 2014. “Increasing CO2 threatens Vermeulen, Sonja J., Andrew J. Challinor, Philip K.
human nutrition.” Nature 510 (7503): 139–142. Thornton, Bruce M. Campbell, Nishadi Eriyagama,
Peng, Shaobing, Jianliang Huang, John E. Sheehy, Re- Joost M. Vervoort, James Kinyangi, Andy Jarvis, Pe-
becca C. Laza, Romeo M. Visperas, Xuhua Zhong, ter Läderach, and Julian Ramirez-Villegas. 2013. “Ad-
Grace S. Centeno, Gurdev S. Khush, and Kenneth G. dressing uncertainty in adaptation planning for agri-
Cassman. 2004. “Rice yields decline with higher night culture.” Proceedings of the National Academy of Sciences 110
temperature from global warming.” Proceedings of the Na- (21): 8357–8362.
tional Academy of Sciences 101 (27): 9971–9975. Von Uexkull, N., M. Croicu, H. Fjelde, and H. Buhaug.
Plucknett, Donald, Nigel J. H. Smith, J. T. Williams, and 2016. “Civil conflict sensitivity to growing-season
N. Murthi Anishetty. 1987. Gene Banks and the World’s Food. drought.” Proceedings of the National Academy of Sciences 113:
Los Baños, Philippines: International Rice Resource 12391–12396.
Institute. Westengen, Ola T., and Anne K. Brysting. 2014. “Crop
Porter, J. R., L. Xie, A. J. Challinor, K. Cochrane, S. M. adaptation to climate change in the semi-arid zone in
Howden, M. M. Iqbal, D. B. Lobell, and M. I. Travasso. Tanzania: The role of genetic resources and seed sys-
2014. “Food security and food production systems.” tems.” Agriculture and Food Security 3 (1): 3.
In Climate Change 2014: Impacts, Adaptation, and Vulnerabil- Westengen, Ola T., Simon Jeppson, and Luigi Guarino.
ity. Part A: Global and Sectoral Aspects. Contribution of Working 2013. “Global ex-situ crop diversity conservation and
Group II to the Fifth Assessment Report of the Intergovernmental the Svalbard Global Seed Vault: Assessing the current
Panel on Climate Change, ed. C. B. Field, V. R. Barros, D. status.” PLOS One 8 (5): e64146.
J. Dokken, K. J. Mach, M. D. Mastrandrea, T. E. Bilir, Ziska, Lewis H., Dana M. Blumenthal, G. Brett Runion,
M. Chatterjee, K. L. Ebi, Y. O. Estrada, R. C. Genova, E. Raymond Hunt Jr., and Hilda Diaz-Soltero. 2011.
B. Girma, E. S. Kissel, A. N. Levy, S. MacCracken, P. “Invasive species and climate change: an agronomic
R. Mastrandrea, and L. L. White, 485–533. Cambridge perspective.” Climatic Change 105 (1–2): 13–42.
University Press.
CHAPTER TWENTY-EIGHT I N T RODUCTION
356
S AV I N G B I O D I V E R S I T Y I N T H E E R A O F H U M A N - D O M I N AT E D E C O S Y S T E M S 357
a larger suite of species increases the pro- ricultural land clearing and land clearing
ductivity, stability, and carbon capture and for logging, as well as hunting and climate
storage abilities of ecosystems. It is for these change (Figure 28.2).
reasons that the loss of biodiversity threat- Moreover, global demands for food and
ens the ability of ecosystems to provide hu- energy seem likely to continue increasing
manity with valued services. until the end of this century, though at a
Most of humanity’s threats to biodiver- much slower rate than in the last century.
sity, and most of our other negative envi- The UN projects that global population
ronmental impacts, are the inadvertent af- will reach 9.7 billion people by 2050 and
fects of how we meet the food and energy will then level off at about 11.2 billion by
demands of the global population (Matson 2100. Although economic growth is slow-
et al. 1997). These environmental impacts ing in richer nations, the vast majority of
have been dramatically escalating since the global population lives in nations that
about 1900 because of massive growth in currently are poor but have economies that
both global population and per capita con- are growing rapidly. When projections of
sumption. From 1900 to 2015 global popu- global economic growth and population
lation increased from 1.7 billion to 7.3 bil- growth are combined, total global eco-
lion and global per capita purchasing power nomic activity may be about four times
increased from about $700 to $8,500. This higher in 2050, and eight times higher in
50-fold increase in the global human econ- 2100, than it was in 2015. Unless policies
omy (i.e., [7.3 × 8,500]/[1.7 × 700] = 52) are implemented that decrease agricultural
has greatly increased global demand for land clearing, logging, excessive hunting,
food and energy. and fossil-fuel combustion, the extinction
The net effect of providing food and en- risks faced by animal and plant species
ergy to the global population is that more are likely to greatly increase throughout
than 4,200 vertebrate species are now the rest of this century. Such policies, and
highly threatened with extinction (Figure their scientific bases, are the focus of this
28.1A). In particular, according to analy- chapter.
ses done by the International Union for the
Conservation of Nature (IUCN), 16 percent
of all bird and mammal species are either Figure 28.1. (A) The number of vertebrate animal species
endangered or critically endangered with that have been driven extinct by human actions, or that
extinction, as are 7 percent of amphib- are classified as critically endangered with extinction or
as endangered, by the IUCN Red List. (B) The percentage
ians and reptiles (Figure 28.1B). The ma- of all known species that are extinct or threatened in
jor causes of these extinction risks are ag- each the four groups, based on the IUCN Red List.
1,500 A 20% B
Global Status Global Status
Percent of Total Group Diversity
Endangered Endangered
1,250 Critically Endangered Critically Endangered
Extinct Extinct
Number of Species
15%
1,000
750 10%
500
5%
250
0 0%
Mammals Birds Amphibians Reptiles Mammals Birds Amphibians Reptiles
358 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
Figure 28.2. Four of the major factors that threaten ver- but abandoned during the past half century.
tebrate species and the number of species threatened by
Moreover, because global crop demand is
each factor, based on the IUCN Red List.
increasing more rapidly than yields, about
E XTI NCTION F ROM F O OD A N D 400 million–700 million additional hect-
E N E RGY PRODUCTION ares of cropland may be needed in the next
40 years to meet global food demand (Til-
People need food and energy; with 7 bil- man et al. 2011; Tilman and Clark 2014). Be-
lion people now, heading toward 11 billion cause this land clearing is mainly occurring
richer people by the end of the century, in high-biodiversity tropical nations (Gibbs
global demand for both food and energy et al. 2009), large numbers of additional
will increase greatly. Global demand for species are likely to face risks of extinction.
