Patrones de distribucin, abundancia e

interacciones entre carnvoros simptridos en un

rea mediterrnea protegida
Distribution, abundance and interaction patterns between
sympatric carnivores in a Mediterranean protected area

Tesis doctoral
Carolina Soto Navarro
Sevilla, Diciembre 2012

Director de tesis
Dr. Francisco Palomares Fernndez
Profesor de Investigacin
Departamento de Biologa de la Conservacin
Estacin Biolgica de Doana-CSIC (EBD-CSIC)
Sevilla-Espaa
Tutor
Dra. M Jos LeivaMorales
Profesora Titular
Departamento de Biologa Vegetal y Ecologa
Universidad de Sevilla
Sevilla-Espaa

ndice

ESTRUCTURA DE LA TESIS DOCTORAL............................................

11

INTRODUCCIN........................................................................................

12

Contexto de la tesis.............................................................................


Marco terico......................................................................................


Teora de la seleccin de hbitat.............................................

13

El concepto de hbitat.............................................................

21

El concepto de nicho ecolgico..............................................

25

El estudio de la seleccin de hbitat.......................................

28

Escalas espacio-temporales de investigacin.........................

29

De lo realmente importante y disponible para los individuos

31

Caractersticas ecolgicas de las especies y mecanismos de

coexistencia.............................................................................

32

De cmo simplemente encontrar el hbitat ms idneo suele



no ser suficiente.......................................................................

36

Objetivos de la tesis doctoral..............................................................

39

Especies y rea de estudio...................................................................

41

21
21

Las especies modelo................................................................

41

rea de estudio........................................................................

46

CHAPTER 1. Non-biological factors affecting track censuses:

implications for sampling design and reliability.............................................

65

CHAPTER 2. Fine-scale habitat use and niche separation in a guild of

sympatric carnivore species that differ in life-history traits ...........................

93

CHAPTER 3. Species abundances in a community of sympatric

carnivores: a trade-off between habitat selection and interspecific
interactions?....................................................................................................

135

CHAPTER 4. Human-related factors regulate dog presence in protected

areas: implications for conservation and management control...........................

157

CHAPTER 5. Surprising low abundance of European wildcats in a protected

area of southwestern Spain.............................................................................

185

CONCLUSIONES........................................................................................

209

AGRADECIMIENTOS...............................................................................

213

ESTRUCTURA DE LA TESIS DOCTORAL

La presente tesis consta de los siguientes apartados: una Introduccin al

estado de la cuestin y al marco terico de la tesis; los Objetivos de la tesis; una

seccin de Especies y rea de estudio en la que se describen las especies modelo
y el rea de estudio; redactados todos en castellano; cinco Captulos Temticos,
presentados en formato de manuscritos publicados y por publicar en revistas
internacionales, y por lo tanto escritos en ingls, aunque acompaados de sus
respectivos resmenes en castellano; y, finalmente, las Conclusiones principales
obtenidas en la tesis redactadas en castellano.

11

Introduccin

Introduccin

Contexto de la tesis

Los seres humanos hemos cambiado la biosfera1 sustancialmente, hasta

el punto que algunos abogan por el reconocimiento de la poca actual en la que

vivimos como una nueva era geolgica, el Antropoceno (Steffen et al. 2011,
Rockstrm et al. 2009, Zalasiewicz et al. 2011). Existen evidencias de que la Tierra
como ecosistema global puede estar acercndose a una transicin crtica o punto
de inflexin que puede suponer un cambio de estado2 rpido a escala planetaria en
un perodo de dcadas o siglos, si no se ha iniciado ya en la actualidad (Barnosky
et al. 2012). A pesar de que vivimos en una era geolgica con la mayor riqueza
de especies y diversidad, nos encontramos en el comienzo de un fenmeno de
extincin en masas (Primack 2006). La principal causa de esta actual crisis a escala
global es el incremento de la alteracin humana de la Tierra (Vitousek et al. 1997,
Primack 2006, Sinclair et al. 2006). De hecho, las tasas de crecimiento actuales de
la poblacin mundial (77 millones de personas al ao) as como el incremento de
las actividades humanas han supuesto cambios rpidos y drsticos en los paisajes
naturales (como la transformacin de entre un tercio y la mitad de la superficie
terrestre (Vitousek et al. 1997, Hoekstra et al. 2005). El consumo excesivo de los
recursos y la expansin de las reas humanizadas han llevado a la destruccin directa

El trmino biosfera fue acuado por el gelogo Eduard Suess en 1875, pero el concepto ecolgico
de biosfera procede de 1920 con Vladimir I. Vernadsky, precediendo a la introduccin en 1935 del trmino
ecosistema por Arthur Tansley. En esta tesis empleo la definicin de biosfera como ecosistema global.
En ecologa, la teora de estados estables alternativos predice que los ecosistemas pueden existir
en mltiples estados (conjunto de condiciones biticas y abiticas nicas). Estos estados alternativos no son
transitorios y por lo tanto se consideran como estables a escalas de tiempo ecolgicamente relevantes. Los
ecosistemas pueden sufrir una transicin de un estado estable a otro, en lo que se conoce como un cambio de
estado (a veces llamado un cambio de fase o cambio de rgimen), cuando son sometidos a perturbaciones. Uno
de los cambios de estado ms rpidos del planeta y el ms reciente, ha sido la transicin desde la ltima era
glacial al presente perodo interglacial (Scheffer et al. 2009, Lenton 2012) que se produjo a lo largo de milenios
(Hoek 2008).

13

Introduccin

y fragmentacin3 de hbitats naturales (por ejemplo, debido a la red de carreteras, la

expansin de las zonas agrcolas y ganaderas, y el desarrollo de las ciudades). De una
manera indirecta, procesos inducidos por el hombre tambin causan la degradacin
de los hbitats de numerosas especies a travs de la contaminacin (pesticidas,
herbicidas, emisiones de industrias qumicas y de automviles que conducen a un
cambio climtico anmalo (Houghton et al. 2001) y la introduccin de especies
invasoras (Primack 2006). Debido a estos cambios drsticos en los ecosistemas,
muchas especies nativas se han extinguido o estn en peligro de extincin debido
al impacto negativo de la reduccin de hbitats y su degradacin (Tilman et al.
1994, Brooks et al. 2002, Primack 2006). La transformacin de la tierra es la fuerza
motriz de la amenaza para la biodiversidad a nivel mundial (Vitousek et al. 1997), y
este proceso est funcionando tan rpido que para la mayora de las especies no hay
tiempo para una adaptacin a tales modificaciones a travs de una compensacin
evolutiva (Teyssdre 2004).

Al modificar el hbitat de numerosas especies, los seres humanos se han

convertido en una parte integral de su entorno. Esta influencia tiene efecto a diferentes
escalas espaciales y niveles biolgicos, tales como la distribucin geogrfica de las
especies, la organizacin espacial de las poblaciones y el comportamiento individual
a escala fina. Por ejemplo, la fragmentacin de hbitats puede implicar la divisin
de las actuales poblaciones en subpoblaciones o metapoblaciones4 alterando as su
dinmica (por ejemplo, Banks et al. 2005). Este escenario puede restringir a muchas
Existe una amplia diversidad de definiciones para el trmino fragmentacin de hbitats. En esta tesis
lo definir como el proceso por el que un hbitat grande se transforma en pequeos parches del mismo hbitat
(vase la revisin de Fahrig 2003).

Usar la definicin de Moilanen et al. (1998): una metapoblacin es un conjunto de poblaciones

locales que habitan distintos parches de hbitat distribuidos en un espacio definido.

14

Introduccin

especies a reservas naturales y reas adyacentes en gran parte del mundo (Woodroffe
& Ginsberg 1998). No obstante, las reas protegidas no estn exentas en muchos
casos de las alteraciones relacionadas con las actividades humanas, bien porque
actividades humanas de alteracin del medio han tenido lugar de forma comn o
bien porque sufren las consecuencias de la influencia de las reas que las circundan.
Adems, para muchos de los ambientes y ecosistemas no existe una informacin
precisa ni fiable sobre las complejas interrelaciones que regulan y determinan la
distribucin y abundancia de las especies, ni los efectos que diferentes factores
tienen sobre el individuo, las poblaciones o las comunidades.

Dentro de este contexto, la comprensin de la relacin entre los organismos

y su ambiente tiene implicaciones fundamentales en varias disciplinas cientficas

como la Ecologa (por ejemplo, cmo los cambios ambientales afectan a los
individuos o a la dinmica poblacional?), Evolucin (por ejemplo, cmo afectan
estos cambios al fitness de los individuos?, cmo se adaptan las especies a estos
cambios, cules son los fenotipos que mejor se adaptan a estos cambios?) y
Biologa de la Conservacin (por ejemplo, cules son los mejores hbitats que
deben ser priorizados para la conservacin de las especies?).

La importancia atribuida a los recursos naturales vara entre generaciones

junto con los cambios en la sociedad y las consecuencias de los niveles de explotacin
anteriores (Conover 2002), pero el reconocimiento de que la biodiversidad desempea
un papel esencial en el bienestar humano y en el equilibrio de los ecosistemas es
creciente, y ha llevado a la generacin de medidas urgentes para incrementar la
conservacin de especies y hbitats en todo el mundo, lo que tambin se traduce en
la cristalizacin de la era moderna de la ciencia y poltica conservacionistas. Desde

15

Introduccin

que se acu el trmino en 1978 y se cre esta disciplina5, los estudios centrados
en la Biologa de la Conservacin han crecido exponencialmente hasta alcanzar la
relevancia actual que tiene esta rea de investigacin (Figura 1).

La distribucin de las especies en el medio ambiente (tambin aplicable

para las poblaciones e individuos, pero me referir a las especies en esta seccin
por conveniencia de lectura) surge de la interaccin entre eventos deterministas
y estocsticos (Corsi et al. 2000). De hecho, es el resultado de la interaccin
entre eventos biolgicos (por ejemplo, alimentacin, reproduccin o dispersin;
deterministas) y eventos impredecibles (por ejemplo, incendios, tormentas;
estocsticos). Durante estos eventos biolgicos, los animales pueden elegir las
reas que mejor satisfagan sus requerimientos ecolgicos (por ejemplo, reas con
abundantes recursos en pocas de escasez), es decir, pueden elegir hbitats a travs
de sus movimientos. En un contexto ecolgico, los requerimientos de las especies
se identifican por tanto en el marco de la seleccin de hbitat, un concepto que
desarrollar ms adelante.

El trmino Biologa de la Conservacin fue introducido como el ttulo de una conferencia que tuvo
lugar en la Universidad de California, San Diego en La Jolla, en 1978, organizado por los bilogos Bruce
Wilcox y Michael E. Soul. Con posterioridad, en 1987, se cre la primera sociedad cientfica profesional (The
Society for Conservation Biology); The Society is a response by professionals, mostly biological and social
scientists, managers and administrators to the biological diversity crisis that will reach a crescendo in the first
half of the twenty-first century. We assume that we are in time, and that by joining together with each other
and with other well-intentioned persons and groups, the worst biological disaster in the last 65 million years
can be averted. Although we have varying philosophies, we share a faith in ourselves, as a species and as
individuals, that we are equal to the challenge. For these reasons we join together in professional alliance, in
the service of each other, but also in the service of the less articulate members of our evolutionary tree (Soul
1987:4-5).

16

Introduccin

Figura 1. Tendencia en el uso del trmino Biologa de la Conservacin en las publicaciones cientficas.
Los resultados proceden de una bsqueda en la ISI Web of Knowledge (http://isiknowledge.com/), con el
trmino conservation biology como topic keyword. La lnea vertical se fij en el ao 1978.

La seleccin de hbitat es un concepto central en Ecologa, debido a su

influencia en la dinmica de las poblaciones y su persistencia, las interacciones entre

especies, y comunidades ecolgicas (Morris 2003). En el contexto de la Biologa de
la Conservacin, el estudio de la seleccin de hbitat tiene por tanto implicaciones
importantes. Adems, los modelos de hbitat son herramientas fiables para la
conservacin y manejo de especies amenazadas que permiten la identificacin de
hbitats relativamente6 idneos a proteger y/o gestionar adecuadamente para la
conservacin de las especies.

He utilizado el trmino relativamente para subrayar que hoy en da, debido a la alteracin humana
del paisaje, la expresin de hbitat idneo debe ser interpretada con cautela. En mi opinin, se refiere a lo
mejor dentro de lo malo ms que simplemente a una calidad de hbitat buena para la especie segn sus
requerimientos. En esta tesis emplear la idoneidad de hbitat en este sentido.

17

Introduccin

El objetivo de esta tesis es comprender los requerimientos ecolgicos de un

gremio7 de mamferos carnvoros silvestres y domsticos a nivel de poblacin en un

rea mediterrnea protegida.

Los mamferos constituyen un foco importante de conservacin en una tasa

desproporcionadamente alta teniendo en cuenta que slo representan un pequeo

porcentaje del nmero total de especies que existen en la tierra (alrededor de 4.500
especies de mamferos de ms de 1.000.000 de especies taxonmicas clasificadas
hasta 1970) (May 1992, Entwistle & Stephenson 2000). Los carnvoros, en
particular, constituyen un grupo de especies muy carismticas que se han utilizado
como especies bandera (flagships)8 en muchos programas de conservacin de
la biodiversidad y hbitats naturales. No obstante, su conservacin se enfrenta a
diversos problemas. Generalmente presentan bajas densidades poblacionales, bajo
rendimiento reproductivo, perodos de gestacin relativamente largos, necesitan
grandes superficies en un buen estado de conservacin (no toleran grandes zonas
urbanas o de alta actividad humana) y tienden a ser muy elusivos (lo que dificulta
su estudio) (Cardillo et al. 2004, Karanth & Chellam 2009, Schipper et al. 2008).
Adems, en algunos casos se consideran una amenaza para el ser humano o
sus actividades surgiendo as un conflicto de intereses (Inskip & Zimmerman
2009). Ese escenario de conflictos relega a estas especies en muchas ocasiones
a reservas naturales y reas adyacentes que deben permanecer ecolgicamente
intactas y su gestin deber adoptar un enfoque metapoblacional que traspase las
fronteras de las reservas para evitar problemas como la endogamia y fenmenos

empleado en el sentido de un grupo de especies que utilizan recursos similares y por lo tanto, pueden

competir.

aunque existen mltiples acepciones emplear la definicin de Heywood 1995; especies

carismticas que sirven como smbolos o iconos para estimular la conciencia conservacionista.

18

Introduccin

estocsticos (como brotes epidmico) (Fernndez et al. 2007) y asegurar as el

xito conservacionista. Muchas especies de carnvoros han sufrido una disminucin
dramtica en los ltimos pocos cientos de aos y algunas se encuentran entre los
mamferos terrestres ms amenazados del planeta (ej. Ceballos & Ehrlich 2002,
Rodrguez & Delibes 2003, Naves et al. 2003). La prdida de hbitat, el agotamiento
de sus presas, la explotacin de pieles y la persecucin directa han contribuido a la
disminucin de muchas de estas especies en el pasado (Woodroffe 2001). Hoy en da,
la persecucin directa y la disminucin de sus potenciales presas son las amenazas
ms inmediatas a corto plazo pero la prdida de sus hbitats naturales y la mortalidad
adicional reciente causada por el trfico son, probablemente, las mayores amenazas a
largo plazo para su persistencia (Ginsberg 2001, Kerley et al. 2002, Burkey & Reed
2006).

Afortunadamente, el reciente reconocimiento de la importancia ecolgica

de los depredadores y la falta de informacin para apoyar las estrategias de

recuperacin ha llevado a una mayor preocupacin por los efectos de los cambios
inducidos por el hombre en estas especies (Miller et al. 2001, Sergio et al. 2008). La
eliminacin de un carnvoro superior en un ecosistema puede tener un impacto en la
abundancia relativa de las especies presa y en el gremio de carnvoros en general (por
ejemplo, Palomares et al. 1996), causando efectos en cascada a travs de las cadenas
trficas hasta las plantas, afectando las interacciones entre las especies as como la
estructura de las comunidades ecolgicas y modificando los procesos bsicos de
funcionamiento de un ecosistema (por ejemplo, Estes & Palmisan 1974, Crooks
& Soul 1999, Duffy et al. 2007, Delibes-Mateos et al. 2008). Un ejemplo clsico
de ello es la disminucin de la densidad poblacional del puma (Puma concolor)
en el Parque Nacional Zion (Utah, EE.UU.), que condujo a un incremento en las
19

Introduccin

densidades de venados mula o bura (su principal especie presa), una mayor presin
de herbivora y en consecuencia un descenso en el reclutamiento de lamos de
rivera, un aumento en las tasas de erosin de los mrgenes de rivera y una reduccin
resultante de la abundancia de especies tanto acuticas como terrestres (Ripple &
Beschta 2006). Otro ejemplo relevante de su importancia es el proceso conocido
como liberacin de mesodepredadores (mesopradator release)9 (por ejemplo,
Palomares et al. 1998, Gehrt & Prange 2007). Es decir, los carnvoros presentan un
papel claro en el mantenimiento directo o indirecto de la biodiversidad, mediante el
control de mesodepredadores y diversificacin de las presas (Terborgh et al. 1999,
Miller et al. 2001). Estos estudios constituyen ejemplos de investigaciones holsticas
a nivel de ecosistema y estimulan enfoques similares en diferentes biomas, a fin de
comprender plenamente el papel de los mamferos carnvoros en el funcionamiento
de los ecosistemas y su representatividad como especies clave (keystone species)10.
Esto slo se puede lograr con investigaciones a largo plazo basadas en protocolos
estrictos de monitoreo de las especies y su medio ambiente (Yoccoz et al. 2001).

10

Las ideas relativas a la liberacin de mesodepredadores se remontan varias dcadas, cuando los
ecologistas comenzaron a observar que la eliminacin de depredadores originaba explosiones poblacionales
de otras especies inferiores (por ejemplo, Paine 1969, Pacala & Roughgarden 1984). El trmino fue acuado
por Soul et al. (1988) para describir un proceso mediante el cual las poblaciones de mamferos carnvoros
de tamao intermedio se hacan ms prevalentes en ausencia de un carnvoro superior, y las poblaciones de
diversas aves se vean deprimidas como consecuencia de ello. En esta tesis empleo el trmino de una manera
ms amplia segn Brashares et al. (2010) para definir la expansin en densidad o distribucin, o el cambio en
comportamiento de un predador de rango medio como resultado de la disminucin en densidad o distribucin
de un predador superior. Aunque la liberacin de mesodepredadores se emplea normalmente en el contexto de
la teora trfica de cascadas (por ejemplo, Berger et al. 2008, Brashares et al. 2010), se trata esencialmente de
una interaccin intragremial entre depredadores.

Paine 1969; especies cuya presencia es crucial en el mantenimiento de la organizacin y

diversidad de las comunidades ecolgicas; especies excepcionales, en relacin con el resto de la comunidad,
en su importancia.

20

Introduccin

Marco terico
Teora de la seleccin de hbitat

Los patrones de distribucin de las poblaciones en el medio natural son

el resultado de procesos que ocurren a diferentes escalas espaciales. La eleccin

individual de las caractersticas del medio ambiente es una de las fuerzas motrices
que opera a escala fina (Turchin 1998). Esta eleccin individual es inherente al
concepto de seleccin de hbitat, definido por Johnson (1980) como el proceso
por el cual un animal elige los componentes del hbitat a usar. No obstante el
concepto de hbitat debe definirse y aclararse antes de desarrollar el concepto de
seleccin de hbitat.