agricultural crops is forecast to increase by Habitat destruction causes extinctions by
70 percent–110 percent from 2005 to 2050 two different mechanisms. First, as habitat
(Tilman et al. 2011). About a third of this is destroyed, there is the possibility that
increase is from population growth; two- the act of destruction will lead to the direct
thirds is from dietary changes associated death of the last living members of a spe-
with higher incomes, especially increased cies. Such direct extinction by habitat de-
per capita meat consumption, the animal struction is most likely to affect rare species
feeds needed to produce it, and increased and to cause the extinction of widespread
caloric consumption. and abundant species only as larger and
Agricultural land clearing is a major larger proportions of a region are destroyed.
threat to biodiversity, with 8,530 species of The potential direct impact of habitat de-
mammals, birds, amphibians, and reptiles struction on biodiversity is often modeled
listed by the IUCN as threatened by this by using the species-area relationship or its
activity. Logging is also important, and it derivatives (Rosenzweig 1995; He and Hub-
is often the initial step toward the conver- bell 2011).
sion of forests into agricultural lands. Cur- Second, the fragmentation that com-
rently 4.9 billion hectares of land, which monly results from habitat destruction can
is 40 percent of the ice-free land area of cause the time-delayed extinction of those
Earth, is in agriculture, with 1.5 billion species that survived (e.g., Gibson et al.
hectares in croplands and the remainder in 2013). This has been termed the “extinction
pastures. An additional 0.5 billion hectares debt” (Tilman et al. 1994) and can lead to
had been cleared and used for agriculture the eventual extinction of even those spe-
S AV I N G B I O D I V E R S I T Y I N T H E E R A O F H U M A N - D O M I N AT E D E C O S Y S T E M S 359
cies that were, at the time of destruction, America successfully migrated toward the
some of the most abundant species in the equator when the last glacial period started,
fragments (Kuussaari et al. 2009). In partic- and then to the north when glaciers re-
ular, the species most susceptible to time- treated. However, because of barriers to
delayed extinction from fragmentation are migration, the glacial cycles contributed to
predicted to be species that are good local tree extinctions in Europe, thus partially
competitors but poor dispersers. The ex- explaining its low tree diversity relative to
tinctions from destruction and from frag- Asia (Latham and Ricklefs 1993). Massive
mentation are each predicted to be sharply human-caused habitat fragmentation has
increasing functions the amount of habi- literally changed the landscapes of all con-
tat that is destroyed (Tilman et al. 1994). tinents, imposing numerous barriers to the
Current theory suggests that the numbers migration of plants and animals. The com-
of species lost by direct extinction and the bination of rapid of human-driven climate
numbers lost by fragmentation may be change and barriers to migration may pose
about equal, but the actual spatial pattern- a severe threat to biodiversity. Finally, two
ing of destruction would have great influ- other ways that people obtain food, hunt-
ence on each type of extinction. ing and fisheries, also pose extinction risks
Climate change is a looming threat to for some types of prey species and espe-
biodiversity, and many species are already cially for other species, called bycatch in
negatively impacted by climate change. The fisheries, that are not directly targeted but
IUCN Red List has a total of 1,880 species of are inadvertently killed.
mammals, birds, amphibians, and reptiles In total, increasing global demand for
threatened by climate change and severe food and energy are greatly increasing the
weather events (Figure 28.2). As climate extinction risks facing life on Earth. Because
change accelerates, so will these threats. agriculture and fossil energy use pose such
Because per capita energy demand is an major risks for biodiversity, the reduction
increasing function of per capita incomes, or elimination of these threats will require
global energy demand is forecast to increase changes in agricultural and energy systems
at a greater rate than population (Interna- worldwide. Fortunately, many such changes
tional Energy Agency 2015). Whether this offer multiple benefits for biodiversity pres-
demand will be met by combustion of fossil ervation, climate stabilization, and pollu-
fuels or by renewable energy will depend tion reduction, and they can be achieved
on the adoption of global greenhouse gas using current knowledge and technology.
policies. Moreover, the greenhouse gas re- These changes provide a pathway toward a
leases from agriculture, which currently more sustainable Earth that would provide
account for 30 percent of annual emissions the food, energy, and livable environments
via CO2 from land clearing and soil tillage, that all of us, and all future generations,
methane from ruminants and rice, and ni- need.
trous oxide from nitrogen fertilizer, are
also increasing.
Climate change can make ecosystems M I N I M I Z I NG E XT I NC T IONS ON A
become unsuitable to the species living in H U M A N-D OM I NAT E D PL A N E T
them. During glacial cycles, most species
successfully migrated so as to remain in re- What will determine how many species sur-
gions with suitable climate, but the ability vive in remnant habitats around the world
of species to do so depended on there be- in 2050 and beyond? The greatest factors
ing few barriers to such migration. Regions are how we obtain food and energy, and
with barriers suffered extinctions. For in- how much of each we demand. There are
stance, most tree species in Asia and North five pathways, all of them feasible because
360 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
they use existing knowledge and technolo- in some of the most diverse ecosystems on
gies, each of which would reduce the risk Earth.
of a massive wave of extinctions. These five Because some cropland is taken out of
actions are outlined below. production each year because of conver-
sion to urban, suburban, transit, or other
uses and because of loss of fertility, the net
Prevent Extinctions by Increasing Yields
increase in global cropland is likely to be
of Developing Nations
less than the amount of land that is cleared.
Because global demand for crops is grow- Tilman and Clark (2014) estimated that the
ing more rapidly than yields, from 1980 total amount of global cropland might in-
to 2000, about 30 million hectares of land crease by from 400 million to 700 million
were cleared annually for agriculture in hectares between 2010 and 2050 if yields
developing nations, at the same time that in yield-gap nations were to continue to in-
about 15 million hectares per year globally crease along their past trajectories.
were abandoned from agriculture (Gibbs et A variety of analyses suggest that closing
al. 2010). Future land clearing will depend the yield gap could provide healthy diets for
on how global per capita demand for crops 9 billion people while requiring little more
increases with income and on how rapidly than the existing amount of global agricul-
the yields of developing nations increase. tural lands (Foley et al. 2011; Tilman et al.
The past trends in both of these variables 2011; Mueller et al. 2012). As such, efforts
are not encouraging. Mainly because of to bring yields up to their potential in all
rapid increases in meat consumption as in- nations could not only help provide all the
comes increase, crop demand may double. people of the Earth with secure and nutri-
However, yield trends in developing nations tious diets but also greatly reduce, or per-
show that many of these nations have yields haps eliminate, land clearing that is a ma-
that are growing slowly and currently are jor threat to global biodiversity. However,
much below their potential, a phenomenon it is imperative that yields be increased via
called the yield gap (Cassman 1999; Lobell, a much more sustainable approach to agri-
Cassman, and Field 2009). cultural intensification, as described below.
In some of the least developed nations,
crop yields are 20 percent–25 percent of
Sustainable Intensification Raises Yields
the yields that could be obtained via in-
with Lower Environmental Impacts
tensification. In a large suite of additional
developing nations, current yields are less Agriculture is of central importance to hu-
than half of those that could be attained. manity, providing most of the 7 billion
Because demand for crops is rising rapidly people of Earth with secure and nutritious
in developing nations, most of which have supplies of food. However, as currently
such yield gaps, two to five times more land practiced, agriculture also has inadvertent
is being cleared every year that would be but globally significant harmful impacts. It
needed if these yield gaps were closed. creates 30 percent of total GHG emissions,
The net effect of current trends is that and pollutes aquifers, lakes, river, and near-
global land clearing could, for decades to shore marine ecosystems with agrochemi-
come, remain at about the 20 million ha cals such as nitrogen and phosphorus from
per year rate reported by Gibbs et al. (2009) fertilizers and various pesticides (Tilman et
for 1980–2000, which is similar to the 20 al. 2001; Foley et al. 2011). The GHG sources
million ha per year of deforestation for are carbon dioxide released as land is
2000–2005 reported by Hansen, Stehman, cleared and soils are tilled; the potent GHG,
and Potapov (2010). This land clearing nitrous oxide, released from nitrogen fer-
would cause large numbers of extinctions tilization; and methane released by cattle,
S AV I N G B I O D I V E R S I T Y I N T H E E R A O F H U M A N - D O M I N AT E D E C O S Y S T E M S 361
sheep, and goat production and during rice Prevent Extinctions with Healthier Diets
cultivation. Moreover, both the GHG and
Adoption of healthy diets could also pre-
the agrochemical impacts of agriculture are
vent future land clearing and resultant spe-
on trajectories to double within the next 40
cies extinctions (Tilman and Clark 2014;
years (Tilman et al. 2001, 2011). These im-
Tilman et al. 2017). The major reason why
pacts and trends mean that the methods of
a 30 percent increase in global population
agricultural intensification used to close the
is forecast to lead to a 70 percent–110 per-
yield gap must be much more sustainable
cent increase in crop demand it that diets
than are current methods of agriculture.
change as incomes rise (Figure 28.3). The
A variety of practices have been identi-
adoption of “Western diets,” which is espe-
fied that reduce the environmental impacts
cially common in urban areas as develop-
of intensive agriculture while still achiev-
ing nations industrialize, leads to increased
ing high yields (Cassman 1999; Snyder,
per capita caloric, meat, processed food
Bruulsema, and Fixen 2009; Robertson and
and sugar consumption, to great increases
Vitousek 2009; Vitousek et al. 2009). These
in production of crops for animal feeds,
practices are called sustainable intensifica-
and to detrimental health impacts caused
tion (Godfray et al. 2010; Tilman et al. 2011;
by this nutrition transition (Popkin 1994;
Godfray and Garnett 2014). Chief among
Drewnowski and Popkin 1997; Popkin,
these are methods to obtain high crop
Adair and Ng 2012).
yields that minimize the amount of nitro-
If people were to adopt healthy diets,
gen fertilizer applied, and thus its negative
rather than typical Western diets, the lower
environmental impacts (e.g., Robertson
demand for meat and thus animal feeds
and Vitousek 2009; Vitousek et al. 2009;
would greatly reduce the need for land
Hoben et al. 2011; Meuller et al. 2014). A
clearing (Tilman and Clark 2014; Figure
foundation of sustainable intensification
28.4). For instance, the traditional Mediter-
is the timely application of the amount of
ranean diet, which is an omnivorous diet
fertilizers needed at various points during
that has high consumption of vegetables
crop growth (Snyder, Bruulsema, and Fixen
and fruits, nuts, and whole grains, and low
2009). For instance, when all the nitrogen
consumption of red meat, could eliminate
fertilizer needed for a year is applied at
the need for about 400 million more hect-
once, often before the crop is planted, 30
ares of cropland by 2050. In general, diets
percent–40 percent of this nitrogen is lost
that have low consumption of meat, espe-
from the system as ammonia that vaporizes
cially beef and other ruminant meats, re-
into the air, as nitrous oxide, and as nitrate
quire less cropland and pasture per capita.
and nitrite that enter waters. In sustain-
In total, if per capita consumption of red
able intensification, because fertilizers are
meat were to decline in developed nations
applied several times during the growing
to healthier levels, and if the current income
season so as to meet current crop demands
dependence of increased meat consumption
for nitrogen and other nutrients, most nu-
were to stop in developing nations, much
trients are taken up by the crop, and there
less land would have to be cleared while
is a much smaller pool of unused nutrient
providing the world with healthy diets in
that can be lost from the system. Because
2050 (Figure 28.4).
of this, the same high yields have been at-
tained using about 30 percent less nitrogen
than in conventional intensive agriculture
Reduce Food Waste to Prevent Extinctions
(Robertson and Vitousek 2009; Vitousek et
al. 2009). Sustainable intensification thus Reduction in food waste could also reduce
offers large GHG and water-quality benefits future land clearing and decrease the other
(Godfray and Garnett 2014). environmental impacts of agriculture
362 H OW C A N C O N S E RVAT I O N A N D P O L I C Y R E S P O N D?
Figure 28.3. Per capita demand for protein (A) and for wasted rather than eaten. If the proportion
calories (B) increase as a function of per capita income
(as estimated by per capita GDP). Each point is the aver-
of crop production that is wasted could be
age of 15 economically similar nations for a given year. cut in half, about 300 million hectares less
Years shown are 1961 through 2005. land would be needed to meet the cur-
rently projected 2050 global food demand.
(Godfray and Garnett 2014). The amounts This estimate is based on a simple calcu-
of crops and foods wasted and the rea- lation; 15 percent less crop production
sons for the wastage vary among nations would require 15 percent less cropland.
and regions (Gustavvsson, Cederberg, and With about 2 billion hectares of cropland
Sonesson. 2011). On average, in developing needed in 2050, the savings would be 300
nations, wastage occurs more from failure million hectares. A 15 percent reduction in
to fully harvest crops, or to preserve or ad- waste would also proportionately reduce
equately store them at the time of harvest, all of the other environmental impacts of
with less wasted once foods reach homes. agriculture.