El concepto de hbitat

Aunque el concepto de hbitat es fundamental en Ecologa, carece de una

clara y consistente definicin, a pesar de los numerosos esfuerzos de unificacin

(Whittaker et al. 1973, Hall et al. 1997, Morris 2003, Kearney 2006), y su utilidad
es a veces incluso controvertida (Mitchell 2005). Bsicamente, este trmino se usa
a menudo para describir el medio fsico de las especies, poblaciones o individuos
a diferentes escalas espaciales. El hbitat se considera a veces slo como una
descripcin de la naturaleza fsica de un lugar (componentes abiticos y biticos)
donde un organismo vive o puede potencialmente vivir (Kearney 2006, Morrison et
al. 2006). En ocasiones la definicin de hbitat incluye los conceptos de persistencia
de especies/poblaciones o la supervivencia y reproduccin de los individuos
(Whittaker et al. 1973, Hall et al. 1997, Morris 2003). Sin embargo, todava no
existe un consenso sobre las propiedades especficas individuales del hbitat, ya
que esto relaciona la presencia de una especie, poblacin o individuos con las
21

Introduccin

caractersticas fsicas y biolgicas de un rea (Hall et al. 1997). En mi opinin,

la definicin de este concepto depende en gran medida del contexto en el que se
emplea. Cuando estamos hablando del hbitat de una especie (o de una poblacin o
individuo) desde una perspectiva evolutiva, es evidente que el hbitat debe incluir el
fitness individual o la persistencia de las especies/poblaciones. En este contexto, la
respuesta a la pregunta: Cul es el hbitat de esta especie? incluye implcitamente
la nocin de buen hbitat en el que la especie puede persistir. Sin embargo, los
estudios de seleccin de hbitat rara vez consideran o miden ningn componente
relacionado con el fitness (pero ver McLoughlin et al. 2007), ya que es difcil
relacionar estas medidas con el hbitat de las especies. De hecho, los investigadores
tradicionalmente describen el hbitat de una especie, relacionando la presencia de
individuos, poblaciones o especies con las caractersticas de un rea, asumiendo
que la presencia de la especie es un buen indicador de la calidad del hbitat. En este
contexto, el concepto de hbitat se emplea en un sentido ms espacial. Por ejemplo,
consideremos una poblacin en una dinmica de fuente-sumidero (para ms detalles
vase Pulliam 1988), donde la poblacin fuente (cuya reproduccin local es mayor
que la mortalidad local) vive en una zona con caractersticas ambientales diferentes
que la poblacin sumidero (cuya reproduccin local no compensa el ritmo de la
mortalidad local). Sin la poblacin fuente, la poblacin sumidero no persistira.
Por tanto, el rea donde vive la poblacin sumidero (a menudo denominado como
hbitat sumidero) no sera un hbitat para la especie en el contexto de la definicin
de rendimiento o persistencia.

Probablemente sea imposible unificar estos conceptos, por lo que ms bien

debera definirse el concepto de hbitat especficamente para cada estudio, segn

el contexto. Creo que la idea de persistencia o rendimiento de las especies debe
22

Introduccin

referirse al concepto subyacente de calidad del hbitat, la capacidad del medio

ambiente para proporcionar las condiciones apropiadas para la persistencia del
individuo y la poblacin (Hall et al. 1997). En este contexto, los hbitats sumidero
siguen siendo hbitats, pero de mala calidad, induciendo una alta mortalidad o una
baja reproduccin. Por lo tanto, preferira una definicin de hbitat que no incluyera
la nocin de persistencia, sino ms bien las caractersticas fsicas y biolgicas de un
rea en la que una especie (poblacin o individuo) puede vivir. Sin embargo, prefiero
pedir prestado parte de la definicin proporcionada por Whittaker (1973), quien
destac la propiedad multivariante de un hbitat: Las m variables del ambiente
fsico [biolgico] y qumico que forman gradientes espaciales en un paisaje o rea
definida como ejes en un hbitat representado como un hiperespacio. La parte de
este hiperespacio que ocupa una determinada especie [en una escala particular
de espacio y tiempo] representa un hipervolumen que define su hbitat. Esta
definicin se acerca bastante a la definicin de nicho ecolgico (vase la seccin
siguiente y Figura 2). Por lo tanto, en esta tesis trato de interpretar la seleccin del
hbitat de los animales haciendo con cautela inferencias acerca de la calidad del
mismo (aunque al tratarse de un rea protegida, a pesar de que se llevan y se han
llevado a cabo de manera tradicional ciertos usos humanos del paisaje en la zona de
estudio, la alteracin en cuanto a fragmentacin y alteracin del mismo se supone
menor que en otro tipo de reas ms humanizadas).

23

Introduccin
E1
Niche
(used)

Environment
(available)

E3

E2

Figura 2. Representacin esquemtica del nicho ecolgico. Las flechas negras representan variables
ambientales (e.g. cobertura de matorral, disponibilidad de presas), y por tanto el espacio ecolgico.
La elipse gris oscuro se corresponde con los valores de esas variables que estn disponibles para
la especie (poblacin o individuos). La elipse gris claro representa el rango de valores usado por la
especie, es decir, su nicho ecolgico.

Siguiendo esta definicin, un parche de hbitat describir un subconjunto

del hbitat de la especie, es decir, una combinacin particular de los componentes

(las variables del hbitat) que constituyen el hbitat de la especie. De acuerdo con
la cuestin de inters, las variables del hbitat pueden referirse como variables
ambientales (por ejemplo, elevacin, tipo de vegetacin, condiciones climticas),
pero tambin puede integrar sus congneres (por ejemplo, la densidad de poblacin)
u otras especies (por ejemplo, la densidad de presas para un depredador). El trmino
parche se utiliza a menudo para describir reas delimitadas que contienen una
cantidad limitada de recursos agregados en un ambiente pobre en recursos a
mayor escala (Cezilly & Benhamou 1996). Estos conceptos se pueden utilizar
para diferentes entidades, tales como especies, poblaciones, individuos o incluso
comunidades.

24

Introduccin

El concepto de nicho ecolgico

Este concepto tambin sufre de la falta de una definicin unificada y se confunde

a menudo con el concepto de hbitat. Fue desarrollado por primera vez por Grinnell
(1917) para referirse a todas las caractersticas del medio ambiente que le permiten a
una especie sobrevivir y reproducirse (ntese la similitud con la definicin de hbitat
mencionada, por ejemplo, Hall et al. 1997). Posteriormente, Elton (1927) introdujo
el papel funcional de la especie dentro de su comunidad, en su nueva definicin
del concepto. Estos autores estn detrs de las controversias pasadas y actuales.
Tenemos en cuenta el impacto de las especies en su medio ambiente y comunidad o
slo el efecto del ambiente en la especie, es decir, el efecto de factores limitantes11en
la especie? Esto tambin depende del contexto. En 1957, Hutchinson formaliz el
concepto de nicho con un modelo geomtrico. Defini el nicho como el hipervolumen
en el espacio multivariado de variables ambientales (el espacio ecolgico, Figura 3)
donde una especie puede persistir (Figura 2). Esta definicin hace hincapi en la gama
de condiciones ambientales necesarias para la persistencia de la especie, es decir, el
nicho Grineliano12 (que es similar al concepto de hbitat). En este contexto, el nicho
ecolgico representa la posicin de la especie en la gama de condiciones ambientales,
de manera que cada dimensin del nicho se corresponde por tanto con un subconjunto
de este rango potencial o realmente importante para la especie. Hutchinson no obstante
reconoci el papel potencial de la especie en su comunidad mediante la descripcin de
dos tipos de nicho: el nicho fundamental y el nicho realizado.

11

los factores limitantes son cualquier proceso [o factores] que afectan de una manera cuantificable
el crecimiento de una poblacin (Messier 1991), tal y como recursos trficos, refugios o condiciones climticas.
.
12

definido por Grinnell 1917.

25

Introduccin

E1
(a)

(b)

E3

E2

Figura 3. (a) Espacio geogrfico y (b) ecolgico. La localizacin (normalmente definida por dos
coordenadas en el espacio; longitud y latitud) de una especie se emplea con frecuencia para analizar
sus propiedades ecolgicas en el espacio ecolgico de variables ambientales (E1 a E3). (Adaptado
de Calenge 2005).

El primer trmino corresponde al lugar ocupado por una especie en ausencia

de competencia. Sin embargo, el nicho fundamental rara vez se produce en la

naturaleza, ya que los ecosistemas estn compuestos por conjuntos de especies
que coexisten e interactan entre ellas. Por lo tanto, la presencia de una especie
no indica necesariamente que el hbitat sea el ptimo, pero es el resultado de un
trade-off entre la calidad del hbitat y la competencia intra e interespecfica que
limita los recursos e interfiere el acceso a ellos (Van Horne 1983, Araujo & Guisan
2006, Sobern 2007). El segundo trmino tiene en cuenta dichas interacciones y
hace referencia a la distribucin de la especie en su entorno, dada la presencia de
competidores, y por tanto se asume como ms estrecho que el nicho fundamental.
Este concepto nos lleva a la idea de la particin de nicho, el mecanismo que permite
la coexistencia entre especies que habitan el mismo biotopo13 (Rosenzweig 1981).

Un biotopo es un rea fsica con condiciones ambientales uniformes dnde viven un conjunto
especfico de plantas y animales.

26

13

Introduccin

Como resultado de la exclusin competitiva (Gause 1934), dos o ms especies que

viven en la misma zona y tienen requisitos similares y utilizan los mismos factores
limitantes, pueden modificar el uso de recursos (al menos una de las especies,
Gause 1934, Rosenzweig 1981), aunque esta idea est sujeta a controversia en la
literatura (Araujo & Guisan 2006). El uso de los conceptos nicho fundamental y
nicho realizado es frecuentemente confuso en la literatura (Sobern 2007) y su
utilidad ampliamente debatida (Araujo & Guisan 2006).

En esta tesis he utilizado el enfoque de Hutchinson para definir el nicho

ecolgico, ya que est ntimamente ligado al concepto de hbitat (cualquiera que

sea la definicin de hbitat, ambas se basan en la relacin entre una especie y las
caractersticas del medio ambiente). Aunque los conceptos de nicho ecolgico y
hbitat estn relacionados con el espacio ecolgico, a menudo estn relacionados
con el espacio geogrfico (Figura 3, Calenge 2005, Araujo & Guisan 2006). De
hecho, el estudio de la ubicacin de las especies en su espacio geogrfico permite
la identificacin de sus propiedades ecolgicas, y la asociacin de las propiedades
ecolgicas con factores espacialmente explcitos da lugar a la distribucin
potencial de la especie (Araujo & Guisan 2006). En el marco de este concepto
se han desarrollado diversos mtodos en relacin con el objetivo de relacionar la
distribucin de las especies con su medio ambiente (Hirzel et al. 2002, Calenge et
al. 2005, Basille et al. 2008, Calenge & Basille 2008, Calenge et al. 2008). Algunos
de estos mtodos utilizan dos interesantes propiedades del nicho: la marginalidad
y especializacin de las especies. La marginalidad es la posicin de las especies en
los gradientes ambientales disponibles. Por lo tanto, se refiere a la excentricidad del
nicho en comparacin con el gradiente de componentes ambientales (Calenge et al.
2005). Por lo tanto, una especie marginal se encontrar en condiciones ambientales
27

Introduccin

ms atpicas (valores extremos del gradiente de una variable), mientras que las
especies no marginales usarn condiciones ambientales medias. La especializacin
es la anchura del nicho, es decir, el grado de tolerancia de la especie al gradiente
ambiental. Cuanto ms grande sea el nicho, mayor tolerancia presentar la especie,
mientras que cunto ms estrecho sea, ms especializada estar la especie en el
uso de ciertos recursos. Estos conceptos son particularmente tiles para describir y
cuantificar la relacin entre una especie y el medio ambiente disponible para ella.
En los ltimos aos, se han desarrollado numerosos anlisis para la estimacin del
nicho ecolgico (Guisan & Zimmermann 2000, Calenge & Basille 2008). Aunque
este concepto ha sido desarrollado y utilizado a nivel de especie o poblacin,
tambin puede ser generalizado a nivel individual.

El estudio de la seleccin de hbitat

Los estudios de seleccin de hbitat para identificar las caractersticas

ambientales que selecciona una especie, en el supuesto de que estas caractersticas

se han seleccionado debido a que proporcionan las mejores condiciones para la
supervivencia y la reproduccin (Thomas & Taylor 2006), se abordan cada vez ms
desde diferentes disciplinas (Evolucin, Ecologa y Conservacin). La seleccin de
hbitat generalmente se investiga utilizando datos sobre el uso del espacio de una
especie dada. Las caractersticas del hbitat utilizado por la especie se comparan con
las de las zonas no utilizadas y ms comnmente con las de reas que se consideran
como disponibles para la especie (Thomas & Taylor 1990, Manly et al. 2002). Se
dice que un hbitat es seleccionado cuando la proporcin utilizada por los animales
es mayor que la proporcin disponible. Por el contrario, se dice que un hbitat se
evita cuando la proporcin de uso es menor que la proporcin disponible. Sin
28

Introduccin

embargo, como se indic anteriormente las interpretaciones resultantes de estos

comparaciones requieren cierta precaucin, porque no son una medida directa de
la calidad del hbitat (lo que requerira algn tipo de medida relacionada con el
fitness14). Sin embargo, aunque la densidad de individuos podra ser un indicador
pobre de la calidad del hbitat en algunas condiciones (Van Horne 1983), la mayora
de las veces es un buen indicador de la idoneidad (calidad) de un rea en particular.

Escalas espacio-temporales de investigacin

La naturaleza misma de la seleccin de hbitat es jerrquica. Diferentes

procesos actan a diferentes escalas espacio-temporales, lo que resulta en una

seleccin diferencial del hbitat de acuerdo con la escala bajo consideracin. Las
caractersticas de un rea en la que se distribuye una poblacin de una determinada
especie (seleccin de primer orden podran no ser congruentes con las caractersticas
de los hbitats disponibles dentro del rea de campeo de los individuos (seleccin
de tercer orden), porque los mecanismos implicados no son los mismos. La eleccin
de la escala de investigacin (por ejemplo el rango geogrfico de la especie, el
rea ocupada por una poblacin, la seleccin de hbitat individual dentro de las
reas de campeo, etc) es crucial, ya que las inferencias a partir de los anlisis a
una escala particular, estn limitada a esa escala (Pendleton et al. 1998). Adems,
la importancia de una escala particular puede ser diferente segn la especie de
estudio. Para las especies generalistas, por ejemplo, las escalas grandes (escala de
paisaje, por ejemplo) podran ser menos importantes que las escalas ms finas (por
ejemplo, la seleccin de hbitat o recursos dentro del rea de campeo) que para las

14

aptitud,adecuacin oeficacia biolgica

29

Introduccin

especies especialistas, para las que el hbitat en el rango geogrfico puede ser de
crucial importancia. Sin embargo, como la mayora de los procesos ecolgicos, la
seleccin de hbitat a menudo se produce a ms de una escala (Levin 1992).

Thomas & Taylor (1990) propusieron diferentes diseos de estudio para

las comparaciones entre hbitats usados y hbitats disponibles (o no usados)

teniendo en cuenta el organismo de estudio (poblacin, individuos) y la escala
de investigacin. El diseo de tipo I se utiliza para estudios a nivel de poblacin
(seleccin de primer orden) cuando no se identifican los individuos. El uso del
hbitat y la disponibilidad de hbitat se miden a nivel de poblacin. Los datos se
supone que son independientes (la presencia de la especie en un sitio particular
no debe influir en su presencia en otros lugares) y el acceso a los recursos es igual
para todos los individuos (deben por lo tanto, seguir una distribucin libre ideal15
(Fretwell & Lucas 1969). Para este diseo se suelen emplear ndices de presencia
de las especies (por ejemplo, heces, pelos, rastros).

Los diseos de tipo II, III y IV se utilizan para estudios a nivel individual

(seleccin de 2, 3 y 4 orden). Los individuos se identifican (por ejemplo,

utilizando telemetra) y los datos de uso se miden por separado para cada uno de
los individuos. En el diseo de tipo II, la disponibilidad de hbitat es la misma
para todos los individuos (por ejemplo, la composicin de reas de campeo dentro
del rango geogrfico de distribucin de las especies), mientras que en los diseos
de tipo III y IV, la disponibilidad de hbitat se mide de forma independiente para
cada individuo. La disponibilidad de hbitat es constante durante el perodo de
estudio en el diseo tipo III, por lo que se define por una medida (por ejemplo,
15

La teora de la distribucin libre (ideal free distribution; IFD) formula que el nmero de individuos
que se agregan en varios parches es proporcional a la cantidad de recursos disponibles en cada uno y predice
que la distribucin de los animales entre parches minimizar la competencia por los recursos y maximizar el
fitness.

30

Introduccin

el rea de campeo de un individuo). En el diseo IV, hay un cambio temporal en

la disponibilidad de hbitat para un individuo dado, lo que requiere una serie de
medidas de disponibilidad (una medida por cada localizacin de un individuo). Este
ltimo diseo fue creado con posterioridad por Erickson et al. (2001) para tener en
cuenta la creciente utilizacin de los nuevos tipos de datos proporcionados por la
telemetra, que permiten localizaciones frecuentes de los animales (por ejemplo,
una localizacin cada 30 minutos). En esta tesis empleo un diseo de estudio tipo
I para determinar los patrones de seleccin de hbitat de especies de carnvoros
silvestres y domsticos en un rea protegida.

De lo realmente disponible e importante para los individuos

Medir la disponibilidad de hbitat para una especie requiere tener en cuenta

dos aspectos importantes: la eleccin de las variables biolgicas significativas para

la especie y los lmites de la zona que consideraremos como disponible para la
especie en cuestin. La eleccin de las variables de hbitat a integrar en los anlisis
es una tarea difcil (Lennon 1999, Guisan & Zimmermann 2000), y debe basarse
en un conocimiento previo profundo de la especie. Todas las variables de hbitat
consideradas limitantes para la especie deben incluirse. Sin embargo, la eleccin de
variables a menudo se basa en consideraciones logsticas, ya que algunas variables
son difciles de medir (Mitchell 2005).

Como se mencion anteriormente, la eleccin de la escala es muy importante

y conduce al problema inherente de definicin de la disponibilidad de hbitat. En

los estudios de seleccin de hbitat, la determinacin de lo que es disponible para la
especie es un reto, ya que slo los animales saben lo que est realmente disponible.
En teora, los investigadores debemos definir objetivamente la disponibilidad de
31

Introduccin

hbitat, desde el punto de vista de la especie. Esto es crtico para la interpretacin

de los anlisis, porque al cambiar la disponibilidad cambiar la proporcin de cada
uno de los componentes hbitat, y por tanto, la comparacin entre la proporcin
de uso y disponibilidad de este componente, especialmente si este componente
muestra agregacin espacial (Porter & Church 1987). En la prctica, sin embargo, la
disponibilidad se define a menudo subjetivamente, debido a la dificultad de evaluar
la percepcin que tiene la especie del medio. Por ejemplo, una zona puede aparecer
disponible para un individuo dado, mientras que las interacciones interespecficas
(por ejemplo, la presencia de depredadores) o interacciones intraespecficas (por
ejemplo, la defensa del territorio) podra prevenir su uso o su acceso, respectivamente.
En algunos casos, la definicin de disponibilidad debe ser reducida si la cuestin de
inters se refiere a los componentes fsicos del hbitat, o si variables referidas a sus
congneres, presas o depredadores deben ser incluidas.

Las escalas de seleccin definidas por Johnson (1980) ayudan a reducir esta

subjetividad, ya que tienen una base biolgica, pero no la eliminan por completo
(Erickson et al. 2001). Por ejemplo, en la escala de establecimiento del rea de
campeo en la distribucin geogrfica de una especie, los lmites del rea de estudio
a menudo abarcan el rea en la que se distribuye la poblacin.