In developed nations, wastage is often In total, increasing yields in low-yield-
small during crop harvesting, storage, and ing nations via sustainable intensification,
processing, but high for perishables in healthier diets, and reduced food waste
grocery stores, and even higher in homes would each decrease the cropland needed
and restaurants. Although the reasons for to feed the world of 2050. Because all three,
wastage differ among nations, on average in unison, would allow the world to be fed
about 30 percent of crop production is with less land than is currently used, partial
S AV I N G B I O D I V E R S I T Y I N T H E E R A O F H U M A N - D O M I N AT E D E C O S Y S T E M S 363
Figure 28.4. (A) Scenarios of additional global cropland sions could prevent a doubling of preindus-
that would be needed in 2050, relative to the amount of trial CO2 levels, and thus avoid some of the
2010. The BAU (business as usual) scenario uses fore-
casted food demand based on income-dependent global
worst climate scenarios and the extinctions
diets and current yield trends. The intensification sce- they would cause. The greatest emitter of
nario assumes the same food demand as BAU but as- GHG is electric power generation, mainly
sumes that yield gaps are closed via sustainable intensi- from coal, that creates 40 percent of total
fication. The Mediterranean diet scenario assumes global global GHG emissions. Renewable power
adoption of this diet current yield trends. The vegetar-
ian diet scenario assumes global adoption of this diet
from wind and solar could displace much
current yield trends. (Based on Tilman et al. 2011 and of current coal-based power and do more
Tilman and Clark 2014.) (B) Similar to part A but show- as power storage capabilities are increased.
ing the resultant changes in GHG emissions from crop Energy-efficient appliances, buildings,
production (but not land clearing) associated with BAU, and vehicles are another part of the Pacala
Mediterranean, and vegetarian scenarios.
and Socolow strategy. As discussed above,
sustainable intensification of agriculture,
adoption of each of these three approaches healthier diets, and reduced food waste
could still provide major biodiversity ben- also provide significant GHG benefits and
efits. Each of these actions would also de- should be part of a global strategy to ad-
crease nutrient pollution of aquifers, lakes, dress climate change.
rivers, and oceans and lead to lower agri-
cultural GHG emissions.
SU M M A RY
intensification of current croplands, healthy He, F., and S. P. Hubbell. 2011. “Species-area relation-
diets, efficient use of food and energy, and ships always overestimate extinction rates from habi-
tat loss.” Nature 463: 368–371.
a focus on renewable energy sources are the Hoben, J. R. J. Gehl, N. Millar, P. R. Grace, and G. P.
keys to creating a global economy and cul- Robertson. 2011. “Nonlinear nitrous oxide (N2O) re-
ture that can nourish and support humanity sponse to nitrogen fertilizer in on-farm corn crops of
and nature for millennia to come. There is the US Midwest.” Global Change Biology 17: 1140–1152.
no unavoidable need for any further land Hu, F. B. 2011. “Globalization of diabetes: The role of
diet, lifestyle, and genes.” Diabetes Care 34: 1249–1257.
clearing, no unavoidable reasons for a mas- International Energy Agency. 2015. World Energy Outlook
sive wave of species extinctions, and no fac- 2015. International Energy Agency.
tors that must necessarily drive the Earth Kuussaari, M., R. Bommarco, R. K. Heikkinen, A. Helm,
into unprecedented climatic changes. J. Krauss, R. Lindborg, E. Ockinger, et al. 2009. “Ex-
tinction debt: A challenge for biodiversity conserva-
tion.” Trends in Ecology and Evolution 24: 564–570.
Latham, R. E., and R. E. Ricklefs. 1993. “Continental
R E FE R E NCE S comparisons of temperate-zone tree species diver-
sity.” In Species Diversity in Ecological Communities: Histori-
Barnosky, A. D., N. Matzkel, S. Tomiya, G. O. U. Wogan, cal and Geographical Perspectives, ed. R. E. Ricklefs and D.
B. Swartz, T. B. Quental, C. Marshall, et al. 2011. “Has Schluter. University of Chicago Press.
the Earth’s sixth mass extinction already arrived?” Na- Lobell, D. B., K. C. Cassman, and C. B. Field. 2009. “Crop
ture 471: 51–57. yield gaps: Their importance, magnitude and causes.”
Cassman, K. C. 1999. “Ecological intensification of cereal Annual Review of Environment and Resources 34: 1–26.
production systems: Yield potential, soil quality, and Matson, P. A., W. J. Parton, A. G. Power, and M. J. Swift.
precision agriculture.” Proceedings of the National Academy of 1997. “Agricultural intensification and ecosystem
Sciences 96: 5952–5959. properties.” Science 277: 504–509.
Cardinale, B. J., J. Duffy, A. Gonzalez, D. Hooper, C. Per- Mueller, N. D., J. S. Gerber, M. Johnston, D. K. Ray, N.
rings, P. Venail, A Narwani, et al. 2012. “Biodiversity Ramankutty, and J. A. Foley. 2012. “Closing yield gaps
loss and its impact on humanity.” Nature 486: 59–67. through nutrient and water management.” Nature 409:
Drewnowski, A., and B. M. Popkin. 1997. “The nutrition 254–257.
transition: New trends in the global diet.” Nutrition Re- Mueller, N. D., P. C. West, J. S. Gerber, G. K. MacDon-
views 55: 31–43. ald, S. Polasky, and J. A. Foley. 2014. “A tradeoff fron-
Foley, J. A., N. Ramankutty, E. Bennett, K. Brauman, S. tier for global nitrogen use and cereal production.”
Carpenter, E. Cassidy, J. Gerber, et al. 2011. “Solutions Environmental Research Letters 9. https://doi.org/10.1088/
for a cultivated planet.” Nature 478: 337–342. 1748–9326/9/5/054002.
Gibbs, H. K., A. Ruesch, F. Achard, M. Clayton, P. Hol- Pacala, S., and R. Socolow. 2004. “Stabilization wedges:
mgren, N. Ramankutty, and J. Foley. 2009. “Tropical Solving the climate problem for the next 50 years with
forests were the primary sources of new agricultural current technologies.” Science 305: 968–972.
land in the 1980s and 1990s.” Proceedings of the National Popkin, B. M. 1994. “The nutrition transition in low-
Academy of Sciences 107: 16732–16737. income countries: An emerging crisis.” Nutrition Reviews
Gibson, L., A. J. Lynam, C. J. A. Bradshaw, F. He, D. P. 52: 285–298.
Bickford, D. Woodruff, S. Bumrungsri, and W. F. Popkin, B. M., L. Adair, and S. W. Ng. 2012. “Global nu-
Laurance. 2013. “Near-complete extinction of native trition transition and the pandemic of obesity in de-
small mammal fauna 25 years after forest fragmenta- veloping countries.” Nutrition Reviews 70: 3–21.
tion.” Science 341: 1508–1510. Robertson, P. G., and P. Vitousek. 2009. “Nitrogen in ag-
Godfray H. C. J., J. R. Beddington, I. R. Crute, L. Had- riculture: Balancing the cost of an essential resource.”
dad, D. Lawrence, J. F. Muir, J. Pretty, S. Robinson, S. Annual Review of Environment and Resources 34: 97–125.
M. Thomas, and C. Toulmin. 2010. “Food security: Rosenzweig, M. L. 1995. Species Diversity in Space and Time.
The challenge of feeding 9 billion people.” Science 327: Cambridge University Press.
812–818. https://doi.org/10.1126/science.1185383. Snyder, C. S., T. W. Bruulsema, T. L. Jensen, and P. E.
Godfray, H. C. J., and T. Garnett. 2014. “Food security Fixen. 2009. “Review of greenhouse gas emissions
and sustainable intensification.” Philosophical Transactions from crop production systems and fertilizer manage-
of the Royal Society B 369. https://doi.org/10.1098/rstb ment effects.” Agriculture, Ecosystems and Environment 133:
.2012.0273. 247–266.
Gustavvsson, J., C. Cederberg, and U. Sonesson. 2011. Tilman, D., C. Balzer, J. Hill, and B. Befort. 2011. “Global
Global Food Losses and Food Waste. FAO. food demand and the sustainable intensification of ag-
Hansen, M. C., S. V. Stehman, and P. V. Potapov. 2010. riculture.” Proceedings of the National Academy of Sciences 108:
“Quantification of global gross forest cover loss.” Pro- 20260–20264.
ceedings of the National Academy of Sciences 107: 8650–8655.
S AV I N G B I O D I V E R S I T Y I N T H E E R A O F H U M A N - D O M I N AT E D E C O S Y S T E M S 365
Tilman, D., and M. Clark. 2014. “Global diets link en- Tilman, D., F. Isbell, and J. M. Cowles. 2014. “Biodiver-
vironmental sustainability and human health.” Nature sity and ecosystem functioning.” Annual Review of Ecology,
515: 518–522. Evolution, and Systematics 45: 471–493.
Tilman, D., M. Clark, D. R. Williams, K. Kimmel, S. Po- Tilman, D., R. M. May, C. L. Lehman, and M. A. Nowak.
lasky, and C. Packer. 2017. “Future threats to biodi- 1994. “Habitat destruction and the extinction debt.”
versity and pathways to their prevention.” Nature 546: Nature 371: 65–66.
73–81. Vitousek, P., R. Naylor, T. Crews, M. B. David, L. E.
Tilman, D., J. Fargione, B. Wolff, C. D’Antonio, A Dob- Drinkwater, E. Holland, P. J. Johnes, et al. 2009. “Nu-
son, R. Howarth, D. Schindler, W. Schlesinger, D. trient imbalances in agricultural development.” Science
Simberloff, and D. Swackhamer. 2001. “Forecasting 324: 1520–1521.
agriculturally driven global environmental change.”
Science 292: 281–284.
This page intentionally left blank
Monika Bertzky
Contributors Independent Consultant
Biodiversity, Ecosystems, and Environment
David Ainley John B. Bradford
Senior Ecologist Research Ecologist
H. T. Harvey & Associates Southwest Biological Science Center
United States Geological Survey
Cecile Albert
Research Officer David D. Breshears
Mediterranean Institute of Biodiversity and Regents’ Professor
Marine Continental Ecology School of Natural Resources and the
Environment
Craig D. Allen Department of Ecology and Evolutionary
Research Ecologist Biology
Fort Collins Science Center University of Arizona
New Mexico Landscapes Field Station
United States Geological Survey Rebecca C. Brock
Senior Programme Officer
Katie Arkema Climate Change and Biodiversity
Lead Scientist United Nations Environment Programme
Natural Capital Project World Conservation Monitoring Centre
Stanford Woods Institute for the (UNEP-WCMC)
Environment
Stanford University Olivier Broennimann
Staff Scientist
Richard B. Aronson Department of Ecology and Evolution
Professor and Head Institute of Earth Surface Dynamics
Department of Biological Sciences University of Lausanne, Switzerland
Florida Institute of Technology
Lauren B. Buckley
Michael Avery Associate Professor
PhD Candidate Department of Biology
Department of Biology University of Washington
Pennsylvania State University
Aline Buri
Grant Ballard Graduate Assistant
Chief Science Officer Institute of Earth Surface Dynamics
Point Blue Conservation Science University of Lausanne, Switzerland
Céline Bellard Kevin D. Burke
CNRS Researcher PhD Candidate
Université Paris-Sud Nelson Institute for Environmental Studies
University of Wisconsin–Madison
367
368 CONTRIBUTORS
373
374 INDEX
invasive species, 120–122, 147, 172, 174f, 258–269; al- Last Glacial Maximum (LGM), 15, 70–73, 284
tered distribution of existing populations, 259, 262; Last Interglacial (LIG) stage, 71
ballast water management practices, 259–260; bio- Latimeria (coelacanth), 116
diversity conservation and restoration, 264; biofuel latitudinal temperature gradients, 201, 204, 211, 212, pl. 4
production and, 261; classifying, 265; climate change Laurentian Mountains, 323–325
and, 258–259; climatic constraints on, 259; coordi- Laurentide Ice Sheet, 133, 133f
nated research and policy efforts, 264–265; extreme leading edges, marine systems, 30–32
weather events and, 261; fish, exotic, 260–261; im- leatherback turtles, 175
pacts of existing populations, 259, 263; intentional life-history traits, 42–44, 165, 258
introduction, 260; Lake Victoria, 242; management lineages, 66–67, 70
effectiveness, 259, 263; marine, 173, 192, 259, 261; Linepithema humile, 261
mechanisms of transport and introduction, 259–261; lithodids, 121
modeling, 259, 262; in mountain regions, 222; or- Little Ice Age, 70
namental, garden-variety plants, 262; pet trade, 260, littoral organisms, 251, 252
265; propagule pressure, 259, 260; shipping and, lizards, Sceloporus, 226
259–260, 265; solutions, 264–265; three axes of im- local impacts of change, 3, 61–63, 174f, 306–307; crop