Caractersticas ecolgicas de las especies y mecanismos de coexistencia

La seleccin natural ha inducido la aparicin de estrategias ms o

menos especializadas entre especies presentando un trade-off evolutivo entre la

especializacin para realizar unas pocas actividades eficazmente o el generalismo
para desarrollar muchas actividades de una manera menos efectiva (Levins 1968).
Algunas especies muestran amplias tolerancias ambientales y presentan una dieta
32

Introduccin

muy variada (generalistas de hbitat y dieta), mientras que otras tienen tolerancias
ambientales muy especficas y estrechas y slo consumen ciertos recursos
(especialistas de hbitat y dieta). Estas dos categoras de especies tienen diferentes
dinmicas poblacionales (Kolasa & Li 2003). Por ejemplo, la variacin en la
densidad de poblacin es mayor en los especialistas que en los generalistas (Kolasa
& Li 2003). Del mismo modo, los especialistas de hbitat utilizan unidades de
hbitat ms pequeas anidadas dentro de unidades de hbitat ms grandes (Kolasa
& Pickett 1989). Esto tiene otra consecuencia ya que las especies que utilizan
pequeas unidades de hbitat tienden a tener bajas densidades poblacionales como
consecuencia de la disminucin de la eficiencia en la bsqueda de los parches
adecuados y de la mortalidad durante la dispersin (Kolasa & Romanuk 2005).
As, la disponibilidad de hbitats adecuados parece afectar ms a las especies
especialistas que a las ms generalistas, las cuales utilizan una gama ms amplia
de tipos de hbitats para satisfacer sus necesidades (Munday et al. 1997, Bean et
al. 2002). Obviamente existe un gradiente continuo en el nivel de especializacin
de una especie entre el especialismo ms extremo (como el caso del lince ibrico;
Lynx pardinus) y el completo generalismo (como el caso del zorro comn; Vulpes
vulpes).

La definicin operativa de especializacin que usar en esta tesis ser el uso

de un subconjunto relativamente restringido de recursos o hbitats en comparacin

con otras especies. Las especies especialistas suelen beneficiarse de ambientes
relativamente homogneos o estables (en el espacio y/o tiempo) mientras que las
especies generalistas suelen beneficiarse de ambientes ms heterogneos (Futuyma
& Moreno 1988, Kassen 2002, Marvier et al. 2004, Ostergard & Ehrln 2005).

Las caractersticas ecolgicas particulares de cada especie pueden afectar

33

Introduccin

la eficiencia de la idoneidad del hbitat o los modelos de distribucin de especies

(Stockwell & Peterson 2002). Por ejemplo, Hepinstall et al. (2002) sugieren que la
amplitud de nicho de las especies es importante porque para las especies generalistas
que utilizan diferentes hbitats se podra predecir su ocurrencia en todas partes
por los mtodos de asociacin de hbitat, mientras que las especies con nichos
estrechos son ms propensas a ser predichas con mayor exactitud. Tsoar et al.
(2007), adems, encontr que los rangos de distribucin de las especies con nichos
ecolgicos estrechos se puede modelar con mayor precisin que los de las especies
ms generalistas. Otros autores (por ejemplo Cowley et al. 2000, Hepinstall et al.
2002, Brotons et al. 2004, Hernndez et al. 2006, Brotons et al. 2007) tambin han
sealado que las especies con nichos ecolgicos restringidos pueden ser modeladas
con mayor precisin que las especies ms generalistas. Brotons et al. (2007) sugiere
que las especies que tienen distribuciones ms amplias o utilizan una amplia variedad
de hbitats en un rea no deben estar limitadas por los factores predictivos medidos
a la escala a la que se ajustan los modelos. Cowley et al. (2000) tambin encontr
que los modelos con mayor rendimiento son los de especies sedentarias que tienen
fuertes asociaciones de hbitat y que presentan una distribucin generalizada en
esos hbitats.

El mecanismo por el que especies con diferentes historias de vida o

caractersticas ecolgicas conviven en la misma rea ha sido foco de estudio en

ecologa. Diversos autores han discutido en detalle como la heterogeneidad espacial
promueve la coexistencia entre especies pertenecientes al mismo nivel trfico (Levin
1974, Yodzis 1978, Hastings 1980). Los primeros modelos tericos mostraron que
dos especies que comparten recursos no pueden coexistir en un solo parche, pero
pueden hacerlo cuando dos o ms parches diferentes estn presentes (Levin 1974).
34

Introduccin

En la dcada de 1980 el concepto de que la heterogeneidad espacial promueve la

coexistencia fue estudiada con ms detalle en una teora desarrollada para explicar
la coexistencia entre especies basndose en las teoras de la seleccin ptima de
los recursos (optimal foraging theory) y la seleccin de hbitat, conocida como la
teora Isoleg (Pimm & Rosenzweig 1981, Rosenzweig 1981). Esta teora explica
cmo dos especies competidoras se distribuyen en hbitats de diferente calidad
segn sus densidades intra e interespecficas. Una de las principales predicciones de
la teora isoleg es que la coexistencia se ve favorecida cuando una de las especies
competidoras es un especialista (comportndose por tanto de manera selectiva),
mientras que el otro es un generalista (actuando por tanto de forma oportunista)
(Rosenzweig 1987). Cuando el especialista es tambin la especie dominante, se
prev que la coexistencia se vea favorecida en reas con una diversidad suficiente
de hbitats en las que la especie generalista subordinada pueda segregarse de la
especie especialista dominante. Un concepto similar se desarroll con anterioridad
para los sistemas predador-presa las presas pueden buscar la seguridad frente a
los depredadores en zonas conocidas como refugios de depredacin, lo que puede
ser crucial para la persistencia de ambas (Hassell & May 1973). Este principio
se puede aplicar igualmente a la competencia interespecfica y, en un entorno
heterogneo, las especies con baja capacidad competitiva puede persistir mediante
el uso de refugios donde la competencia se reduce (Durant 1998).

No obstante, en reas aparentemente homogneas desde una perspectiva

o escala mayor, los mecanismos de coexistencia que deben desarrollar especies

similares (es decir, pertenecientes al mismo nivel trfico) que les permitan segregarse
y por tanto subsistir no resultan tan evidentes. Estos mecanismo pueden reflejarse
en patrones diferenciados de uso del hbitat relativamente homogneo a una escala
35

Introduccin

ms fina (a escala de parche) y debe verse favorecido por la presencia de especies

con diferentes grado de especializacin en el gremio. Conocer el efecto del hbitat
en la coexistencia de las especies puede ser especialmente relevante cuando puede
afectar a las polticas de conservacin y gestin, incluso en reas protegidas.

De cmo simplemente encontrar el hbitat ms idneo puede no ser suficiente

Como ya referimos con anterioridad con las ideas Hutchinsonianas de nicho

fundamental y nicho realizado, la presencia de una especie no indica necesariamente

que el hbitat sea el ptimo debido a la existencia de un trade-off entre la calidad
del hbitat y las interacciones intra e interespecfica que limitan los recursos e
interfieren el acceso a ellos. As, los modelos de seleccin de hbitat puros16 no
siempre reflejan la idoneidad del hbitat para una especie, o al menos para aquellas
especies para las cuales la presencia de otras especies pertenecientes al mismo nivel
trfico y las interacciones con ellas puedan alterar su uso del espacio y limitar su
abundancia o distribucin (e.g. Laurenson 1995, Lindstrm et al. 1995).

As, bajo condiciones de competencia por el uso de los recursos17, la densidad

de individuos puede ser un indicador pobre de la calidad del hbitat para algunas
especies.

La competencia se suele clasificar tanto en competencia por explotacin

como en competencia por interferencia (Park 1962). La competencia por explotacin

se produce cuando una especie utiliza un recurso (por ejemplo consume una

con modelos de seleccin de hbitat puros hago referencia a modelos en los que no se tienen en
cuenta las potenciales interacciones competitivas existentes entre especies pertenecientes al mismo nivel trfico
que la especie bajo estudio, sino solamente las caractersticas anatmicas del paisaje que las rodea.
16

17

entindase como recurso cualquier fuente de la cual la especie obtenga un beneficio, es decir tanto
las caractersticas del paisaje como la vegetacin o disponibilidad de presas, como el propio espacio en s
mismo.

36

Introduccin

presa especfica) y con ello reduce la oportunidad de usar ese mismo recurso a
otra especie. La competencia por interferencia implica sin embargo interacciones
comportamentales entre las especies como la predacin intragremial (Polis et al.
1989) en el caso ms extremo o bien mecanismos ms sutiles como evitar los
hbitats ms usados por el predador principal y mostrar preferencias por hbitats
menos productivos (Harrison et al. 1989, Thurber et al. 1992, Durant 1998, Fedriani
et al. 1999, Fuller & Keith 1981), ajuste de los patrones de actividad para reducir
los encuentros con el predador principal (Litvaitis 1992, Johnson et al. 1996) o
bien formar grupos para competir de una manera ms exitosa por los recursos
y/o obtener ventajas antipredadoras (Kruuk 1975, Eaton 1979, Lamprecht 1981,
Gittleman 1989).

El tamao corporal se considera normalmente el factor ms influyente

que determina la direccin y fuerza de la dinmica intragremial (Polis et al. 1989,

Donadio & Buskirk 2006), siendo las especies de tamao superior capaces de
excluir a las ms pequeas de los parches de hbitat o recursos trficos.

Las especies de carnvoros de pequeo y mediano tamao a menudo se

ven perjudicados por interacciones agresivas con otros miembros simptridos de

mayor tamao del gremio al que pertenecen (Palomares et al. 1996, Crooks & Soule
1999, Palomares & Caro 1999, Fedriani et al. 2000, Donadio & Buskirk 2006). Las
interacciones agresivas o predacin intragremial entre los mamferos carnvoros
es muy frecuente y en algunas especies, adems de suponer un cambio en el uso
del hbitat de las especies menos aventajadas competitivamente tiene tambin un
impacto considerable en las tasas de mortalidad (Ralls & White 1995, Sovada et
al. 1995) y por tanto en las densidades relativas de dichas especies en presencia
de un competidor superior. En esta tesis analizaremos la existencia de potenciales
37

Introduccin

interacciones agresivas o predacin intragremial entre las especies que conforman

el gremio de mamferos carnvoros de nuestra rea de estudio para determinar cmo
predadores inferiores pueden mostrar una respuesta numrica negativa reflejada en
sus densidades relativas en presencia de un predador superior.

38

Introduccin

OBJETIVOS DE LA TESIS DOCTORAL

El captulo I est dedicado a consideraciones metodolgicas, particularmente
en lo relativo a la necesidad de tener en cuenta variables metodolgicas y/o
climticas en los censos de rastros en sustratos arenosos y cmo su inclusin como
variables adicionales en modelos de seleccin de hbitat puede mejorar la fiabilidad
de los resultados obtenidos.

Los objetivos biolgicos de esta tesis cubren la seleccin de hbitat a nivel

de rea de campeo de los individuos y la biologa de la conservacin y se presentan

en los captulos del II al V.

En el captulo II, el objetivo es determinar los patrones de seleccin de

hbitat a escala fina de las distintas especies de carnvoros silvestres considerndolas

en su conjunto. Particularmente el objetivo consiste en determinar cmo especies
con diferentes historias de vida (es decir, grado de especialismo de hbitat y
dieta) pueden convivir en un rea estudiada relativamente homognea a escala de
macrohbitat. Nos centramos en determinar el espacio ecolgico ocupado por cada
una de las especies (nicho ecolgico) dentro del hiperespacio representado por las
variables ambientales medidas en el rea de estudio y en evaluar cmo factores
relacionadas con el tipo de vegetacin, la disponibilidad de presas, la estructura del
paisaje o las perturbaciones humanas pueden afectar a los patrones diferenciales de
seleccin de hbitat entre las distintas especies a una escala fina.

En el captulo III pretendemos determinar si la presencia de un predador

superior puede afectar negativamente las densidades relativas de especies inferiores

una vez controlado por la seleccin de hbitat a escala fina de cada especie; efecto
potencialmente atribuible a la existencia de depredacin intragremial.
39

Introduccin

En el captulo IV nos planteamos determinar si las rastros de perros

detectados durante los censos se correspondan con perros domsticos procedentes

de la matriz humanizada circundante al rea protegida o se correspondan con perros
asilvestrados residentes en el interior del Parque, as como en evaluar las variables
ambientales y/o relacionadas con la presencia humana que determinan su uso del
espacio. Asimismo, discutimos las implicaciones de los resultados en trminos de
gestin y manejo de especies domsticas en el rea.
El captulo V est dedicado a la presencia de gato monts en el Parque
Nacional. Esta especie resulta de especial inters debido a que no se han llevado a
cabo estudios relacionados con su presencia, abundancia y/o seleccin de hbitat
con anterioridad en el rea protegida asumindose una especie muy escasa a pesar
de la potencial idoneidad del hbitat para la especie. En este captulo realizamos
un estudio de fototrampeo para determinar su presencia y estatus de conservacin.
Asimismo, discutiremos las posibles razones relacionadas con el actual estado de
la poblacin en el rea. Estos resultados sirven adems como una herramienta para
ayudar a los gestores en la toma de decisiones en materia de conservacin y gestin
de la poblacin de dicha especie.

40

Introduccin

ESPECIES & REA DE ESTUDIO

Las especies modelo

Esta tesis se centra en el estudio de un gremio de carnvoros compuesto

por cuatro especies nativas (el lince ibrico Lynx Pardinus, el tejn europeo
Meles meles, el zorro comn Vulpes vulpes y el gato monts Felis silvestris) y dos
introducidas en tiempos histricos (la gineta comn Genetta genetta y el meloncillo
Herpestes ichneumon) (Figura 4) en un rea protegida del suroeste de Espaa; el
Parque Nacional de Doana.

El zorro comn (5-7 Kg.), debido a su marcado generalismo de hbitat y

dieta (Mitchell-Jones et al. 1999, Pita et al. 2009), es la especie ms extendida en el

Mediterrneo. Presenta una alta resistencia ecolgica gracias a su alta movilidad y
capacidad reproductiva (Blanco 1998, Piero 2002) as como una alta flexibilidad
en su ecologa espacial; puede ocupar cualquier tipo de hbitat que le ofrezca un
mnimo de refugio y alimento adaptndose a cualquier tipo de cambio (Fedriani
1996, Piero 2002). Como depredador oportunista, consume una amplia variedad de
recursos trficos y se alimenta de acuerdo con la disponibilidad de presas (Carvalho
& Gomes 2001, Cavallini & Lovari 1991, Delibes-Mateos et al. 2008).

El tejn europeo (7-8 Kg.) es un carnvoro social y territorial con una amplia

distribucin en el Palertico (Revilla & Palomares 2005). De actividad en su mayor

parte nocturna, es considerado como un generalista trfico por la gran variedad
de recursos disponibles de los que hace uso (lombrices, insectos, frutos, gazapos;
Revilla & Palomares 2002). Pese a que en Europa central muestra preferencias por
los bosques templados y pastos asociados, se ha demostrado que tambin selecciona
las masas arbustivas como hbitat principal en climas ms secos, como es el caso
del suroeste de la Pennsula Ibrica (Revilla et al. 2000). Su presencia parece
41

Introduccin

condicionada por la existencia de cobertura vegetal que oculte las madrigueras

(Revilla & Palomares 2005). El tamao del rea de campeo puede variar mucho
entre individuos y en funcin de la productividad de los hbitats ocupados, siendo
el tamao medio de 525 hectreas en el suroeste de la Pennsula Ibrica (Rodrguez
et al. 1996).

El meloncillo (3 Kg.) es un carnvoro diurno procedente de frica (Riquelme

Cantal-et al. 2008) cuya distribucin en Europa se restringe al suroeste de la

Pennsula Ibrica. Se alimenta en grupos presentando una dieta de amplio espectro
que incluye conejos, roedores, aves, reptiles y carroa (Palomares & Delibes 1991,
Palomares 1993, Zapata et al. 2007). Debido a su carcter diurno, la especie muestra
preferencias por zonas de matorral denso con una alta densidad de presas y refugio
evitando las zonas abiertas, por lo que se le ha considerado en cierto grado como
especialista de hbitat en zonas mediterrneas (Palomares & Delibes 1990, 1993).
El tamao medio de su rea de campeo es de 300 hectreas (Palomares 1994).

Al igual que en el caso del meloncillo, la gineta (2 Kg.) tambin en una especie

introducida de frica (Riquelme-Cantal et al. 2008). Es un carnvoro nocturno que

se alimenta de pequeos mamferos, aves, reptiles y artrpodos, aunque la presa
principal puede variar en diferentes zonas de distribucin (Virgs et al. 1999). En
nuestra zona de estudio, el Parque Nacional de Doana, el anlisis de la dieta ha
mostrado una preferencia por micromamferos, seguido de aves, insectos, anfibios,
conejos y reptiles (Palomares & Delibes 1991). Aunque es muy adaptable en cuanto
al hbitat se ha descrito su uso de formaciones arboladas con cobertura arbustiva
(Palomares & Delibes 1994, Virgs & Casanovas 1997), as como que evita zonas
abiertas a no ser que exista vegetacin de matorral cercana o parches aislados con
rboles y vegetacin de sotobosque que puedan actuar como refugios para la especie
42

Introduccin

(Rosalino & Santos-Reis 2002, Galantinho & Mira 2009). Debido a que presentan
ciertos patrones de seleccin de hbitat definidos en funcin de la zona y cierta
especializacin local en la dieta, la gineta se considera entre el tpico generalista y
el tpico especialista de hbitat y dieta (Virgs, Llorente & Corts 1999). El tamao
medio del rea de campeo en Doana de la especie es de 541 hectreas (Palomares
& Delibes 1994).

El gato monts (3-7 Kg.) es un carnvoro del cual, desde un punto de vista

cientfico, se ha sabido bastante poco hasta aos muy recientes. Aunque puede
vivir en una gran variedad de hbitats, tradicionalmente se ha considerado como
una especie tpicamente forestal (Guggisberg 1975, Ragni 1978, Blanco 1998). No
obstante, en climas ms secos como la cuenca Mediterrnea la especie se encuentra
en paisajes constituidos por mosaicos de matorral y pastizales con abundantes cursos
de agua y presas adems de una alta cobertura de arbustos a escala de microhbitat
(Lozano et al. 2003). Se considera como un especialista facultativo de dieta; siendo
el conejo de monte (Oryctolagus cuniculus) la presa ms abundante en su dieta
cuando est presente, pero consumiendo una alta proporcin de roedores cuando los
conejos son escasos o ausentes (Moleon & Gil Snchez 2003, Lozano et al. 2006,
Sarmento 1996).

Por ltimo, el lince ibrico (9-15 Kg.) es la especie de felino ms amenazada

del planeta (UICN 2008); es endmico de la Pennsula Ibrica (Rodrguez & Delibes
1992, Ferreras et al. 2010) y en la actualidad existe nicamente en dos poblaciones
estables y reproductoras: Doana y Sierra Morena oriental (Ferreras et al. 2010).
El lince ibrico es un especialista trfico, estrictamente dependiente del conejo de
monte (Delibes et al. 2000). La densidad de conejos determina la densidad de linces
(Palomares 2001, Palomares et al. 2001). Adems, asociado entre otras cosas a su
43

Introduccin

sistema de caza y a la distribucin natural del conejo, el lince ibrico tambin es un

especialista de hbitat, estando asociado al matorral mediterrneo (Palomares et al. 1991,
2000, Palomares 2001). La combinacin de zonas de matorral denso y de extensiones
herbceas o maquis abiertos que permitan la caza y el desplazamiento parece esencial
(Delibes et al. 2000). Esta especie constituye un claro ejemplo de contraccin rango
extremo en las ltimas dcadas. Hay evidencias que indican un descenso continuo de la
especie debido al agotamiento severo de su principal presa por enfermedades (mixomatosis
y enfermedad hemorrgica vrica (Villafuerte et al. 1995, 1997) y el exceso de caza
(Rodrguez & Delibes 1992). La reduccin y fragmentacin del matorral mediterrneo
del que tanto el lince como su presa principal dependen, tambin ha causado un impacto
negativo sobre la especie (Rodriguez & Delibes 1992). Hoy en da, las poblaciones de
lince ibrico en declive persisten bajo condiciones de seria presin humana, incluyendo
reforestaciones con especies no nativas y rozas de matorral, persecucin directa de la
especie, expansin de la red vial, construccin de presas y la urbanizacin del medio
natural (Rodriguez & Delibes 2004). El volumen creciente de trfico es tambin el
responsable de las altas tasas de mortandad adicionales no naturales en el lince ibrico;
particularmente en los alrededores del Parque Nacional de Doana (Ferreras et al. 1992).
De manera adicional, recientes estudios han demostrado que el deterioro concomitante
tanto de los atributos demogrficos de la especie como de las caractersticas genticas de
la misma (es decir, tasas de mortalidad no traumtica, tamao de camada, supervivencia
de cras, edad de adquisicin de territorio, proporcin de sexos, tasas de endogamia y
diversidad gentica) es consistente con un vrtice de extincin, y que esa coocurrencia,
con o sin interaccin, del deterioro demogrfico y gentico podra explicar la falta de xito
de los esfuerzos de conservacin de la poblacin de lince ibrico o su lenta recuperacin
en nuestra rea de estudio (Palomares et al. 2012).
44

Introduccin

Figura 4. Especies de estudio en el Parque Nacional de Doana; (a) tejn europeo (Meles meles),
(b) zorro comn (Vulpes vulpes), (c) gato monts (Felis silvestris), (d) lince ibrico (Lynx pardinus),
(e) gineta comn (Genetta genetta) y (d) meloncillo (Herpestes ichneumon).