pact, 263 adaptation, 351–352; TV weathercaster coverage, 341,
Iriartea (palm), 143 343
iron fertilization, 328 lodgepole pine (Pinus contorta), 83
irrigation, 300 loggerhead turtles, 175
islands, 120, 125–127, 126f, 147; 2°C too much, 327; long-term observations, 28; range and abundance
megafaunal populations, 135 changes, 26, 27t, 32–33
isotherms, 31 Lovejoy, Thomas, 188
Italian Alps, 224 low carbon fuel standard (LCFS), 261
Luquillo Forest (Puerto Rico), 9
Japan: CO2 vents, 191; flowering phenology, 49 Lyme disease, 262
jet stream, 16
jump dispersal, 145, 146f mackerel, 177
Juniperus, 133f, 134, 137 macroinvertebrates, marine, 171–175
Juniperus monosperma, 83 Madagascar, 196
Magellanic penguin (Spheniscus magellanicus), 175
kangaroo rat species, 72 magmas, 103
Kangerlussuaq, Low Arctic Greenland, 44 maize, 347, 348–349, 350, 353
Keeling, M. J., 106, 271 malaria, 270, 271–280, 273f, 306; avian, 275; highland,
krill, 121–122 East Africa, 275–277
K-T extinction event, 15, 21f, 101, 328, 329f mammals, 357; body size, 131; marine, 176; megafauna,
118, 135, 147; mountain regions, 223, 227; uphill and
lakes: East Africa, 242; ecosystems, 251–252; eutrophi- poleward shifts, 214. See also individual species
cation, 242, 243; Gerzensee (Switzerland), 133f, 134; managed relocation, 262
high-latitude, 228, 238; Kråkenes (Norway), 133f, management-based actions, 4, 6–10, 86–87; Australian
134–135; Kyoga Basin (Uganda), 242; Malawi, 242; tuna fisheries, 255–257; marine biodiversity, 176–177;
Nawampassa, 242; papyrus swamp, 242; sediments, post-modern marine communities, 118–119; rewild-
133–134; Soppensee (Switzerland), 133f, 135; strati- ing, 118; spatial, 255–257. See also protected-area
fication, 240, 242, 24; Tana, 242; Tanganyika, 242; management
Titicaca, 144; Victoria, 238, 242; Yoa, 137 mangroves, 6, 191, 198f, 303, 305
La Meseta Formation, 119–120 manufactured landscapes, 147–148
land clearing, 7, 205, 308, 315, 332, 357 marine biodiversity, 168–182, 302–303; acidification
LANDIS-II, 164 and, 168–169, 171, 175, 178, 185–195; conservation,
landraces, 352 176–177; habitat complexity, 191; invasive species,
landscape genetics, 67–71, 68f, 69t, 70–71; contempo- 173, 192, 259, 261; key challenges, 177–178; mac-
rary time frame, 73–74 roinvertebrates and fishes, 171–175; mammals, 176;
landscape genomics, 67–68 mechanisms of climate change effects on, 169–170;
land-sharing approaches, 8 oxygen, need for, 169–170, 170f; phytoplanktons,
land-use change, 161, 163, 358 macroalgae, and seagrasses, 171, 188; poleward shifts,
large igneous provinces (LIPs), 103–104, 108 171, 172, 175; reproduction, 188; reptiles, mammals,
large marine ecosystems (LMEs), 173 and birds, 175–176; responses to climate change,
larvae: insect, 27t, 32–33, 225, 235, 271; marine, 120, 170–176; seabird populations, 91–92, 175–176; spe-
121, 169, 170, 173, 188, 189, 191, 240; mountain re- cies richness, 169, 170, 171, 173. See also acidification;
gions, 225 benthic systems; oceans
INDEX 381
marine protected areas (MPAs), 62, 122, 177, 287f 165; extinction risk, 294; fish body size, 173–175; gap
Marine Stewardship Council, 177, 343 models, 164; general circulation model (GCM) uncer-
marine systems: Antarctica, future of, 120–121; climate tainty, 158–159, 165, 198f; generalized dissimilarity
and biodiversity in, 117; in deep time, 114–124; lead- modeling (GDM), 158, 160; habitat preference, 256;
ing edges, 30–32; mass-extinction events and, 97; HadCM2 and HadCM3 climate models, 271; invasive
overharvesting, 114, 118–119, 121; post-modern, 118– species, 259, 262; landscape genetics, 73; landscape
119; predation, 116, 117; teleosts, 116, 119, 120; three models, 164; last deglaciation, 132–135; MaxEnt
evolutionary faunas, 115–116 (maximum entropy theory model), 158, 160; niche
Marmota flaviventris, 227 envelope models, 237; niche evolution, 73; niche
marmots, yellow-bellied (Marmota flaviventris), 227 models, 183–184, 237; from paleoecological perspec-
marten (Martes americana), 227 tive, 106–107; protected-area management, 288; sea
Martes americana, 227 level rise, 126–127; spatial autocorrelation, 274; spe-
Marxan, 158 cies distribution models (SDMs), 160–164, 272–274;
Mauritia (palm), 142–143 susceptible (S), infected (I), and recovered (R) (SIR)
maximum entropy (MaxEnt) model, 158, 160 models, 271; tree line advance, 222–224. See also deep
mean temperature of catch (MTC), 172–173, 172f time; projections; time stamp metaphor
meat, per capita demand for, 361, 362f modeling species and vegetation distribution, 157–167;
mechanistic links, 26, 32 adding information on population processes, 164;
Medieval Warm Period, 70 area metrics, 161–162; Earth-system modeling, 157,
Mediterranean diet, 361, 363f 165; ensemble approaches, 158–160; gap models,
Mediterranean mountains, 224 164; general circulation model (GCM) uncertainty,
Mediterranean Sea, 14, 191 158–159, 165, 198f; landscape models, 164; lessons
Meerfelder Maar, 133f, 134 learned, 160–164; macroscale modeling, 165–166;
megadiversity, 59–61 population-level metrics, 162–163; reasons for, 157–
megadroughts, 130t, 137, 150 158; species distribution models (SDMs), 158–164;
megafauna, 118, 135, 147 velocity metrics, 163–164, 163f
megafires, 81–82 Modern fauna, 115–116
Mekong River, 239t; dam, 318 MODI S satellite observations, 43
Melanoplus femurrubrum, 332–335 Mojave Desert species, 72
Mesozoic era, 104 mollusks, 59, 86f, 188, 190, 240
“Mesozoic marine revolution,” 116 monkey flower species, 72
messages. See awareness monsoons, 132, 133, 137
meta-analyses, 25, 28, 30t, 212–213, 261 montane ecosystems, 83, 85, 201–202. See also mountain
metazoans, 101–102 biodiversity
meteor strikes, 98, 101 Monterey Bay Aquarium, 342–343
methane (CH4), 4, 13, 20, 100, 105, 312, 330, 360–361 Montreal, Canada, 323–325
Mexico, 240 mortality: consumer-resource dynamics, 247f, 248–249;
microarthropods, 225 trees, 81–83, 217
microclimates, 135, 146, 149, 284–285, 318; broad-scale mosquitos, 271–280, 306, 344f; Aedes aegypti, 271, 272f,
ecosystem changes and, 81, 83, 86, 87; connectivity 274; Anopheles gambiae, 271, 277; Anopheles maculipennis,
and, 289 271; Triatoma brasiliensis, 274
microrefugia, 147, 149–150 mountain avens (Dryas sp.), 46
migration: assisted, 86–87, 262, 289; upslope and mountain biodiversity, 221–233; alpine and nival areas,
downslope, 145, 146f, 149, 201–204 222, 222t, 239f; alpine and subalpine grasslands, 49,
migration rates, 134, 135, 139. See also velocity of climate 221, 223f, 224; amphibians, 226; birds, 223, 227; cli-
change (VoCC) mate change in mountain areas, 221–222; freshwater
Milankovitch cycles, 107 species, 228; global area of bioclimatic mountain belts,
Ministry of Sustainable Development, Environment, and 222t; high-altitude habitat “squeeze,” 239t; insects,
Parks (Quebec), 323 225–226; mammals, 227; microclimatic variation, 224;
Mississippi Basin fauna, 240 mountain forests and tree lines, 222–224, 223f; percent
Mississippi Delta, 330 of land area, 221; reptiles, 226–227; soil microorgan-
mitigation. See ecosystem-based adaptation (EbA); terres- isms, 224–225; tropical montane ecosystems, 85, 224;
trial ecosystems, climate change mitigation warming rate, 221–222. See also montane ecosystems
mitochondrial (mtDNA), 67, 68–69, 71, 72 muscid flies, 46
mobility of species, 246, 249–250, 251f mutualistic interactions, 45, 50–52; coral reefs, 60, 61
modeling: climate system, 17–18, 21f; coral bleaching, mycorrhizal processes, 328
61–62; correlative, 164, 272–274, 273f; crop yield,
348–349; disease transmission, 271; diversity-flow National Academy of Sciences, 326–327
relationship, riverine systems, 237; Earth-system, 157, nationally appropriate mitigation actions (NAMAs), 315
382 INDEX
National Oceanographic and Atmospheric Administra- tory role, 55–56; sea surface temperatures (SSTs), 57,
tion, 14 77, 173, 175; thermocline, 249; upwelling intensity,
natural breakwaters, 302 175, 178; volume of 1.3 billion km3, 56; water col-
natural disturbances, 216–217 umn, 59, 60, 116, 118, 168, 173. See also Bering Sea;
natural integrity, 197 Chukchi Sea; coral bleaching; coral reefs; marine bio-
Nature Conservancy, 330 diversity; sea level rise
Nebraska Sand Hills, 137 Ochotona princeps, 227
Neogene, 70, 328, 329f oil palm cultivation, 316, 316f
Neolithodes, 121 oil wells, 327
neotropical perspective, 142–153; path to homogeniza- omnivory, 250–251
tion, 147–148 Ontong Java Plateau eruption, 105
New Climate Economy report, 2015, 330 ophiuroids, 116
New Guinea, 196, 197, 227 optimal foraging theory, 170
New York Declaration on Forests (restoration), 311 orbital cycles, 14, 104–105, 131, 137, 143
next-generation sequencing (NGS) platforms, 67, 69, Ordovician period, 103, 116
73–74 organelle genomes, 67, 68, 72
niche conservatism, 46, 52 organism physiology, 169
niche evolution, 72–73 Ostreococcus tauri, 188
niche models, 183–184, 237, 272–274, 273f oxygen, in waters, 56, 103
niches: invasive species and, 264; thermal, 201, 204, 214 oxygen and capacity-limited thermal tolerance (OCLTT),
Nile perch (Lates niloticus), 242–243 192
nitrogen deposition, 4–5, 13 ozone layer, 3, 20
nitrogen fertilizer, 360–361
nitrous oxide (N2O), 13, 360 Pacific Ocean, 39, 57, 60, 77–78, 106, 161
no-analogue communities, 117–118, 148 paleobiology, 117–118
non-forest ecosystems, 311–313 Paleocene-Eocene Thermal Maximum (PETM), 102f,
Nordic Genetic Resource Centre, 353 104–105, 107, 117, 129–131, 130t, 138, 199, 294; acid-
North America, 16, 26, 28, 30, 66, 72–73; freshwater ification and, 186
fishes, 240–241; ice sheet, 106; land bridge, 14 Paleocene epoch, 328, 329f; clathrates, 129–130
North Atlantic deep water (NADW), 106 paleoclimatic record, 14, 27t, 69t
North Atlantic Ocean, 132–134; “conveyor belt for salt,” paleoecological perspective, 97–113; biodiversity crises
106; northward movement of marine life, 172. See also in Earth history, 101–104; biodiversity in a change-
Atlantic Ocean able climate, 107–108; Earth system and carbon cycle,
Northern Hemisphere land areas: range and abundance 99–101; geologic time scale, 98–99, 98f, 99f, 128;
figures, 30; summer temperatures, 15–16, 16f modeling from, 106–107
North Pacific Fisheries Management Council, 40–41 paleoecology, 81, 294
North Sea, 173 Paleozoic era, 98, 99f, 104, 107, 115–116
Norway, 43, 133f, 134, 225; genebanks, 353 Palmer Deep, 121
Norway, High Arctic, 43 palms, 142–143
nutrient enrichment, 176, 228 panda, 227
Papua New Guinea, 191
oak, California (valley oak, Quercus lobata), 73 Paraguay, 312f
Oak Ridge experiments, 9 Paris Climate Agreement, 20, 62, 63, 209, 310, 319, 327
ocean anoxic events (OAEs), 104–105, 108 Paris Climate Change Conference of the Parties, 330
oceans, 12; absolute rate of temperature rise, 31; acidifi- parrotfish, 6–7
cation, 40, 61–63, 100, 103, 106, 114, 168–169, 171, past abrupt ecosystem changes, 128–141, 130t, 144;
175, 178, 185–195; asphyxiation, 103, 104; bathyal examples, 132–135, 133f; hydrological variability,
zone, 114, 116, 119; biological responses to rapid 135–138, 136f; Paleocene-Eocene Thermal Maxi-
ocean change, 58–59, 58f; carbon dioxide (CO2) ab- mum (PETM), 129–131. See also broad-scale ecosystem
sorption, 55–57, 57f; chemistry, 57; CO2 equilibrium, changes, rapid
185–186, 186f; conveyor-belt model for salt, 106; pastoral societies, 137
dead zones, 4–5, 56, 170; deep sea “living fossils,” peatlands, 131, 304, 311, 313, 329
116; early studies, 56; heterogenous warming, 255; pelagic systems, 39, 118, 251
layers, 56–57; local and global warming and acidifica- penguin optimum, 91
tion, 61–63; mobility of organisms, 59; oxygen con- penguins, 121–122; Adélie (Pygoscelis adeliae), 91–92, pl.