45

Introduccin

rea de estudio

El nombre de Doana surge para designar el conjunto de cotos de caza

mayor con paisaje de matorral y bosque situado en el entorno de las marismas del
ro Guadalquivir en el suroeste de Espaa, entre las provincias de Huelva y Sevilla
(Figura 5). El inters despertado por la fauna de este enclave conlleva la creacin
de un rea protegida en 1964, as como un centro de investigacin, la Estacin
Biolgica de Doana. En 1969 el rea bajo proteccin aumenta hasta las 35.000
ha con la creacin del Parque Nacional de Doana. En 1982 el rea protegida ve
aumentada de nuevo su superficie con la creacin del Parque Natural de Doana en
el entorno del Parque Nacional (Garca-Novo & Martn-Cabrera 2005).

En la actualidad el Parque Nacional ocupa una extensin de 550 Km2

El rea est designada como Reserva de la Biosfera (UNESCO), Humedal de

Importancia Internacional (Ramsar), Zona de Especial Proteccin para las Aves
Figura 5. Imagen general del rea de estudio Parque Nacional de Doana y su localizacin en el
suroeste de Espaa.

46

Introduccin

(ZEPA - Natura 2000) y Patrimonio de la Humanidad (UNESCO). El clima en

Doana es Mediterrneo sub-hmedo, con inviernos lluviosos y veranos secos. El
paisaje de Doana ha sido transformado por la accin del hombre, con un mayor
alcance de estas transformaciones a lo largo del siglo XX. Como en muchos otros
humedales europeos, desde el siglo XIX se observan intentos de desecacin y
puesta en cultivo de las marismas de Doana. Sin embargo, la transformacin en el
caso de Doana fue particularmente lenta hasta mediados del siglo XX. Aunque las
primeras repoblaciones con pino pionero en Doana datan del siglo XIX, destacan
las llevadas a cabo con esta especie en la dcada de 1950, seguida de repoblaciones
de eucaliptos en la dcada de los 70 (Garca-Novo & Martn-Cabrera 2005). As, el
alcornoque, el enebro y la sabina han sido sustituidos en muchas reas de Doana por el
pino y el eucalipto. El rea ocupada por las marismas de agua dulce y salada tiene una
extensin actual de 30.000 ha (Enggass 1968, Garca-Novo & Martn-Cabrera 2005)
(Figura 6), suponiendo la quinta parte de las marismas naturales que an subsisten en
Espaa (Garca-Novo & Martn-Cabrera 2005). Adems, el Parque Nacional presenta un
sistema de lagunas temporales y permanentes que consta de ms de 3000 cuerpos de agua
en periodos de mxima inundacin (Gmez-Rodrguez 2009). Estas zonas inundadas
constituyen uno de los hbitats de hibernacin y cra ms utilizados por miles de aves
europeas y africanas.
Figura 6. Vista area de la marisma de Doana desde la denominada Torre del Palacio en poca
otoal antes de las primeras lluvias.

47

Introduccin

48

Introduccin

49

Introduccin
Figura 7. Mosaico de ecosistemas del Parque Nacional de Doana. (a) monte blanco, (b) monte
negro, (c) pastizal en La Vera de Doana, (d) laguna temporal inundada en poca invernal, (e)
pastizal en zona inundable desecada durante la poca estival, (f) zonas de repoblacin de pinos
pioneros, (g) eucaliptar, (h) marisma inundada en poca invernal, (i) inicio del tren de dunas
mviles, (j) pinares en corrales entre dunas, (k) dunas mviles (cerro de Los nsares de Doana),
(l) vista de pinares de repoblacin desde el inicio del tren de dunas mviles, (m) vista general de
zona de sabinares y matorral mediterrneo (n) vista general de La Vera de Doana con pastizal y
alcornoques remanentes del bosque primigenio de Doana.

Actualmente Doana est formada por un mosaico de ecosistemas (playa, dunas,

cotos, marisma, matorral mediterrneo) (Figura 7). El monte o los cotos (Figura 8) es
el ecosistema ms estable del Parque. Se localiza en su zona oeste y esta cubierto
de un espeso matorral mediterrneo que ocupa ms del 50% de la superficie del
rea protegida y que segn las especies dominantes, se denomina localmente como
monte blanco (jaras, jaguarzos, tomillos y romeros) o monte negro (brecinas,
brezos, tojos y zarzas) (Figura 5). Este paisaje de monte mediterrneo constituye un
ambiente bastante homogneo a escala de macrohbitat.

El tren de dunas mviles de Doana, que avanzan hasta 6 metros por ao,

est separado por depresiones intermedias con vegetacin de matorral y pino

pionero, que se denominan corrales, y se sita en la zona sur occidental del Parque
Nacional. Tambin es de destacar el ecosistema conocido localmente como La
Figura 8. Vista general de la zona de cotos o matorral mediterrneo en el Parque Nacional de
Doana. El matorral predominante en la imagen se corresponde con el denominado monte blanco
(aulagas, jaguarzos y romeros como especies predominantes) aunque tambin se puede apreciar
algunas manchas dispersas de monte negro (dominadas por brezales) (ver texto).

50

Introduccin

Vera, uno de los biotopos con ms biodiversidad de Doana (Figura 8).

Es un terreno de pastizales que constituye la zona de contacto entre la

marisma y el monte, y recorre el Parque Nacional de norte a sur. La capa fretica

est muy prxima a la superficie, de modo que forma numerosas lagunas temporales
en la estacin hmeda. En la seca, sin embargo, solo aflora en los lugares ms
deprimidos, generalmente ahondadosex profesopara proporcionar agua a la fauna
(zacallones).
Para dar una idea de la enorme riqueza de este espacio, baste mencionar que en
Doana han sido avistadas 400 especies de aves y en lo que respecta a los dems
vertebrados, se encuentran 29 especies de mamferos, 19 de reptiles, 12 de anfibios
y 7 de peces, a las que aadir otras 60 del Estuario del Guadalquivir (GarcaNovo & Martn-Cabrera 2005). Estas cifras, ciertamente altas para Espaa, son
excepcionales para el subcontinente europeo.

REFERENCIAS
Arajo MB, Guisan A (2006) Five (or so) challenges for species distribution modelling. Journal of
Biogeography 33 (10):1677-1688.
Banks SC, Finlayson GR, Lawson SJ, Lindenmayer DB, Paetkau D, Ward SJ, Taylor AC (2005) The
effects of habitat fragmentation due to forestry plantation establishment on the demography
and genetic variation of a marsupial carnivore, Antechinus agilis. Biological Conservation
122 (4):581-597.
Barnosky AD, Hadly EA, Bascompte J et al (2012) Approaching a state shift in Earths
biosphere. Nature 486 (7401):52-58.
Basille M, Calenge C, Marboutin E, Andersen R, Gaillard JM (2008) Assessing habitat selection
using multivariate statistics: Some refinements of the ecological-niche factor analysis. Ecol
Model 211 (1-2):233-240.
Bean K, Jones GP, Caley MJ (2002) Relationships among distribution, abundance and microhabitat
51

Introduccin
specialisation in a guild of coral reef triggerfish (family Balistidae). Mar Ecol-Prog Ser
233:263-272.
Berger KM, Gese EM, Berger J (2008) Indirect effects and traditional trophic cascades: A test
involving wolves, coyotes, and pronghorn. Ecology 89 (3):818-828.
Blanco JC (1998). Mamferos de Espaa. Geoplaneta, Barcelona.
Brashares JS, Prugh LR, Stoner CJ, Epps CW (2010) Ecological and conservation implications
of mesopredator release. In Terborgh J, Estes JA, eds. Trophic Cascades. Island Press.
Forthcoming.
Brooks TM, Mittermeier RA, Mittermeier CG, da Fonseca GAB, Rylands AB, Konstant WR, Flick
P, Pilgrim J, Oldfield S, Magin G, Hilton-Taylor C (2002) Habitat loss and extinction in the
hotspots of biodiversity. Conserv Biol 16 (4):909-923.
Brotons L, Herrando S, Pla M (2007) Updating bird species distribution at large spatial scales:
applications of habitat modelling to data from long-term monitoring programs. Diversity
and Distributions 13 (3):276-288.
Brotons L, Thuiller W, Araujo MB, Hirzel AH (2004) Presence-absence versus presence-only
modelling methods for predicting bird habitat suitability. Ecography 27 (4):437-448.
Burkey TV, Reed DH (2006) The effects of habitat fragmentation on extinction risk: mechanisms
and synthesis. Songklanakarin Journal of Science and Technology 28 (1):9-37.
Calenge C (2005) Des outils statistiques pour lanalyse des semis de points dans lespace cologique.
Universit Claude Bernard Lyon 1.
Calenge C, Basille M (2008) A general framework for the statistical exploration of the ecological
niche. Journal of Theoretical Biology, 252: 674685.
Calenge C, Darmon G, Basille M, Loison A, Jullien JM (2008) The factorial decomposition of the
Mahalanobis distances in habitat selection studies. Ecology 89 (2):555-566.
Calenge C, Dufour AB, Maillard D (2005) K-select analysis: a new method to analyse habitat
selection in radio-tracking studies. Ecol Model 186 (2):143-153.
Cardillo M, Purvis A, Sechrest W, Gittleman JL, Bielby J, Mace GM (2004) Human population
density and extinction risk in the worlds carnivores. PLoS Biology 2 (7):909-914.
Carvalho JC, Gomes P (2001) Food habits and trophic niche overlap of the red fox, european wild
cat and common genet in the Peneda-Geres National Park. Galemys, 13, 39-48.
52

Introduccin
Cavallini P, Lovari S (1991) Environmental-Factors Influencing The Use of Habitat In The Red Fox,
Vulpes-Vulpes. J Zool 223:323-339.
Ceballos G, Ehrlich PR (2002) Mammal population losses and the extinction crisis. Science 296
(5569):904-907.
Cezilly F, Benhamou S (1996) Optimal foraging strategies: A review. Rev Ecol-Terre Vie 51 (1):43-86.
Conover M (2002) Resolving human-wildlife conflicts: the science of wildlife damage management.
Resolving human-wildlife conflicts: the science of wildlife damage management. Lewis
Publishers.
Corsi F, de Leeuw J, Skidmore A (2000) Modeling species distribution with GIS. Research techniques
in animal ecology: controversies and consequences. Columbia University Press.
Cowley MJR, Wilson RJ, Leon-Cortes JL, Gutierrez D, Bulman CR, Thomas CD (2000) Habitatbased statistical models for predicting the spatial distribution of butterflies and day-flying
moths in a fragmented landscape. Journal of Applied Ecology 37:60-72.
Crooks KR, Soule ME (1999) Mesopredator release and avifaunal extinctions in a fragmented
system. Nature 400 (6744):563-566.
Delibes M,Rodrguez A,Ferreras P (2000)Action plan for the conservation of the Iberian lynx in
Europe (Lynx pardinus).Council of Europe Publishing, Strasbourg.
Delibes-Mateos M, de Simon JF, Villafuerte R, Ferreras P (2008) Feeding responses of the red
fox (Vulpes vulpes) to different wild rabbit (Oryctolagus cuniculus) densities: a regional
approach. Eur J Wildl Res 54 (1):71-78.
Donadio E, Buskirk SW (2006) Diet, morphology, and interspecific killing in carnivora. Am Nat
167 (4):524-536.
Duffy JE, Cardinale BJ, France KE, McIntyre PB, Thebault E, Loreau M (2007) The functional role
of biodiversity in ecosystems: incorporating trophic complexity. Ecol Lett 10 (6):522-538.
Durant SM (1998) Competition refuges and coexistence: an example from Serengeti carnivores. J
Anim Ecol 67 (3):370-386.
Eaton RL (1979) Interference Competition among Carnivores - Model For The Evolution of SocialBehavior. Carnivore 2 (MAR):9-16.
Elton G (1927) Animal Ecology. Animal Ecology. London (Sidgwick & Jackson).
Enggass PM (1968) Land reclamation and resettlement in the Guadalquivir Delta-Las Marismas.
53

Introduccin
Economic Geography 44, 125-143.
Entwistle AC (2000) Flagships for future? Oryx 34: 239-240.
Entwistle AC, PJ Stephenson (2000) Small mammals & the conservation agenda. Pages 119-140
in A. Entwistle & N. Dunston, editors. Priorities for the Conservation of Mammalian
Diversity: Has the P&a Had its Day? Cambridge University Press, Cambridge, UK.
Erickson WP, McDonald TL, Gerow KG, Howlin S, Kern JW (2001) Statistical issues in resource selection
studies with radio-marked animals. Radio tracking and animal populations. Academic Press.
Estes JA, Palmisan JF (1974) Sea Otters - Their Role in Structuring Nearshore Communities.
Science 185 (4156):1058-1060.
Fahrig L (2003) Effects of habitat fragmentation on biodiversity. Annu Rev Ecol Evol Syst 34:487-515.
Fedriani JM (1996) Dieta anual del zorro, Vulpes vulpes, en dos hbitats del Parque Nacional de
Doana. Doana Acta Vertebrata, 23, 143-152.
Fedriani JM, Fuller TK, Sauvajot RM, York EC (2000) Competition and intraguild predation among
three sympatric carnivores. Oecologia 125 (2):258-270.
Fedriani JM, Palomares F, Delibes M (1999) Niche relations among three sympatric Mediterranean
carnivores. Oecologia 121 (1):138-148.
Fernandez N, Delibes M, Palomares F (2007) Habitat-related heterogeneity in breeding in a
metapopulation of the Iberian lynx. Ecography 30 (3):431-439.
Ferraras P, Rodriguez A, Palomares F, Delibes M (2010) Iberian lynx: the uncertain future of a
critically endangered cat. Biology and conservation of wild felids. Oxford University Press.
Ferreras P, Aldama JJ, Beltran JF, Delibes M (1992) Rates And Causes of Mortality in a Fragmented Population
of Iberian Lynx Felis-Pardina Temminck, 1824. Biological Conservation 61 (3):197-202.
Fretwell SD, Lucas HL (1969) On territorial behavior and other factors influencing habitat
distribution in birds. 1. Theoretical development. Acta Biotheoretica.
Fuller TK, Keith LB (1981) Woodland Caribou Population-Dynamics in Northeastern Alberta.
Journal of Wildlife Management 45 (1):197-213.
Futuyma DJ, Moreno G (1988) The Evolution of Ecological Specialization. Annu Rev Ecol Syst
19:207-233.
Galantinho A, Mira A (2009) The influence of human, livestock, and ecological features on the
occurrence of genet (Genetta genetta): a case study on Mediterranean farmland. Ecol Res
54

Introduccin
24 (3):671-685.
Garca-novo, F. y C. Martn-Cabrera -2005- Doana: Agua y Biosfera. Doana 2005, Confederacin
hidrogrfica del Guadalquivir, Ministerio de Medio Ambiente. Madrid.
Gause GF (1934) The struggle for existence. The struggle for existence. (Williams & Wilkins).
Gehrt SD, Prange S (2007) Interference competition between coyotes and raccoons: a test of the
mesopredator release hypothesis. Behav Ecol 18 (1):204-214.
Ginsberg JR (2001) Setting Priorities for Carnivore Conservation: What Makes Carnivores Different? In:
Macdonald D, Gittleman J, Wayne R (Eds) Carnivore Conservation. Cambridge University Press.
Gittleman JL (1989) Carnivore group living: comparative trends. Carnivore behavior, ecology, and
evolution. Chapman & Hall, London & Cornell University Press, New York.
Gmez-Rodrguez C (2009) Condicionantes ecolgicos de la distribucin de anfibios en el Parque
Nacional de Doana. Tesis Doctoral, Departamento de Biologa Animal, Ecologa,
Parasitologa, Edafologa y Qumica Agrcola, Universidad de Salamanca, Salamanca, Espaa.
Grinnell J (1917) The Niche-Relationships of the California Thrasher. The Auk 34 (4):427-433.
Guisan A, Zimmermann NE (2000) Predictive habitat distribution models in ecology. Ecol Model
135 (2-3):147-186.
Guggisberg CAW (1975). Wild cats of the world. David & Charles, Newton Abbot, London.
Hall LS, Krausman PR, Morrison ML (1997) The habitat concept and a plea for standard terminology.
Wildl Soc Bull 25 (1):173-182.
Harrison DJ, Bissonette JA, Sherburne JA (1989) Spatial Relationships Between Coyotes and Red
Foxes in Eastern Maine. Journal of Wildlife Management 53 (1):181-185.
Hassell MP, May RM (1973) Stability in Insect Host-Parasite Models. J Anim Ecol 42 (3):693-726.
Hastings A (1980) Disturbance, Coexistence, History, And Competition For Space. Theor Popul
Biol 18 (3):363-373.
Hepinstall JA, Krohn WB, Sader SA (2002) Effects of niche width on the performance and agreement
of avian habitat models. Predicting Species Occurrences: Issues of Accuracy and Scale.
Island Press, Washington.
Hernandez PA, Graham CH, Master LL, Albert DL (2006) The effect of sample size and species
characteristics on performance of different species distribution modeling methods.
Ecography 29 (5):773-785.
55

Introduccin
Heywood, V.H. (1995) Global biodiversity assessment. Cambridge University Press, Cambridge.
Hirzel AH, Hausser J, Chessel D, Perrin N (2002) Ecological-niche factor analysis: How to compute
habitat-suitability maps without absence data? Ecology 83 (7):2027-2036.
Hoek WZ (2008) The Last Glacial-Interglacial Transition. Episodes 31 (2):226-229.
Hoekstra JM, Boucher TM, Ricketts TH, Roberts C (2005) Confronting a biome crisis: global
disparities of habitat loss and protection. Ecol Lett 8 (1):23-29.
Houghton, JT, Y Ding, DJ Griggs, M Noguer, PJ Van der Linden, D Xiaosu, Eds., 2001: Climate
Change 2001: The Scientific Basis: Contributions of Working group I to the Third
Assessment Report of the Intergovernmental Panel on Climate Change.Cambridge
University Press, 881 pp.
Hutchinson GE (1957). Concluding remarks. Cold Spring Harb. Symp. Quant. Biol. 22, 415427.
Inskip C, Zimmermann A (2009) Review Human-felid conflict: a review of patterns and priorities
worldwide. Oryx 43 (1):18-34.
Johnson DH (1980) The Comparison of Usage and Availability Measurements for Evaluating
Resource Preference. Ecology 61 (1):65-71.
Johnson WE, Fuller TK, Franklin WL (1996) Sympatry in canids: a review and assessment. Carnivore
behaviour, ecology and evolution, volume 2. Cornell University Press.
Karanth KU, Chellam R (2009) Carnivore conservation at the crossroads. Oryx 43 (1):1-2.
Kassen R (2002) The experimental evolution of specialists, generalists, and the maintenance of
diversity. J Evol Biol 15 (2):173-190.
Kearney M (2006) Habitat, environment and niche: what are we modelling? Oikos 115 (1):186-191.
Kerley LL, Goodrich JM, Miquelle DG, Smirnov EN, Quigley HB, Hornocker NG (2002) Effects of
roads and human disturbance on Amur tigers. Conserv Biol 16 (1):97-108.
Kolasa J, Li BL (2003) Removing the confounding effect of habitat specialization reveals
the stabilizing contribution of diversity to species variability. Proc R Soc B-Biol Sci
270:S198-S201.
Kolasa J, Pickett STA (1989) Ecological-Systems and the concept of Biological Organization. Proc
Natl Acad Sci USA 86 (22):8837-8841.
Kolasa J, Romanuk TN (2005) Assembly of unequals in the unequal world of a rock pool
metacommunity. Metacommunities: spatial dynamics and ecological communities.
56