tent, 56, 103, 169, 176; percentage of coverage, 55, 5; emperor (Aptenodytes forsteri), 92, 175; king (Apteno-
168; pH of waters, 20, 56, 57, 185–186; physical and dytes patagonicus), 249; Magellanic (Spheniscus magellanicus),
chemical changes in, 56–57; regime shifts, 77; regula- 175
INDEX 383
per capita consumption and purchasing power, 357, 359, 44–45; implications of differential phenological ad-
362f vance in, 45–50; mutualistic interactions, 45, 50–52;
permafrost, 82, 131, 330 plant-flowering phenology, 46; pollen limitation in
Permian period, 101, 103 flowering plants, 46–47, 47f; recent trends, 43–44;
Peru, 350f snowmelt and frost damage, 234–235; species-spe-
pest and pathogen outbreaks, 6, 81 cific phenological dynamics, 49–50, 50f, 51f
Phanerozoic Eon, 103, 108, 115 pollution, 3–4, 61, 63, 108, 176, 317
phenology, 42–54, 173; community-averaged rates, 44; population genetics, 67, 69t
in consumer-resource interactions, 44–45; mountain population increase, 7, 8, 357, 358, 359, 363
biodiversity and, 225; multiannual studies, 44; mu- population-level metrics, 162–163
tualistic interactions, 45, 50–52; recent trends related post-modern marine communities, 118–119
to climate change, 43–44; seabirds, 175–176; shifts in poverty alleviation, 306–307
synchrony, 218; temperate and boreal regions, 214– precipitation, 16, 29, 33, 199, pl. 1; agricultural produc-
215; trophic match-mismatch, 44–52. See also pollina- tion and, 300–301; Amazon region, 143, 201, 208;
tor-plant associations, phenological dynamics in base flow reduction, 239t, 241; freshwater biodiver-
pH of ocean waters, 20, 56, 57, 185–186 sity and, 228; invasive species and, 258–259; link to
photoperiod cues, 214 temperature increase, 199–201; past abrupt changes,
photosynthesis, 106, 328; coral reefs and, 60; effect of 136f, 137–138; United States, 241
heat on vegetation, 349; leaf-level responses, 216; pre-Columbian peoples, 147–148
plankton, 188 predation, 116, 117, 170; acidification and, 192; carbon
photosynthesizing organisms, 103 effects of, 332–335; mountain regions, 227; spider
phylogenetics, 67, 70 species, 252; trophic levels, 44, 92, 117, 250–251.
phylogenomics, 67 See also food webs; food webs, asymmetrical thermal
phylogeography, 66–67, 68f; comparative, 67, 68f, 70; response
spatial and temporal scales, 68–70, 68f, 69t predators, 116, 122
physiological sensitivities to climate change, 212–213 “pride campaigns,” 344
Picea critchfieldii, 135 projections, 13; acidification of oceans, 190; Amazon Re-
picoplankton Ostreococcus tauri, 188 gion, 84; coral bleaching, 62–63; extinction resilience
pied flycatcher (Ficedula hypoleuca), 44 and risk, 74; extreme weather changes, 19f; insects,
pika, American (Ochotona princeps), 227 mountain regions, 225–226; mass-extinction data,
pine beetle (Dendroctonus ponderosae), 83, 216, 305 98, 98f; mountain regions, 228–229; temperate and
pine processionary moth (Thaumetopoea pityocampa), 262 boreal regions, 218–219; through the twenty-first cen-
pinnipeds, 41f, 176 tury, 17–20; vegetation change, 81, 82f. See also model-
piñon-juniper (Juniperus monosperma) woodlands, 83 ing; modeling species and vegetation distribution
piñon pine (Pinus edulis), 82–83, 137 prokaryotic cyanobacteria, 103
Pinus, 82–83, 133f, 134, 137 protected-area management, 7–8, 283–293; connectiv-
Pinus contorta, 83 ity, 286–287; connectivity between present and future
Pinus edulis, 82–83, 137 populations, 289; conservation in changing climate,
planetary boundaries, 4–5 290–291; ecosystem gap, 286; enduring geophysi-
plankton, 59, 116; acidification and, 187–188; picoplank- cal features, 289; enhanced synergies, 290; ensure
ton Ostreococcus tauri, 188; poleward shift, 171, 172 permanence and quality of existing protected areas,
plantations, 314, 316 288; expand or modify protected areas and networks,
plant functional types (PFTs), 165–166 288–289; first principles, 288; four components,
Plasmodium falciparum. See malaria 283–284; gaps in design and management, 285–286;
plate tectonics, 99, 117 geographic gap, 285–286; “gray” literature, 285; habi-
Pleistocene epoch, 66, 71–72, 99, 102f; evolution under, tat corridors, 287; innovative approaches to adapt pro-
296; glaciations, 117 tected areas to climate change, 288–290; magnitude,
Pleistocene-Holocene transition, 70 rate, timing, and intensity of climate change and pro-
policy making, 4, 9; agriculture and, 347–355; aware- tected areas, 284–286; models, 288; scale commen-
ness and, 9, 336–346; coral reefs, 62; freshwater fish surability, 287–288; science-management gap, 285;
management, 243–244; invasive species, 264–265; stepping-stones, 289–290, 324–325, 324f; surround-
range and abundance changes, 25; SDM simulations, ing matrix, 286–287, 288; theory of, 286–288. See also
161; terrestrial ecosystems, 318–319 conservation; management-based actions
pollen, 46–47, 133–134, 144 protected areas (PAs), 161, 205; marine (MPAs), 122,
pollination capacity, 299–300 176–177, 287f; mountain regions, 227; terrestrial pro-
pollinator-plant associations, phenological dynamics in, tected areas (TPAs), 287f, 290
42–54; abiotic conditions, 42, 43, 44, 52; conserva- Proterozoic eon, 115
tion strategies, 52; consumer-resource interactions, pteropod species, 92, 188
384 INDEX