Introduccin
University of Chicago Press.
Kruuk H (1975) Functional aspects of social hunting by carnivores. Function and evolution in
behaviour. Essays in honour of Professor Niko Tinbergen, F.R.S. i-xxxi, l-393. Clarendon
Press, Oxford.
Lamprecht J (1981) The Function of Social Hunting in Larger Terrestrial Carnivores. Mammal
Review 11 (4):169-179.
Laurenson MK (1995) Implications of high offspring mortality for cheetah population dynamics.
Serengeti 2: dynamics, management and conservation of an ecosystem. University of
Chicago Press.
Lennon JJ (1999) Resource selection functions: taking space seriously? Trends Ecol Evol 14
(10):399-400.
Lenton TM, Livina VN, Dakos V, Scheffer M (2012) Climate bifurcation during the last deglaciation?
Clim Past 8 (4):1127-1139.
Levin SA (1974) Dispersion and Population Interactions. Am Nat 108 (960):207-228.
Levin SA (1992) The Problem of Pattern and Scale in Ecology. Ecology 73 (6):1943-1967.
Levins R (1968) Evolution in Changing Environments. Princeton University Press, Princeton, NJ.
Lindstrom ER, Brainerd SM, Helldin JO, Overskaug K (1995) Pine Marten Red Fox Interactions - A
Case of Intraguild Predation. Ann Zool Fenn 32 (1):123-130
Litvaitis JA (1992) Niche relations between coyotes and sympatric Carnivora. Pages 7386 in
A. H. Boer, ed. Ecology and management of the eastern coyote.Wildlife Research Unit,
University of Brunswick, Canada.
Lozano J, Moleon M, Virgos E (2006) Biogeographical patterns in the diet of the wildcat,
Felis silvestris Schreber, in Eurasia: factors affecting the trophic diversity. Journal of
Biogeography 33 (6):1076-1085.
Lozano J, Virgs E, Malo AF, Huertas DL, Casanovas JG (2003) Importance of scrubpastureland
mosaics for wild-living cats occurrence in a Mediterranean area: implications for the
conservation of the wildcat (Felis silvestris). Biodiversity & Conservation 12 (5):921-935.
Manly BFJ, McDonald LL, Thomas DL, McDonald TL, Erickson WP (2002) Resource selection by
animals: statistical design and analysis for field studies. Second edition. Resource selection
by animals: statistical design and analysis for field studies. Second edition. Kluwer
57

Introduccin
Academic Publishers.
Marvier M, Kareiva P, Neubert MG (2004) Habitat destruction, fragmentation, and disturbance promote
invasion by habitat generalists in a multispecies metapopulation. Risk Anal 24 (4):869-878.
May RM (1992) How many species inhabit the earth? Sci Amer 10: 1824.
McLoughlin PD, Gaillard JM, Boyce MS, Bonenfant C, Messier F, Duncan P, Delorme D, Van
Moorter B, Said S, Klein F (2007) Lifetime reproductive success and composition of the
home range in a large herbivore. Ecology 88 (12):3192-3201.
Messier F (1991) The Significance of Limiting and Regulating Factors on The Demography of
Moose and White-Tailed Deer. J Anim Ecol 60 (2):377-393.
Miller B, Dugelby B, Foreman D, Martinez del Rio C, Noss R, Phillips M, Reading R, Soule ME,
Terborgh J, Willcox L (2001) The importance of large carnivores to healthy ecosystems.
Endangered Species Update 18 (5):202-210.
Mitchell SC (2005) How useful is the concept of habitat? a critique. Oikos 110 (3):634-638.
Mitchell-Jones AJ, Amori G, Bogdanowicz W et al. (1999)The Atlas of European Mammals.
Academic Press, London, UK.
Moilanen A, Smith AT, Hanski I (1998) Long-term dynamics in a metapopulation of the American
pika. Am Nat 152 (4):530-542.
Moleon M, Gil-Sanchez JM (2003) Food habits of the wildcat (Felis silvestris) in a peculiar habitat:
the Mediterranean high mountain. J Zool 260:17-22.
Morris DW (2003) Toward an ecological synthesis: a case for habitat selection. Oecologia 136 (1):1-13.
Morrison ML, Marcot BG, Mannan RW (2006) Wildlife-habitat relationships: concepts and
applications. Third edition. Wildlife-habitat relationships: concepts and applications. Third
edition. Island Press.
Munday PL, Jones GP, Caley MJ (1997) Habitat specialisation and the distribution and abundance of
coral-dwelling gobies. Mar Ecol-Prog Ser 152 (1-3):227-239.
Naves J, Wiegand T, Revilla E, Delibes M (2003) Endangered species constrained by natural and
human factors: The case of brown bears in northern Spain. Conserv Biol 17 (5):1276-1289.
Ostergard H, Ehrlen J (2005) Among population variation in specialist and generalist seed predation
- the importance of host plant distribution, alternative hosts and environmental variation.
Oikos 111 (1):39-46.
58

Introduccin
Pacala S, Roughgarden J (1984) Control Of Arthropod Abundance By Anolis Lizards On StEustatius (Neth Antilles). Oecologia 64 (2):160-162.
Paine RT (1969) A Note on Trophic Complexity and Community Stability. Am Nat 103 (929):91-&.
Palomares F, Delibes M (1990) Habiatat preference of large grey mongoose, Herpestes ichneumon,
in Spain. Acta Theriologica, 35, 1-6.
Palomares F (1993) Opportunistic Feeding of The Egyptian Mongoose, Herpestes-Ichneumon, (L)
in Southwestern Spain. Rev Ecol-Terre Vie 48 (3):295-304.
Palomares F (1994) Site Fidelity and Effects of Body-Mass on Home-Range Size of Egyptian
Mongooses. Can J Zool-Rev Can Zool 72 (3):465-469.
Palomares F (2001) Vegetation structure and prey abundance requirements of the Iberian lynx:
implications for the design of reserves and corridors. Journal of Applied Ecology 38 (1):9-18.
Palomares F, Caro TM (1999) Interspecific killing among mammalian carnivores. Am Nat 153
(5):492-508.
Palomares F, Delibes M (1991) Dieta del meloncillo, Herpestes ichneumon, en Coto del Rey, Norte
del Parque Nacional de Donana. Dofiana, Acta Vertebrata, 18, 187-194.
Palomares F, Delibes M(1991) Alimentacin del meloncilloHerpestes ichneumony de la
gineta Genetta genettaen la Reserva Biolgica de Doana, SO de la pennsula
Ibrica.Doana Acta Vertebrata, 18 (1): 5-20.
Palomares F, Delibes M (1993) Resting Ecology and Behavior of Egyptian Mongooses (HerpestesIchneumon) in Southwestern Spain. J Zool 230:557-566.
Palomares F, Delibes M (1994) Spatiotemporal Ecology and Behavior of European Genets in
Southwestern Spain. J Mammal 75 (3):714-724.
Palomares F, Delibes M, Revilla E, Calzada J, Fedriani JM (2001) Spatial ecology of Iberian lynx and
abundance of European rabbits in south-western Spain. Wildlife Monographs 148, 136.
Palomares F, Ferreras P, Fedriani JM, Delibes M (1996) Spatial relationships between Iberian lynx and
other carnivores in an area of south-western Spain. Journal of Applied Ecology 33 (1):5-13.
Palomares F, Ferreras P, Travaini A, Delibes M (1998) Co-existence between Iberian lynx and
Egyptian mongooses: estimating interaction strength by structural equation modelling and
testing by an observational study. J Anim Ecol 67 (6):967-978.
Palomares F, Godoy JA, Lopez-Bao JV, Rodriguez A, Roques S, Casas-Marce M, Revilla E, Delibes
59

Introduccin
M (2012) Possible Extinction Vortex for a Population of Iberian Lynx on the Verge of
Extirpation. Conserv Biol 26 (4):689-697.
Palomares F, Rodriguez A, Laffitte R, Delibes M (1991) The Status and Distribution of The Iberian
Lynx Felis-Pardina (Temminck) in Coto Donana Area, Sw Spain. Biological Conservation
57 (2):159-169.
Park T (1962) Beetles, Competition, and Populations - Intricate Ecological Phenomenon is Brought
into Laboratory and Studied as an Experimental Model. Science 138 (3548):1369.
Pendleton GW, Titus K, DeGayner E, Flatten CJ, Lowell RE (1998) Compositional analysis and
GIS for study of habitat selection by goshawks in southeast Alaska. Journal of Agricultural
Biological and Environmental Statistics 3 (3):280-295.
Pimm SL, Rosenzweig ML (1981) Competitors and Habitat Use. Oikos 37 (1):1-6.
Pinero JR (2002) Mamferos Carnvoros Ibricos (Iberian Mammalian Carnivores). Lynx Edicions
Pita R, Mira A, Moreira F, Morgado R, Beja P (2009) Influence of landscape characteristics on carnivore
diversity and abundance in Mediterranean farmland. Agric Ecosyst Environ 132 (1-2):57-65.
Polis GA, Myers CA, Holt RD (1989) The Ecology and Evolution of Intraguild Predation - Potential
Competitors that Eat Each Other. Annu Rev Ecol Syst 20:297-330.
Porter WF, Church KE (1987) Effects of environnemental pattern on habitat preference analysis.
Journal of Wildlife Management 51, 681-685.
Primack RB (2006) Essentials of conservation biology. Fourth edition. Essentials of conservation
biology. Fourth edition. Sinauer Associates.
Pulliam HR (1988) Sources, Sinks, and Population Regulation. Am Nat 132 (5):652-661.
Ragni B (1978) Observations on the ecology and behaviour of the wild cat (Felis silvestris Schreber,
1777) in Italy. Carnivore Genetics Newsletters, 3: 270274.
Ralls K, White PJ (1995) Predation on San-Joaquin Kit Foxes by Larger Canids. J Mammal 76
(3):723-729.
Revilla E, Palomares F (2002) Does local feeding specialization exist in Eurasian badgers? Can J
Zool-Rev Can Zool 80 (1):83-93.
Revilla E, Palomares F (2005) Patrones generales de organizacion social en el tejon eurasitico. In: Virgos E,
Revilla E, mangas JG, Domingo-Roura X (Eds.) Ecologa y conservacin del tejn en ecosistemas
mediterrneos. Sociedad Espanola para la Conservacion y Estudio de los Mamiferos SECEM.
60

Introduccin
Revilla E, Palomares F, Delibes M (2000) Defining key habitats for low density populations of
Eurasian badgers in Mediterranean environments. Biological Conservation 95 (3):269-277.
Ripple WJ, Beschta RL (2006) Linking a cougar decline, trophic cascade, and catastrophic regime
shift in Zion National Park. Biological Conservation 133 (4):397-408.
Riquelme-Cantal JA, Simon-Vallejo MD, Palmqvist R, Cortes-Sanchez M (2008) The oldest
mongoose of Europe. J Archaeol Sci 35 (9):2471-2473.
Rockstrom J, Steffen W, Noone K, et al. (2009) A safe operating space for humanity. Nature 461
(7263):472-475.
Rodriguez A, Delibes M (1992) Current Range and Status of The Iberian Lynx Felis-Pardina
Temminck, 1824 in Spain. Biological Conservation 61 (3):189-196.
Rodriguez A, Delibes M (2003) Population fragmentation and extinction in the Iberian lynx.
Biological Conservation 109 (3):321-331.
Rodriguez A, Delibes M (2004) Patterns and causes of non-natural mortality in the Iberian lynx
during a 40-year period of range contraction. Biological Conservation 118 (2):151-161.
Rodriguez A, Martin R, Delibes M (1996) Space use and activity in a mediterranean population of
badgers Meles meles. Acta Theriol 41 (1):59-72.
Rosalino LM, Santos-Reis M (2002) Feeding habits of the common genet Genetta genetta (Carnivora
: Viverridae) in a semi-natural landscape of central Portugal. Mammalia 66 (2):195-205.
Rosenzweig ML (1981) A Theory of Habitat Selection. Ecology 62 (2):327-335.
Rosenzweig ML (1987) Habitat selection as a source of biological diversity. Evol Ecol 1 (4):315-330.
Sarmento P (1996) Feeding ecology of the European wildcat Felis silvestris in Portugal. Acta Theriol
41 (4):409-414.
Scheffer M, Bascompte J, Brock WA, Brovkin V, Carpenter SR, Dakos V, Held H, van Nes EH, Rietkerk
M, Sugihara G (2009) Early-warning signals for critical transitions. Nature 461 (7260):53-59.
Schipper J, Chanson JS, Chiozza F, et al. (2008) The status of the worlds land and marine mammals:
Diversity, threat, and knowledge. Science 322 (5899):225-230.
Sergio F, Caro T, Brown D, Clucas B, Hunter J, Ketchum J, McHugh K, Hiraldo F (2008) Top
Predators as Conservation Tools: Ecological Rationale, Assumptions, and Efficacy. In:
Annual Review of Ecology Evolution and Systematics, vol 39. Annual Review of Ecology
Evolution and Systematics. Annual Reviews, Palo Alto, pp 1-19.
61

Introduccin
Sinclair ARE, Fryxell JM, Caughley G (2006) Wildlife ecology, conservation, and management. Second
edition. Wildlife ecology, conservation, and management. Second edition. Blackwell Publishing.
Soberon J (2007) Grinnellian and Eltonian niches and geographic distributions of species. Ecol Lett
10 (12):1115-1123.
Soul ME (1987) History of the Society for Conservation Biology: how and why we got here.
Conserv Biol 1 (1):4-5
Soul ME, Bolger DT, Alberts AC, Wright J, Sorice M, Hill S (1988). Reconstructed dynamics
of rapid extinctions of chaparral-requiring birds in urban habitat islands. Conservation
Biology 2: 7591.
Sovada MA, Sargeant AB, Grier JW (1995) Differential Effects of Coyotes and Red Foxes on Duck
Nest Success. Journal Of Wildlife Management 59 (1):1-9.
Steffen W, Grinevald J, Crutzen P, McNeill J (2011) The Anthropocene: conceptual and historical
perspectives. Philos Trans R Soc A-Math Phys Eng Sci 369 (1938):842-867.
Stockwell DRB, Peterson AT (2002) Effects of sample size on accuracy of species distribution
models. Ecol Model 148 (1):1-13.
Terborgh J, Estes JA, Paquet P, Ralls K, Boyd-Heger D, Miller BJ, Noss RF (1999) The role of top
carnivores in regulating terrestrial ecosystems. Continental conservation: scientific foundations
of regional reserve networks. Island Press.
Teyssdre A (2004) Vers une sixime crise dextinctions?, in Barbault R. et Chevassus-au-Louis B.,
(eds.).Biodiversit et changements globaux. Enjeux de socit et dfis pour la recherche,
ADPF, Paris, France.
Thomas DL, Taylor EJ (1990) Study Designs and Tests for Comparing Resource Use and Availability.
Journal Of Wildlife Management 54 (2):322-330.
Thomas DL, Taylor EJ (2006) Study designs and tests for comparing resource use and availability
II. Journal of Wildlife Management 70 (2):324-336.
Thurber JM, Peterson RO, Woolington JD, Vucetich JA (1992) Coyote Coexistence With Wolves on
The Kenai Peninsula, Alaska. Can J Zool-Rev Can Zool 70 (12):2494-2498.
Tilman D, May RM, Lehman CL, Nowak MA (1994) Habitat Destruction and the Extinction Debt.
Nature 371 (6492):65-66.
Tsoar A, Allouche O, Steinitz O, Rotem D, Kadmon R (2007) A comparative evaluation of presence62

Introduccin
only methods for modelling species distribution. Diversity and Distributions 13 (4):397-405.
Turchin P (1998) Quantitative analysis of movement: measuring and modeling population
redistribution in animals and plants. Quantitative analysis of movement: measuring and
modeling population redistribution in animals and plants. Sinauer Associates, Inc.
Vanhorne B (1983) Density as A Misleading Indicator of Habitat Quality. Journal of Wildlife
Management 47 (4):893-901.
Villafuerte R, Calvete C, Blanco JC, Lucientes J (1995) Incidence of viral hemorrhagic disease in
wild rabbit populations in Spain. Mammalia 59 (4):651-659.
Villafuerte R, Lazo A, Moreno S (1997) Influence of food abundance and quality on rabbit
fluctuations: Conservation and management implications in Donana National Park (SW
Spain). Rev Ecol-Terre Vie 52 (4):345-356.
Virgs E, Casanovas JG (1997) Habitat selection of genet Genetta genetta in the mountains of
central Spain. Acta Theriol 42 (2):169-177.
Virgs E, Llorente M, Cortes Y (1999) Geographical variation in genet (Genetta genetta L.) diet: a
literature review. Mammal Review 29 (2):117-126.
Vitousek PM, Mooney HA, Lubchenco J, Melillo JM (1997) Human domination of Earths
ecosystems. Science 277 (5325):494-499.
Whittake.Rh, Levin SA, Root RB (1973) Niche, Habitat, and Ecotope. Am Nat 107 (955):321-338.
Woodroffe R (2001) Strategies for carnivore conservation: lessons from contemporary extinctions.
In: Gittleman JL, Funk SM, Macdonald DW, Wayne RK (eds) Carnivore Conservation, vol
5. Conservation Biology Series. Cambridge Univ Press, Cambridge, pp 61-92.
Woodroffe R, Ginsberg JR (1998) Edge effects and the extinction of populations inside protected
areas. Science 280 (5372):2126-2128.
Yoccoz NG, Nichols JD, Boulinier T (2001) Monitoring of biological diversity in space and time.
Trends Ecol Evol 16 (8):446-453.
Yodzis P (1978) Competition for space and the structure of ecological communities. Springer-Verlag.
Zalasiewicz J, Williams M, Haywood A, Ellis M (2011) The Anthropocene: a new epoch of geological time?
Introduction. Philos Trans R Soc A-Math Phys Eng Sci 369 (1938):835-841.
Zapata SC, Travaini A, Ferreras P, Delibes M (2007) Analysis of trophic structure of two carnivore
assemblages by means of guild identification. Eur J Wildl Res 53 (4):276-286.
63

64

CHAPTER 1

Non-biological factors affecting track

censuses: implications for sampling design
and reliability
Factores no biolgicos afectan los censos de rastros:
implicaciones para el diseo muestral y fiabilidad

Soto C, Desnia S, Palomares F (2012) Non-biological factors affecting track

censuses: implications for sampling design and reliability. European Journal of
Wildlife Research 58:117126
65

66

Chapter 1

ABSTRACT

Track census is a widely used method for rapid faunal assessments, which

assumes that differences in track count numbers mainly reflect differences in

species abundance due to some biological factors. However, some methodological
and climatic variables might affect results of track censuses. Here, we tested the
effect of climatic variables, such as maximum temperature, humidity, wind speed
or days since last rain, and methodological factors, such as censusing day period,
distance from transect to vegetation edge, substrate condition or observer on the
number of tracks of mammal carnivores and some of their potential prey detected in
sandy substrates. We sampled 2 x 2 km squares located within the scrubland area of
Doana National Park (southwestern Spain) for carnivore and several potential prey
tracks. Our results showed differences in the number of tracks detected between
observers and a significant interaction between observers and the day period when
censuses were carried out. Moreover, the variables increasing the quality of the
substrate (higher environmental humidity, lower wind speed and days since last
rain) led to a greater detection of carnivore tracks, but depending on the size of the
species sampled other variables, such as distance from transects to the vegetation
border, also affected results. We recommend restricting sampling to certain fixed
weather conditions when planning to monitor relative animal abundance from
track censuses. When not possible, climatic or methodological variables should be
included as covariates in analyses that try to test for the biological factors affecting
wildlife abundance, taking into account that these variables, which affect the number
of tracks detected could vary between years.

67

Chapter 1

RESUMEN

Los censos de rastros constituyen un mtodo ampliamente utilizado en

estudios de fauna y asumen que las diferencias en el nmero de rastros detectados

reflejan diferencias en la abundancia de las especies debido a factores biolgicos.
Sin embargo, algunas variables metodolgicas y/o climticas pueden afectar
los resultados de dichos censos de rastros. En este estudio analizamos el efecto
potencial de variables climticas tales como la temperatura mxima, la humedad,
la velocidad del viento o los das transcurridos desde la ltima lluvia, as como de
variables metodolgicas como el periodo del da en el que se realiza el censo, la
distancia desde el transecto al borde de la vegetacin, las condiciones del sustrato
o el observador, en el nmero de rastros de diferentes especies de mamferos
carnvoros as como de algunas de sus potenciales presas detectados en censos de
rastros en sustratos arenosos.

Censamos rastros de carnvoros y de diferentes presas potenciales en

cuadrculas de 2 x 2 km2 localizadas dentro del rea de matorral del Parque Nacional
de Doana (situado en el suroeste de Espaa). Nuestros resultados mostraron
diferencias en el nmero de rastros detectados en los censos dependiendo del
observador, as como una interaccin significativa entre el observador y el perodo
del da en el que se realiz el censo. Adems, las variables que incrementaban la
calidad del sustrato (una mayor humedad ambiental y una menor velocidad del
viento, as como un nmero bajo de das transcurridos desde la ltima lluvia),
permitieron una mayor deteccin de rastros de carnvoros. No obstante, dependiendo
del tamao de la especie, otras variables como la distancia del transecto al borde
de la vegetacin tambin afectaron los resultados. Recomendamos restringir los
censos a ciertas condiciones climticas determinadas cuando se planteen realizar
68

Chapter 1

monitoreos de la abundancia relativa de especies basndose en datos obtenidos por

censos de rastros. En los casos en los que no sea posible, las variables climticas
y/o metodolgicas se deben incluir como covariables en los anlisis que tengan
como objetivo analizar los factores biolgicos que determinan la abundancia de las
especies, teniendo en cuenta que esas variables que pueden afectar los resultados de
censos, adems pueden variar entre distintos aos.

69

Chapter 1

INTRODUCTION

Determining occurrence and estimating population abundance of species

is fundamental for their conservation, research and management (Caughley

and Sinclair 1994; Silveira et al. 2003; Sadlier et al. 2004). This is particularly
difficult and poses many practical problems on a large spatial scale and in longterm monitoring for cryptic species with nocturnal and solitary habits, large home
ranges, low-density populations and an elusive nature, such as most mammalian
carnivore species and their prey.

Few methods are suitable for monitoring elusive, low-density species (Mills

et al. 2000) in spite of the amount of available monitoring methods (Williams et

al. 2002; Liebenberg 2010), but indirect sampling methods such as track counts on
suitable natural substrates (e.g. snow, mud or sand) have been traditionally used to
overcome these problems (e.g. Stephenson 1986; Smallwood and Fitzhugh 1995;
Zaumyslova 2000; Gusset and Burgener 2005; Datta 2008; Funston et al. 2010).

The broad application of natural sign surveys such as track counts has firmly

established their use as a tool for wildlife detection. Track surveys do not rely upon
special technology or equipment, can be relatively straightforward and quick to
conduct, and can easily incorporate multispecies and large geographic area objectives.
Moreover, track counts do not require a behavioural response to attractants or other
survey equipment, thus there are potentially fewer species-specific limitations
and biases inherent to track surveys (Long et al. 2008). Nevertheless, field-based
species identification may be ambiguous or unfeasible so additional efforts and
highly skilled and experienced trackers are needed to validate the identification of
species or individuals. This weakness related to species identification combined
with the limited availability of appropriate tracking mediums or conditions, the
70

Chapter 1

ephemeral and weather-dependent character of tracks and the inconsistent survey

designs and quality control procedures, have resulted in a growing criticism of track
surveys and the need to improve survey efforts to meet more rigorous standards.

Although strong relationships between kilometric abundance indices (KAI),

obtained by spotlight counts, and population size have been previously reported
(e.g. Newsome et al. 1989; Short et al. 1997; Garel 2010), it is important to highlight
that track censuses can only be taken as indices of presence, relative abundance
or density estimators (Anderson 2001) and that such indices are rated closely to
true animal abundance across habitat types, observers, and other factors (see Gibbs
200). Herein, the number of tracks of certain species encountered on a transect will
depend on biological factors, such as their abundance, food density and distribution,
vegetation structure and intra- or interspecific interference, including humans (e.g.
Odonoghue et al. 1997; Shapira et al. 2008; Bayne et al. 2008; Blaum et al. 2009)
but there are other classes of variables that affect the index (Buckland et al. 1993).
These variables are related to the observer, including the observers training and
experience, eyesight and fatigue level, the environment (i.e. climatic conditions and
local habitats) and aspects of the species itself, such as their body size (Anderson
2001; Mackenzie and Kendall 2002). Among the variables associated with the
environment, such as wind speed, temperature, humidity, cloud cover, time of
sunrise or days from the last rain or snow, many have been previously suggested as
potential influences on the results of track sampling (Norton 1990; Hayward et al.
2002; Long et al. 2008). These non-biological factors constitute an important source
of error, as they affect the probability of detection and therefore the count. If they are
not considered when designing a monitoring program they can increase variance or
uncertainty for the estimates of relative abundance indices (Thompson et al. 1989).
71

Chapter 1

Despite the potential influence that the above-mentioned factors may have

on track counts, specific studies on the subject are scarce (Jennelle et al. 2002;
Karanth et al. 2003, but see Stander 1998; Balme et al. 2009; Zielinski and Schlexer
2009). Nevertheless, there is a growing suggestion to include measures of precision
and estimates of the detection probability when using indices values (usually raw
counts) purporting to measure relative abundance (e.g. Anderson 2001; Rosenstock
et al. 2002; Engeman 2003).

Here, we studied how methodological and climatic factors affected the number

of tracks detected in surveys carried out on sand-based substrates. Specifically, we

examined whether some climatic variables such as maximum temperature, average
relative humidity, maximum wind speed or days since last rain and methodological
factors such as censusing day period, distance from transect to vegetation edge,
substrate condition or observer may affect the number of tracks detected for a set of
seven mammalian carnivores and some of their potential prey on sandy substrates
in South-western Spain and how, depending on the size of the animals surveyed,
different factors could affect the detection probability.

72

Chapter 1

METHODS
Study area

The study was carried out in Doana National Park in southwestern Spain

(379N, 626W). This is a 550 km2 flat sandy area at sea level bordered to the south
and west by the Atlantic Ocean and to the east by the Guadalquivir River mouth.
The climate is Mediterranean subhumid (i.e. characterised by mild wet winters
and hot dry summers), with an average annual rainfall of 500-600 mm. There are
three main environmental units in the park: marshland, dunes and Mediterranean
scrubland (Fig. 1). Track censuses were restricted to the scrubland area, which
is mainly characterised by heterogeneous patches of xerophytic species such as
Halimium sp. and Cistus sp., and hygrophytic ones such as Erica sp., with some
patches of Juniperus phoenicea and Pistacia lentiscus shrubs. Interspersed with the
scrubland there are scattered cork oak trees (Quercus suber) and wild olive trees
(Olea europea), and a few patches of pine Pinus pinea and eucalyptus Eucalyptus
sp. plantations. The Mediterranean scrubland represents approximately half of the
National Park surface area.

Carnivore species in our study area are the red fox (Vulpes vulpes), the

Eurasian badger (Meles meles), the Egyptian mongoose (Herpestes ichneumon), the
common genet (Genetta genetta), the polecat (Mustela putorius) the Iberian lynx
(Lynx pardinus), the European otter (Lutra lutra), wild and domestic cat (Felis sp.),
and domestic dog (Canis familiaris). Polecats and otters were excluded from our
study because of their low abundance.

Potential target prey species for hunting or consumption as carrion by the

carnivore community and sampled in our study were small mammals (i.e. Garden
dormouse (Eliomys quercinus), Southern Water Vole (Arvicola sapidus), bush rat
73

Chapter 1

(Rattus rattus), Long-tailed Field Mouse (Apodemus sylvaticus) and other mice
(Mus spp.), but the most common are A. sylvaticus (Kufner and Moreno 1989)),
European rabbits (Oryctolagus cuniculus), red partridges (Alectoris rufa), domestic
cows (Bos Taurus) and horses (Equus caballus) and wild ungulates such as the
fallow deer (Dama dama), the red deer (Cervus elaphus) and the wild boar (Sus
scrofa).

Track sampling

Under a wider project that aims to study biological and anthropic

factors affecting wild and domestic carnivore abundance and distribution within
Doana National Park, during the wet seasons of 2007-08 and 2008-09 (from

Figure 1. Map of the study area showing Doana National Park in the southwestern
Spain and the 2x2 km2, where carnivore and prey track censuses were carried out
74

Chapter 1

November 2007 to May 2008 and from October 2008 to April 2009) three and two
observers sampled 59 and 57, respectively, 2 x 2 km squares all with at least 40% of
their surface located within the scrubland area of the park (Fig.1). Marshland area
was not sampled as its clay soils make it unsuitable for track censuses. The squares
with 40% area of open dune were excluded for the present study since vegetation
is clearly different from the rest of the scrubland area, which would add an extra
source of variability to data. The three observers of the first year were a fieldworker
with 15 years of experience and two without previous experience, but that were
trained for two months by the experienced fieldworker. During the second year, the
experienced observer and one of the others carried out the surveys.

We sampled for carnivore tracks in each square by slowly walking zigzag

(ca. 1.5 km/h) in at least 3 km-long sandy paths (in car tracks or firebreaks). Once
a continuous track, that crossed side to side across the pathway, was detected, we
georeferenced it using a GPS. We noted location as a grid reference, date, and the
methodological variables, censusing day time (we established three block schedules;
early morning (from 8 a.m. to 12 a.m.), afternoon (from 12 a.m. to 3 p.m.) and
evening (from 5 p.m. to sunset), start time and end time for each census, and the
observer who carried out the census. In order to homogenize the number of tracks
detected per grid and maximize the probability of detection for each carnivore
species, we re-sampled the same path (leaving at least 7 days between samplings) a
second time in a few squares until completing 3 km if during the first sampling there
was not enough available path within the square to achieve this distance. Thus, we
had more censuses than total number of 2x2 km squares. We always carried out
surveys at least 3 days after any rainfall.

75

Chapter 1

We concentrated prey sampling within on month to avoid strong intermonthly

variations in abundance for some species (e.g. see Kufner 1986; Palomares et al.
2001 for small mammals and European rabbits, respectively). Thus, we carried out
the sampling of prey tracks in April 2009 along transects in every 2x2 km square
sampled for carnivore tracks. These transects for prey species were walked as they
were for carnivore tracks, were 25 m in length and approximately 1.7 m wide (i.e.
the area of a four-wheel-drive car) and were located in the middle of the census path
and separated by at least 300 m. We recorded the location as a grid reference, date,
observer, and the following methodological variables: distance from nearest border
of census transect to the closest vegetation border (not recorded for carnivore track
censuses as they were carried out by zigzag walking), pathway where transects
were established (firebreaks or car tracks) and quality of the substrate for detecting
tracks based on the presence of grass (we established two categories; good (when
grassy groundcover was less than 10% in any part of the 25 m transect) and fair
(when grassy groundcover in any part of the 25 m transect was between 11-30%).
We considered unsuitable for prey counts transects in which grassy groundcover
was more than 30% in some part of the transect.

The climatic variables result from an average of the maximum temperature

(calculated as the average of the maximum temperature measured on the census day
and the maximum temperature measured two consecutive days before the census
day), relative humidity, maximum wind speed, and the number of days since last
rain. The data was obtained from a station located inside Doana National Park
(Control RM1 Meteorological Station; Latitude: 37 118, Longitude: 6 33 17
http://icts.ebd.csic.es).

76

Chapter 1

Data analyses

In order to avoid linearity assumptions, we preliminary explored the shape of the

response for each landscape variable before fitting them into the final equations (Austin,
2002). With this aim we fit Generalised Additive Models (GAMs) (Hastie and Tibshirani,
1990) using carnivore Kilometric Abundance Indexes and the number of prey tracks as
response variables and fitting smoothing splines with 3 degrees of freedom to model
every climatic and methodological continuous effect. The smoothed variables were then
turned into suitable parametric terms guided by visual inspection of the partial residual
plots (Crawley, 2005). The postulated models were then fit to the track-census dataset
using general linear models (GLM) with the log link, negative binomial error structure
and linear and non-linear responses to fixed effects in accordance with the GAM results.
We analyzed the effect of methodological variables observer (observer) and censusing
day time (day_time), and climatic variables maximum temperature (max_temp), average
relative humidity (humidity), maximum wind speed on census day (wind_speed), and
days since last rain (last_rain) on the carnivore Kilometric Abundance Index (KAI)).
We also included in the models the interactions between observer and day_time, as the
number of samplings carried out by each observer in each daily time period was different.
Correlations between predictors were always low (r < 0.6) so we fitted full models (i.e.
models including all the methodological and climatic variables). As for some 2x2 km
squares we carried out more than one census, our sampling unit was the census and not
the square. We examined the effect of the above variables on total carnivore abundance
index, small carnivore (from 1-5 kg of body mass) abundance index (including the
common genet, wild and domestic cats and the Egyptian mongoose) and medium-sized
carnivore (>7 kg of body mass) abundance index (including the red fox, the Eurasian
badger, the Iberian lynx and the domestic dog).
77

Chapter 1

We followed a backward regression-model selection procedure excluding

variables contributing least to the model (i.e. variables with P > 0.3) before models
were refitted. Only variables with P 0.05 were interpreted as statistically significant.
Overdispersion was not a problem ( was close to 1 (1.14 - 1.21)) for any of the models
(Zuur at al. 2009). We analyzed data separately for each study year as the number of
observers changed and to maximize the variability between conditions, which could
affect the number of tracks detected on sand substrates.

We also performed general linear models (GLM) with negative binomial errors

and log link function to analyse the effect of observer, distance from border of census
transects to the closest vegetation border (dist_veg.), type of path (place) where prey
censuses were carried out (firebreaks or car tracks), quality of the substrate (quality) and
climatic variables last_rain, max_temp, humidity and wind_speed on track counts data
of prey species. We also included in the models the interactions between observer and
quality and observer and place. Prey data were grouped as total prey, small prey (small
mammals), medium-sized prey (rabbits and partridges) and large prey (cows, horses and
ungulates). Correlation between predictors was low (r < 0.5), so we fitted full models.
Overdispersion was not a problem ( = 1.04 - 1.24). The sample unit to adjust GLM was
the 25 m transects.

To simplify models and their understanding we also followed a backward

regression-model selection procedure excluding variables with P > 0.3, and then refitted
the models.

All statistical analyses were performed using the SAS 9.2 statistical software

(SAS Inst. Inc., Cary, NC), GAM and GLM were fitted using the gam and genmod
procedures, respectively.

78

Chapter 1

RESULTS

A total of 471 km were walked and 8,373 carnivore tracks were found during

surveys, with the red fox, the Eurasian badger and the Egyptian mongoose being
the most recorded species (Table 1). For prey, 5,000 tracks were detected on 11,575
m sampled (Table 2). Prey species more often detected were ungulates (i.e. fallow
deer, red deer, wild boar) and rabbits.

For the first year, there were differences in the number of tracks detected

among observers for all small and medium carnivores, and the number of tracks
decreased when wind speed increased for total carnivores, increased when humidity
was higher for small carnivores, and was highest during the afternoon for mediumsized carnivores (Table 3, Fig. 2). A significant interaction was also detected between
the censusing day period and the observer, with the three observers finding more
medium-sized carnivore tracks in the evening than in the morning or afternoon
(Table 3, Fig. 2d).
Table 1. Carnivore kilometric abundance index (tracks / km; KAI) in the scrubland area of
Doana National Park during the wet seasons of 2007-2008 and 2008-2009.

Species

Positive
censuses

2007-2008
Total
number of
tracks

KAI
(meanSD)

Range

Positive
censuses

2008-2009
Total
KAI
number
(meanSD)
of tracks

Range

Lynx
pardinus

19

65

0.4 0.9

0 - 5.4

14

75

0.4 0.9

0 - 5.0

Meles meles

68

599

3.9 5.1

0 - 20.6

53

666

3.7 3.8

0 - 17.3

60

631

4.0 4.4

0 - 21.1

48

544

2.8 3.3

0 - 12.5

76

3,138

17.7 9.1

2.4 - 44.2

61

2,333

11.9 6.5

1.2 - 32.6

21

160

0.8 2.3

0 - 15.9

20

66

0.4 0.7

0 - 2.8

Felis sp.

25

0.2 0.8

0 - 5.7

17

27

0.1 0.3

0 - 1.3

Canis
familiaris

12

23

0.2 0.7

0 - 5.9

11

21

0.1 0.3

0 - 2.0

Herpestes
ichneumon
Vulpes
vulpes
Genetta
genetta

Number of censuses was 76 and 62 for each study year, respectively.

79

Chapter 1

During the second year, more tracks were detected when humidity increased for total
and medium-sized carnivores, and fewer small carnivore tracks were recorded when
daily maximum temperature and days since last rain increased (Table 3, Fig. 3).

Most of the variables considered affected the total number of prey tracks

detected (Table 4). Thus, the number of tracks for total prey increased when the
daily maximum temperature, humidity and days since last rain increased (Fig.
4c), and decreased when wind speed and distance to vegetation edge was highest.
Furthermore, the number of total tracks was higher when samplings were carried
out in car tracks than in firebreaks (Fig. 4b), one observer detected more tracks
than the other (Fig. 4a), and there was a significant interaction between observer
and place (Fig. 4d). Some interesting exceptions to this general pattern were found
when data were separated by type of prey (Table 4). For small prey, number of
tracks was not affected by wind speed, humidity, maximum daily temperature and
days since last rain, and the interaction between observer and place was not found.
For medium-sized prey the number of tracks found was not affected by wind speed,
daily maximum temperature and days from the last rain, but in this case quality of
the road did affect results, with a higher number of tracks being detected at transects
without grass. Finally, the number of large prey tracks was not affected by daily
maximum temperature, distance to vegetation, type of path, or observer, and there
was no interaction between observer and place.

DISCUSSION

Track censuses can provide a rapid and cost-effective assessment of relative

abundance of a species, but improving their performance and robustness by

understanding the factors that potentially affect them is a key step for the design of
80

Chapter 1

sound monitoring programmes. Indexes derived from track surveys are partially a
function of animal abundance, but are also a function of a long list of methodological
and climatic variables and characteristics of the species being surveyed. Our results
support the hypothesis that non-biological factors can affect the number of animal
tracks detected in surveys, and must be taken into account when planning to study
animal abundance, distribution or the biological factors determining them.

The aim of this study is not to establish a standard protocol to carry out

track censuses in sandy substrates but to identify how different methodological

and climatic variables can affect the number of tracks detected in censuses.
Nevertheless, as a rule, our study shows how variables increasing the quality of the
substrate (e.g. higher environmental humidity, lower wind speed and few days since
last rain) allowed the detection of a greater number of tracks. Previous studies had
suggested that these types of variables could influence results, but no quantitative
data had been provided (but see Hayward et al. 2002; Bayne et al. 2008). For
instance, to determine the presence of mountain lions Felis concolor californica,
through track surveys, Van Dyke et al. (1986) and Smallwood and Fitzhugh (1995)
recommended excluding desert areas or windy conditions. Other authors (Zielinski
and Kucera 1995; Bayne et al. 2005) have made recommendations for snow
tracking to detect the presence of a wide diversity of species such as American
Table 2. Total number of tracks detected and the mean index of abundance (tracks / 25 m) for each
group of potential prey species along 463 25 m transects.
Species group
Small mammals
Rabbits
Partridges
Cows/horses
Ungulates

No of tracks
233
2,132
260
210
2,165

Mean SD
0.5 1.4
4.6 9.4
0.6 1.3
0.4 1.6
4.7 9.4

Range
0 - 13
0 - 115
0 - 11
0 - 18
0 - 33
81

Chapter 1

martens (Martes americana), Canada lynx (Lynx canadensis), wolverines (Gulo

gulo), coyotes (Canis lastrans), red foxes (Vulpes vulpes), bobcats (Lynx rufus),
grey wolves (Canis lupus) or mountain lions (Puma concolor) among others.
Recommendations include avoiding tracking during snowfall and waiting until the
second morning after a snowfall, carrying out tracking early in the morning during
periods of melting and freezing and not tracking on south-facing slopes.

Moreover, in Europe and other arid regions of Australia, India, South

America or South Africa where track surveys occurring in dust, mud or sand are more
broadly employed, many authors have also tried to standardize survey design and
Table 3. Results of GLM analysis to test for the effect of several methodological and climatic
variables on abundance indeces of total, small and medium-sized carnivores in Doana
National Park. Standard errors have been omitted to simplify the table. Variables with P > 0.3
excluded from the models are represented as (-). Least squares means (LS-Means) of the categorical
fixed effects observer and day time are shown. Non-est. means that the model could not calculate the
parameter because of little data. (*P < 0.05; **P < 0.01; ***P < 0.001; ns not significant).
Effects

Intercept
wind_speed
humidity

2007-2008

2008-2009
Small
carnivores

Mediumsized
carnivores

Total
carnivores

Small
carnivores

Medium-sized
carnivores

Total
carnivores

3.7238 ***

0.5065

3.2957***

2.4779**

2.0154

1.4906

-0.1329*

-0.1316

0.1071

-0.1751

-0.0711

0.0269**

0.0194**

0.0261**

-0.0376

-0.0937**

-0.0095

-0.0160

-0.048**

-0.0023

early morning

24.5214

15.1443**

16.5645

15.1596

afternoon

28.3601

22.8087**

20.3801

16.6357

evening
observer

35.5493

27.8652**

29.2495

23.1351

20.7509***

3.4544***

16.1252**

3.5478

20.0273

27.8207***

2.4517***

25.1415**

2.4450

16.1825

38.8229***

11.5468***

28.3257**

ns

***

max_temp
last_rain
day_time

observer*day_time

82

Chapter 1

effort by suspending searches of tracks for 24 hours after precipitation or periods

of high winds ( 24 km/h) and also by using only early morning observations for
analysis (e.g. Stander et al. 1998; Sargeant et al. 2005; Evans et al. 2009; Funston
et al. 2010).

There was some exception to the general rule stated. A positive correlation

was found between number of tracks recorded and days since last rain for large prey
species. Their large size rendered their tracks easily recognizable in poor substrate
conditions.
Table 4. Results of GLMM analysis to test for the effect of several methodological and climatic
variables on abundance indeces of total, small and medium-sized and large prey in Doana
National Park. Standard errors have been omitted to simplify the table. Variables with P > 0.3
excluded from the models are represented as (-). Least squares means (LS-Means) of the categorical
fixed effects place, quality and observer are shown. Non-est. means that the model could not
calculate the parameter because of little data. (*P < 0.05; **P < 0.01; ***P < 0.001).
Effects

Total prey

Small prey

Medium-sized prey

Large prey

Intercept

2.4520***

-2.7311***

0.7317*

2.4978***

wind_speed

-0.0005**

-0.0004

-0.0006**

humidity

0.0001***

0.0001**

0.0001**

max_temp

0.0002**

0.0002

last_rain

0.0439***

0.0622***

distance_veg

-0.0666***

-0.1018*

-0.1234***

2.4818**
2.8317**

-1.0304**
-2.3403**

1.2910*
1.7615*

place
firebreak
car track
quality
without grass
with grass
observer
1
2

2.5409
2.7726

-1.3780
-1.9927

1.8563**
1.1962**

1.5986***
2.3358***

3.1145***
2.1989***

-1.2665**
-2.1042**

2.3213***
0.7312***

observer*place

***

***

Non-est.

Non-est.

Non-est.

Non-est.

observer*quality

83

Chapter 1

Other methodological variables also affected results depending on the size

of the species sampled. A higher number of prey species tracks were detected when
transects ran near a vegetation border or in car tracks. This result could be caused
by two different reasons. Vegetation must exert a protective effect against wind and
maintain a higher level of moisture on the sand, thereby increasing substrate quality
for detecting tracks. Also, medium and small prey may prefer to remain near a
vegetation edge to decrease predation risk (Hughes and Ward 1993, but see Moreno
et al. 1996).

Figure 2. Effects of wind speed on total carnivore tracks per km detected (a), of observer (least
square means and their standard errors are represented) on total carnivore tracks per km (b), of
relative humidity on small carnivore tracks per km (c), and differences in the number of mediumsized carnivore tracks per km detected by observers in each period of the day given as estimated
least square means and their standard errors (d) during the first study year. F values computed as
MSModel / MSError (i.e. Mean SquareModel/Mean SquareError), and their respective p-values are shown.
84

Chapter 1

Nevertheless, it is necessary to note that our results were not consistent

among size groups or between years. This could be due to the fact that many of the
variables that affect detectability, and therefore the count, exhibit time trends (i.e.
vary between years) further confounding the value and interpretation of the index.
This is an important issue as it makes track censuses incomparable across different
climatic and methodological conditions.

Our results also showed differences in the number of tracks detected

amongst observers. Moreover, the differences were more apparent in poor

substrate conditions, when the tracks of nocturnal species would have suffered the
greatest deterioration due to time, sun or wind, as was suggested by the significant
interaction found between observer and the census day time. This result reflects
the fact that some observers are able to detect more tracks at one period of the day
than others, probably depending on their performance. The effect of the observer in
the number of tracks detected could also be partially related to the census speed as
differences among observers in their average speed as a function of their experience
were detected (data not shown). Quality of data had been previously questioned

Figure 3. Effects of maximum temperature on small carnivore tracks per km (a), and of days
since last rain on small carnivore tracks per km detected (b) for the second study year. F
values computed as MSModel / MSError, and their respective p-values are shown.
85

Chapter 1

depending on skill level of observers (Bider 1968; Smallwood and Fitzhugh 1995;
Anderson 2001; Wilson and Delahay 2001; Silveira et al. 2003). For this reason,
suggestions have been recently proposed to decrease differences among observers
or to evaluate observer skills (Sadlier et al. 2004; Evans 2006, 2009; Zielinski and
Schlexer 2009).

Different approaches might be used to reduce the possible influence

of methodological and climatic variables on track counts. One would be to limit

track censuses to some given weather conditions, which would ensure a constant
substrate quality for detecting tracks. Furthermore, to keep constant the time of the

Figure 4. Effects of observer (least square means and their standard errors are represented) on
total prey tracks per 25 m transects (a), of censusing place on total prey tracks per km (least square
means and their standard errors are represented) (b), of days since last rain on number of large prey
tracks (c), and the interaction between observer and type of pathway sampled given as estimated
least square means and their standard errors (d). F values computed as MSModel / MSError, and their
respective p-values are shown.
86

Chapter 1

day, days from the last rain or snow, observers involved in the sampling, or distance
to vegetation borders will also help to diminish variability in the number of tracks
recorded. Some of these suggestions have been approached by maintaining constant
substrate quality through the use of artificial substrates and by erasing and resampling
newly left tracks on transects for a given fixed number of days (Gruber at al. 2008;
Watts et al. 2008; Russell et al. 2009; Zielinski and Schlexer 2009).

Avoiding such potentially confounding effects in data collection should be

a fundamental design concept (Engeman 2003). However, it is not always possible

to consider these meteorological or methodological issues when carrying out largescale samplings or when large volumes of data are needed for population estimation
procedures. Thus, we also propose that these possible meteorological or methodological
variables be recorded and to include them as covariates in any further statistical
analysis aiming to test for biological factors affecting relative animal abundance (also
see Smallwood and Fitzhugh 1995). Survey design and statistical rigor are important,
but we agree with previous authors (e.g. Thompson et al. 1998;Nichols et al. 2000;
Anderson 2001; Yoccuz et al. 2001; Mackenzie and Kendall 2002) that estimating
relative abundance based on indices alone (e.g., raw counts) is naive, and that the
focus of efforts ought to be on estimating detection probabilities as well.

These results could be applied to a variety of research fields, both for testing

validity of pre-existing data and improving the suitability and performance of future
studies based on track surveys. Although the indices derived from track censuses
are only partially a function of animal abundance (Anderson 2001), if the variables
associated with the observer, the environment, and characteristics of the species
being surveyed are controlled, the reliability of the information extracted from these
methods may be improved.
87

Chapter 1

ACKNOWLEDGMENTS

This research was funded by the projects CGL2004-00346/BOS (Spanish Ministry

of Education and Science) and 17/2005 (Spanish Ministry of the Environment; National Parks
Research Programme). Land-Rover Espaa S.A. lent two vehicles for this work. We are very grateful
especially to J.C. Rivilla for their assistance during fieldwork. C. Soto was also supported by a JAEPredoc grant from the CSIC.

REFERENCES
Anderson DR (2001) The need to get the basics right in wildlife studies. Wildl Soc Bull 29:1294-1297.
Balme GA, Hunter LTB, Slotow R (2009) Evaluating Methods for Counting Cryptic Carnivores. J

Wildl Manag 73(3):433-441.

Austin MP (2002) Spatial prediction of species distribution: an interface between ecological theory

and statistical modelling. Ecol. Model. 157, 101118.

Bayne E, Moses R, Boutin S (2005) Evaluation of winter tracking protocols as a method for

monitoring mammals in the Alberta Biodiversity Monitoring Program. Rep Alta

Biodivers Monit Progr, Vegreville, Alberta, Canada.

Bayne EM, Boutin S et al. (2008) Ecological factors influencing the spatial pattern of Canada lynx

relative to its southern range edge in Alberta, Canada. Can J Zool-Rev Can Zool 86(10):

1189-1197.
Bider JR (1968) Animal activity in uncontrolled terrestrial communities as determined by a sand

transect technique. Ecol Monogr 38(3):269-308.

Blaum N, Tietjen B et al. (2009) Impact of Livestock Husbandry on Small- and Medium- Sized

Carnivores in Kalahari Savannah Rangelands. J Wildl Manag 73(1): 60-67.

Buckland ST, Anderson DR, Burnham KP and Laake JL (1993) Distance Sampling: Estimating

Abundance of Biological Populations. Chapman and Hall, London. 446pp

Caughley G, Sinclair ARE (1994) Wildlife ecology and Management. Blackwell Science

Publications, Oxford.

Crawley MJ (2005) Statistics: An Introduction using R. JohnWiley & Sons Ltd., Chichester.
Datta A, Anand MO, Naniwadekar R (2008) Empty forests: Large carnivore and prey

88

abundance in Namdapha National Park, north-east India. Biol Conserv 141(5):1429-1435.

Chapter 1
Engeman RM (2003) More on the need to get the basics right: population indices. Wildl Soc Bull

31: 286-287.

Evans J (2006) Observer error in identifying species using indirect signs: analysis of a river otter

track survey technique. Thesis, Texas A&M University, College Station, USA.

Evans et al. (2009) Determining Observer Reliability in Counts of River Otter Tracks. J Wildl

Manag 73: 426-432.

Funston PJ, Frank L et al. (2010) Substrate and species constraints on the use of track incidences to

estimate African large carnivore abundance. J Zool 281(1):56-65.

Garel M, Bonenfant C, Hamann JL, Klein F, Gaillard JM (2010) Are abundance indices derived

from spotlight counts reliable to monitor red deer populations? Wildl Biol 16(1)

Gruber B, Reineking B et al. (2008) A new method for estimating visitation rates of cryptic animals

via repeated surveys of indirect signs. J Appl Ecol 45(2): 728-735.

Gibbs JP (2000)Monitoring populations. Pp. 213-247 in Research techniques in animal ecology:

controversies and consequences (L. Boitani & T. Fuller, eds.). Columbia University Press,

New York.

Gusset M, Burgener N (2005) Estimating larger carnivore numbers from track counts and

measurements. Afr J Ecol 43(4): 320-324.

Hastie, TJ, Tibshirani RJ (1990) Generalized Additive Models. Chapman & Hall, London.
Hayward GD, Miquelle DG et al. (2002) Monitoring Amur tiger populations: characteristics of track

surveys in snow. Wildl Soc Bull 30(4): 1150-1159.

Hughes JJ,Ward D (1993) Predation risk and distance to cover affect foraging behaviour in Namib

Desert gerbils. Anim Behav 46: 1243-1245.

Jennelle CS, Runge MC MacKenzie DI (2002) The use of photographic rates to estimate densities

of tigers and other cryptic mammals: a comment on misleading conclusions. Anim

Conserv 5:119-120.

Karanth KU, Nichols JD et al. (2003) Science deficiency in conservation practice: the monitoring of

tiger populations in India. Anim Conserv 6:141-146.

Kufner MB (1986) Tamao actividad, densidad relativa y preferencias de hbitat de los pequeos

y medianos mamferos de Doana, como factores condicionantes de su tasa de predacin.

Thesis, Universidad Autnoma, Madrid, Spain.

89

Chapter 1
Kufner MB, Moreno YS (1989) Abundancia y amplitud de los desplazamientos de Apodemus
sylvaticus en cuatro biotopos de Doana que difieren en cobertura vegetal. Doana, Acta
Vertebrata, 16: 179-181.
Liebenberg et al. (2010) Practical Tracking. Stackpole Books.
Long et al. 2008. Noninvasive Survey Methods for Carnivores. Island Press.
Mills LS, Citta J et al. (2000) Estimating animal abundance using non-invasive DNA sampling:

promise and pitfalls. Ecol Appl 10(1):283294.

MacKenzie DI, and WL Kendall(2002) How should detection probability be incorporated into

estimates of relative abundance. Ecology 83: 2387-2393

Moreno S, Villafuerte R, Delibes M (1996) Cover is safe during the day but dangerous at night:

the use of vegetation by European wild rabbits. Can J Zool 74: 1656-1660.

Newsome AE, Parer I, Catling PC (1989) Prolonged prey suppression by carnivores-predator

removal experiments. Oecologia 78:458467.

Nichols JD, Hines JE, Sauer JR, Fallon FW, Fallon JE, and Heglund PJ (2000) A double-observer

approach for estimating detection probability and abundance from counts. Auk 117:393408.

Norton PM (1990) How many leopards? A criticism of Martin and de Meulenaers population

estimates for Africa. S Afr J Sci 86:218220.

Odonoghue M, Boutin S et al. (1997) Numerical responses of coyotes and lynx to the snowshoe hare

cycle. Oikos 80(1): 150-162.

Palomares F, Delibes M et al. (2001) Spatial ecology of Iberian lynx and abundance of European

rabbits in South western Spain. Wildl Monogr 148:1-36.

Rosenstock SS, Anderson DR, Giesen KM, Leukering T, and Carter MF (2002) Landbird counting

techniques: Current practices and an alternative. Auk 119:4653.

Russell JC, Hasler N et al (2009) Automatic track recognition of footprints for identifying cryptic

species. Ecol 90, 20072013.

Sadlier LMJ, Webbon CC et al. (2004) Methods of monitoring red foxes Vulpes vulpes and badgers

Meles meles: are field signs the answer? Mamm Rev 34: 75-98.

Sargeant GA, Sovada MA et al. (2005) Markov chain Monte Carlo estimation of species distributions:

a case study of the swift fox in western Kansas. J Wildl Manag 69:483 497.

Shapira I, Sultan H et al. (2008) Agricultural farming alters predator-prey interactions in nearby
90

Chapter 1

natural habitats. Anim Conserv 11(1): 1-8.

Short J, Turner B, Risbey DA, Carnamah R (1997) Control of feral cats for nature conservation. II.

Population reduction by poisoning. Wildl Res 24:703714.

Silveira L, Jcomo ATA, Diniz-Filho JAF (2003) Camera trap, line transect census and track

surveys: a comparative evaluation. Biol Conserv 114: 351-355.

Smallwood KS, Fitzhugh EL (1995) A track count for estimating mountain lion Felis concolor
californica population trend. Biol Conserv 71:251259.
Stander PE (1998) Spoor counts as indices of large carnivore populations: the relationship between

spoor frequency, sampling effort and true density. J Appl Ecol 35:378385.

Stephenson RO (1986) Development of lynx population estimation techniqucs. Alaska Department

of Fish and Game, Project W-22-2, W-22-3, W-22-4, and W-22-5, Res. Final Report, 84 pp.

Thompson ID, Davidson IJ et al. (1989) Use of track transects to measure the relative occurrence of

some boreal mammals in uncut forest and regenerations stands. Can J Zool 67:1816-1823.

Thompson WL, White GC, and Gowan C (1998) Monitoring vertebrate populations. Academic

Press, San Diego, California, USA.

Van Dyke FG, Brocke RH, Shaw HG (1986) Use of road track counts as indexes of mountain lion

presence. J Wildl Manag 50 (1):102-109.

Watts CH, Thornburrow D et al. (2008) Tracking tunnels: a novel method for detecting a threatened

New Zealand giant weta (Orthoptera: Anostostomatidae). N Z J Ecol 31:9297.

Williams BK, Nichols JD, Conroy MJ (2002) Analysis and management of animal populations:

modeling, estimation, and decision making. Academic Press, San Diego, USA: 11040.

Wilson GJ, Delahay RJ (2001) A review of methods to estimate the abundance of terrestrial

carnivores using field signs and observation. Wildl Res 28(2): 151-164.

Yoccuz NG, Nichols JD, and Boulinier T (2001) Monitoring of biological diversity in space and

time. Trends in Ecol. and Evol. 16:446453.

Zaumyslova O.Yu. (2000) Long term population-dynamics of ungulates in the Sikhote-Alin Reserve

from winter transect count data. In Analysis of long-term monitoring data of nature

components in Nature Reserves in the Russian Far East: 7079. Astafiev, A.A. (Ed.).

Vladivostok: Dalnauka, USAID.

Zielinski WJ, Kucera TE, Barrett RH (1995) American marten, fisher, lynx, and wolverine: survey
91

Chapter 1

methods for their detection. USDA forest service, Pacific Southwest Research Station

General Technical Report PSW-GTR-157, Albany, CA.

Zielinski WJ, Schlexer FV (2009) Inter-Observer Variation in Identifying Mammals from Their

92

Tracks at Enclosed Track Plate Stations. Northwest Science 83(4): 299-307.

CHAPTER 2

Fine-scale habitat use and niche separation

in a guild of sympatric carnivore species that
differ in life-history traits
Uso del hbitat a escala fina y segregacin de nicho en un
gremio de carnvoros simptridos que difieren en sus rasgos
de historia de vida

93

94

Chapter 2

ABSTRACT

Heterogeneous landscapes offer many choices in habitat selection but in

landscapes where a general habitat type dominates, the resources distribution is

limited and opportunities for coexistence are likely to be the most restrictive. Subtle
patterns of habitat partitioning as well as coexistence between species with different
degree of specialisation may promote ecological separation and therefore sympatry
under this scenario. We tested the hypothesis of fine-scale spatial partitioning as a
function of the species degree of specialisation by comparing habitat use of five
sympatric carnivore species (Iberian lynx, Eurasian badger, Egyptian mongoose,
common genet and red fox) within a Mediterranean protected area dominated by
Mediterranean scrubland in south-western Spain. We first developed logistic and
regression generalized linear mixed models to analyse habitat use and to assess the
environmental features that best describe the relationship between each species and
their environment and secondly we used OMI analysis to illustrate the ecological
space filled by each species in a multidimensional space. Our results suggested that
sympatry of carnivore species in DNP appears to be mediated by fine-scale habitat
selection only for the most specialist species (lynxes and genets). Lynxes and
genets showed the narrowest and most marginal niches as well as niche segregation.
Lynxes were associated with zones of high density of bushes and ecotones between
bushes and pastureland whereas genets showed association to areas with high small
mammal abundance and pine forest with undergrowth bushes. Mongooses were
positively associated with dense bush cover whereas foxes and badgers showed a
non-specific habitat use pattern and a wide distribution across the environmental
conditions sampled in the study area. Foxes, badgers and mongooses overlapped in
niche position, but foxes and badgers had broader niches than mongooses suggesting
95

Chapter 2

that mongooses exhibit certain degree of habitat specialization.

Differences in fine-scale habitat use and ecological separation through

segregation along the ecological niche dimensions have been shown as key
mechanisms for the most specialist species (genets and lynxes) allowing coexistence
whereas for the most generalist species such as mongooses, foxes and badgers,
another mechanisms such as spatial-temporal segregation of activity patterns might
help to explain coexistence among them.

96

Chapter 2

RESUMEN

Los paisajes heterogneos ofrecen diversas oportunidades a las especies

para la seleccin de hbitat, sin embargo, en paisajes dominados por un tipo de

hbitat principal, la distribucin de los recursos es limitada y las oportunidades para
la coexistencia entre especies probablemente ms restrictivas. Patrones sutiles de
particin de hbitat as como la coexistencia entre especies que difieren en su grado
de especialismo de hbitat y/o trfico deben fomentar la simpatra y segregacin de
nicho bajo este escenario. Comprobamos la hiptesis de la particin espacial a escala
fina en funcin del grado de especialismo de la especie en cuestin comparando
los patrones de uso del hbitat de cinco especies de carnvoros simptridos (lince
ibrico, tejn, zorro, gineta y meloncillo) en un rea Mediterrnea protegida en la
que la zona de estudio estuvo dominada por un tipo de hbitat principal; matorral
Mediterrneo, en el suroeste de Espaa. En primer lugar ajustamos modelos lineares
generalizados mixtos logsticos y de regresin para analizar el patrn de seleccin
de hbitat y determinar las caractersticas ambientales que mejor describen las
relaciones de cada especie con su ambiente. Posteriormente realizamos un anlisis
OMI para ilustrar el espacio ecolgico ocupado por cada especie en un espacio
multimensional.

Nuestros resultados sugieren que la simpatra entre especies de carnvoros

en el Parque Nacional de Doana parece estar mediada por seleccin de hbitat

a escala fina solo para las especies ms especialistas (linces y ginetas). Linces y
ginetas mostraron los nichos ecolgicos ms estrechos y marginales as como una
segregacin de nicho entre ambas especies. Los linces mostraron asociacin con
zonas con una densidad elevada de matorral alto y ecotonos entre matorral alto y
pastizales mientras que las ginetas aparecieron asociadas con reas con una elevada
97

Chapter 2

disponibilidad de micromamferos y pinares con sotobosque denso. Los meloncillos

estuvieron asociados a zonas con una cobertura densa de matorral alto mientras que
los zorros y tejones no mostraron un patrn especfico de seleccin de hbitat a
escala fina. Zorros, tejones y meloncillos solaparon en la posicin espacial de sus
nichos ecolgicos, pero los zorros y tejones mostraron nichos ms amplios que el
de los meloncillos, sugiriendo probablemente un mayor especialismo de hbitat de
estos ltimos. Las diferencias en el uso del hbitat a escala fina y la segregacin
de nicho resultaron ser mecanismos clave de coexistencia para las especies ms
especialistas (linces y ginetas), mientras que para las especies ms generalistas como
los meloncillo, zorros y tejones, otros mecanismos como la segregacin espaciotemporal en sus patrones de actividad deben ayudar a explicar los mecanismos de
coexistencia entre estas especies.

98

Chapter 2

INTRODUCTION

The competitive exclusion principle or Gauses law (Gause 1934, Hardin

1960) states that two ecologically similar species cannot coexist unless they differ
sufficiently in niche separation, that is, in the way they use resources. Thus, some
degree of partitioning has to occur in the realized niche of coexisting species which
can occur at the temporal, trophic and/or habitat selection level. This is particularly
obvious in heterogeneous landscapes where spatial heterogeneity foments
coexistence between similar species (i.e., within the same trophic level), selection
of different habitats being one of the main processes that promotes sympatry (Levin
1974, Rosenzweig 1984). Nevertheless, mechanisms that facilitate coexistence
between similar species in apparently homogeneous landscapes where a general
habitat type dominates have been poorly explored.

Coexistence may be dependent on the availability of patches that support a set

of resources sufficient to fulfil species basic needs. Hence, accurate discrimination

and quality assessment of suitable habitats may require more detailed landscape
information to detect crucial habitat features not obvious at broader scales
(Fernndez et al. 2003, Martin et al. 2010).

Additionally, life-history trait differences between coexisting species in

apparently homogeneous landscapes (i.e., where the spatial variation in the species
biotic or abiotic environment is low) may help to understand the way species
exploit resources. Hence, besides fine-scale habitat use segregation, coexistence is
therefore possible if species exhibit differences in their life history traits that allow
niche differences in space (Brown and Wilson 1956, Hutchinson 1959, Chesson
2000). In other words, cohabitation may be privileged when species with different
degrees of habitat and/or trophic specialisation are present.
99

Chapter 2

To test the hypothesis of fine-scale spatial partitioning as a function of

the species degree of specialisation we compared habitat use of five sympatric

carnivore species (Iberian lynx Lynx pardinus, Eurasian badger Meles meles,
Egyptian mongoose Herpestes ichneumon, common genet Genetta genetta and red
fox Vulpes vulpes) within a Mediterranean protected area with a main habitat type
in south-western Spain. We hypothesized that in an area with a main habitat type,
fine-scale habitat use differences may allow sympatry of species within the same
thropic level but that differ in the degree of specialisation. If we graph the ecological
space filled by each species within an n-dimensional hypervolume representing
environmental variability, spatial segregation in their realized niches of the species
as a function of their degrees of specialisation may be revealed. We first developed
habitat models at fine-spatial scales to assess the environmental features that best
describe the relationship between each species and its environment. Second, we
used an ordination method to illustrate the ecological space filled by each species in
a multidimensional space. We fitted a set of fine-scale habitat models considering
prior knowledge of species ecology to test five hypotheses: (a) vegetation and
landscape structure are the best explanatory factors of species habitat use; (b) since
our study area is quite homogeneous in vegetation, we hypothesized that prey
availability is the single explanatory factor of habitat use; (c) human disturbance
is the most important variable describing habitat selection; (d) vegetation and
landscape structure and prey availability interact to explain habitat use; and (e) all
landscape descriptors, prey availability and human disturbance are relevant.

The study of the carnivore guild considered here is particularly interesting

because it will allow to compare different species within a continuum of

specialisation ranging from complete specialisation (i.e., the Iberian lynx) to full
100

Chapter 2

generalisation (i.e., the red fox). The Eurasian badger and the red fox are habitat and
thropic generalist species (Carvalho and Gomes 2001, Cavallini and Lovari 1991,
Kruuk and Parish 1985, Roper and Lups 1995, Balestrieri et al. 2004, Rosalino et
al. 2005, Amores 1975). The Egyptian mongoose is a habitat specialist but trophic
generalist (Palomares and Delibes 1991, Palomares and Delibes 1993, Zapata et al.
2007,) whereas the Iberian lynx is a highly habitat and trophic specialist species
(Palomares et al. 1991, Delibes et al. 2000, Palomares et al. 2001). The genet
meanwhile is considered between typical generalists and typical specialists (Virgs
et al. 1999).

We predicted that habitat use in lynxes and genets should be explained by

multiple variables due to their higher degree of habitat and trophic specialisation
whereas badgers and foxes should have fewer requirements and respond to single
explanatory factors such as prey availability or vegetation and landscape structure
because of their omnivorous diets and habitat generalism. For mongooses, as
habitat specialists but trophic generalists vegetation and landscape structure should
explain their pattern of habitat use. Therefore, genets and lynxes should show the
most marginal (i.e., located in extreme values of variable gradients) and narrowest
realized niches whereas foxes and badgers would have the most widespread and
widest niches among those possible. Mongooses meanwhile may show a nonmarginal niche, wider than lynxes and genets but narrower than foxes and badgers.

101

Chapter 2

Figure 1. The study area. Doana National Park and its location in south-western Spain.

102

Chapter 2

METHODS
Study area

This study was located in Doana National Pak (DNP), a fully protected

area in southwestern Spain (550 km2 379N, 626W) (Fig.1). DNP is a flat sandy
area at sea level bordered to the south and west by the Atlantic Ocean and to the
east by the Guadalquivir River mouth. The climate is Mediterranean subhumid (i.e.,
characterized by mild wet winters and hot dry summers), with an average annual
rainfall of approximately 550 mm. Approximately half of the surface of DNP is
covered by Mediterranean scrubland, and the other half by marshland. There is
a dune system in the southern part of the scrubland area. Track censuses were
restricted to the scrubland biotope.

Mediterranean scrubland is dominated by hygrophytic species consisting

of very dense clumps of heathers Erica sp. up to 3 m high (Erico scoparidae

Ullicetum australis and Erico ciliaris Ullicetum (minaris) lusitanici associations)
and xerophytic species of up to 1.5 m high mainly consisting of Halimium sp, and
other such as Cistus sp., gorses Ulex sp. and rosemary Rosmarinus officinalis,
(Halimio halimifolii Stauracanthetum genistoidis association). More mature
shrubland areas with Pistacia lentiscus and Myrtus communis can be found mainly
in the north and in the valleys of the dune system. Interpested between purely
scrubland areas there are a few and small patches of Eucalyptus camaldulensis and
pine, Pinis pinea plantations. Most of the dune valleys are also colonized by pines
Pinus pinea.

Human access into the park is regulated, but some low impact traditional

uses are maintained under control including cattle-raising, apiculture, and fishing
with traditional methods. The northern and western edges of the protected area are
103

Chapter 2

in close contact with human settlements, crop fields and a highly intensive use paved
road. There are two villages close to the edge of the park limits. Populations in the
main suburban settlement (situated in the western vicinity)vary greatlybetween
winterand summer, as this area ismainlya summer resort occupied by about
250.000 people during the summer seasons. The other village situated in the north
is occupied by about 1,635year-round residents, although on holydays the numbers
of people increase considerably and duringa spring pilgrimage may even reach up
to one million people. There are also private large and medium-sized farms used for
agriculture as well as six visitor centers, hiking and cycling paths, recreation zones
and bird observatories in the nearby area.

Sampling protocol

The study area was divided into 69 2 x 2 km quadrants following UTM

coordinates. Quadrants were sampled throughout the wet seasons, from November
2007 to May 2008 and from October 2008 to April 2009. In each quadrant, a 3 kmlength survey route was slowly walked searching for carnivore and prey tracks. The
survey routes were located along firebreaks and car roads between 2 and 12 m wide.
Once a continuous track that crossed from one side to the other across the pathway
was detected, we georeferenced it using a global positioning system. We resampled
21 squares over the two sampling periods a second time (leaving at least 7 days
between samplings) until completing 3 km, if during the first sampling there were
insufficient available paths within the square to achieve this distance. Thus, as we
had more censuses than quadrants we averaged track counts to get a single value per
quadrant. We always carried out surveys at least 3 days after any rainfall.

104

Potential target prey species of the carnivores studied were also sampled

Chapter 2

by track censuses. Preys sampled were small mammals (mostly long-tailed field
mouse (Apodemus sylvaticus) according to Kufner and Moreno 1989), European
rabbits (Oryctolagus cuniculus), red partridges (Alectoris rufa), domestic cows
(Bos Taurus) and horses (Equus caballus) and wild ungulates such as the fallow
deer (Dama dama), the red deer (Cervus elaphus) and the wild boar (Sus scrofa).
Wild and domestic ungulates are rarely prey of any of the carnivore species studied
here, but we sampled them since they may provide on occasions an important food
source as carrion for some of them. For the first study year, the prey censuses were
carried out at each quadrant at the same time as the carnivore censuses. For the
second study year, we concentrated prey sampling in a one-month period to avoid
particularly apparent inter-monthly variations in abundance for some species (i.e.,
small mammals and European rabbits) (Kufner 1986, Palomares et al. 2001). Thus,
we carried out the sampling of prey tracks in April (corresponding to the intraannual abundance peak in both species) along transects in every quadrant sampled
for carnivore tracks. As Kilometric Abundance Index (KAI) of prey calculated
for each quadrant was obtained from censuses carried out in different months,
we applied a correction in order to homogenize prey indexes for the two study
years relativizing the first years Kilometric Abundance Indexes to April using
the abundance trend curve of the species that likely exhibit higher inter-monthly
variation in their relative densities (rabbits and small mammals) throughout the year
(Moreno et al. 2007, Moreno and Kufner 1988, Villafuerte et al. 1993, Villafuerte
and Moreno 1997). These transects for prey species were located in the middle of
the same routes walked for carnivore tracks, were 25 m in length and approximately
1.7 m wide (i.e., the area of a four-wheel-drive car) and separated by at least 300 m..
Thus, between 7 and 10 prey censuses were carried out per quadrant.
105

Chapter 2

Vegetation characteristics of quadrants were also sampled in the same places

where we censused prey species. There, we visually estimated the cover of short
shrubs (species such as Halimium sp. and Cistus sp.), tall shrubs (species such as
Erica sp., Juniperus phoenica and Pistacia lentiscus) and trees in a circle of 25 m
diameter around the sampling point. In addition we also measured other variables
related to habitat structure: average tree height and average tall- and short-shrub
height.

Predictive variables

Fourteen variables were selected to study habitat selection patterns for each

species (Table 1; see next section for arguments on variable selection).

For each quadrant, we averaged the value obtained at the sampling points both for
prey and vegetation indexes. We then used an index (Kilometre Abundance Index
of tracks (KAI) and percentage of coverage) to categorize different prey abundance
and vegetation type coverage in each quadrant.

The distance to water, distance to La Vera and distance to the anthropic edge

(see Table 1) were calculated from digitized roads, urban settlements, and water
source cover layers in DNP using a Euclidean distance-based approach (Perkin and
Conner 2004, Benson and Chamberlain 2007). Roads included firewalls and car
roads inside DNP. Urban settlements included towns and villages surrounding the
National Park. Water sources included natural and artificial ponds (i.e., dug for the
cattle in zones were the water table is higher) permanently flooded. Traffic index
per quadrant was derived from data on a previous study on the effect of traffic
on biodiversity in DNP (Romn et al. 2010). Ecotones between pastureland and
scrubland were defined using a 1:10 000 fine-scale vegetation map for the years 1996106

Chapter 2
Table 1. Explanatory variables used to model relative abundance of Egyptian mongoose, Eurasian
badger, and red fox and the probability of Iberian lynx and common genet presence.

Variable

Code

Definition

Units

Vegetation

tall shrub

%B

short shrub + tall

shrub

%SB Mean cover of short shrub and bushes per quadrant

trees

%T

Mean cover of trees per quadrant

Landscape

Mean cover of bushes per quadrant

edges between tall

shrub and pastureland

Linear measure of the density of the ecotones between

eBP patches with bush cover >50% and patches with
pasture cover >50%

distance to water

DW

Measured in meters from the quadrant centre to the

nearest permanent flooded natural or artificial pond

distance to La Vera

DV

Measured in meters from the quadrant centre to the

ecotone between the marshland and the Mediterranean
scrubland

Prey availability

m/ha

rabbits

Ra

Kilometric Abundance Index of rabbits per quadrant

Tracks/km
calculated as the mean of censuses per quadrant

small mammals

SM

Kilometric Abundance Index of small mammals

per quadrant calculated as the mean of censuses per Tracks/km
quadrant

Total prey

Tot

Kilometric Abundance Index of rabbits + partridges +

small mammals + ungulates per quadrant calculated Tracks/km
as the mean of censuses per quadrant

distance to antropic
edge

DH

Measured in meters from the quadrant centre to the

nearest protected area border influenced by humans
(iexcluding the beach and marshland edges)

Human disturbance

Traffic index

Mean daily traffic (MDT), adjusted for road length

(m) and type of road (i) in the quadrant (= (MDTi
* mi))

Humidity

Hum Environmental humidity on census day

Observer

Obs Observer who carried out censuses

car/m

%
no units

107

Chapter 2

2006 obtained from the Sistema de Informacin Ambiental de Andaluca for the
Doana area. We reclassified vegetation units or polygons based on four vegetation
attributes of physiognomy, species composition and density of each vegetation layer
within the polygon: (1) trees (P. pinea, Quercus suber and Eucaliptus spp.); (2)
tall shrubs (subsequently referred to as bushes) of mature Mediterranean shrubland
(e.g., P. lentiscus, M. communis) and also tall, thicket Erica spp.; (3) short shrubs
(H. halimifolium, Ulex spp., Stauracanthus genistoides) and (4) pastures. The result
was a reclassified vegetation digital map with 316 polygons and four vegetation
attributes per polygon. The projection for all GIS layers and data was UTM 30S,
datum European 1950 (ED50). Hawths tools and Geopreocessing extension in
ArcInfo 9.3 (ESRI, Redlands, California, USA) was used to calculate distancebased variables, to identify ecotones, and to calculate their density (Table 1).

Statistical analysis

We based our analyses on information-theoretic methods guided by the

view that ecological inference can best be approached by weighing evidence for
multiple working hypotheses simultaneously (Hilborn and Mangel 1997, Burnham
and Anderson 1998, Johnson and Omland 2004). In essence, these methods consist
of identifying a priori the alternative hypotheses for habitat selection and their
mathematical formulation, and then testing their support by fitting the relevant
equations to species distribution data and examining penalized maximum-likelihood
estimates (e.g., Fernndez et al. 2003, Johnson et al. 2004).

We first specified a set of 16 candidate models that could potentially

predict species habitat use and distribution in DNP, therefore restricting the model
selection process to a few meaningful combinations of predictors of the species. For
108

Chapter 2

selecting predictor variables and formulating the candidate models, we considered

five working hypotheses addressing the critical points of the different species lifehistory traits and requirements: (1) vegetation and landscape structure, (2) prey
availability, (3) human disturbance, (4) vegetation and landscape structure + prey
availability and (5) global. We designed fine-scale habitat models considering prior
knowledge of species ecology. The Eurasian badger and the red fox are generalist of
habitat use and diet (Carvalho and Gomes 2001, Cavallini and Lovari 1991, DelibesMateos et al. 2008, Kruuk and Parish 1985, Roper and Lups 1995, Balestrieri et al.
2004, Rosalino et al. 2005, Amores 1975) although a certain degree of local feeding
specialisation has been reported for several badger populations and for different
resources (Kruuk and Parish 1981, Martn et al. 1995) as well as negative responses
to certain types of landscape fragmentation patterns (Virgs 2002). The Egyptian
mongoose is a trophic generalist (Zapata et al. 2007, Palomares and Delibes 1991),
but habitat specialists in Mediterranean areas (Palomares 1985, Palomares and
Delibes 1993). Mongooses actively avoid open areas and select those with dense
vegetation (bushes) for foraging and resting (Palomares and Delibes 1993). The
Iberian lynx is the most specialised in terms of habitat use and trophic niche of this
carnivore guild (Palomares et al. 1991, Ferreras et al. 1997, Delibes et al. 2000,
Palomares et al. 2001). Lynxes feed almost exclusively upon European wild rabbits
(Oryctolagus cuniculus) and need shrub vegetation patches to rest and breed.
Habitats sustaining stable lynx populations should ideally include 40% cover of
understorey vegetation half of which should be bushes (Palomares 2001, Delibes et
al. 2000). Moreover, the density of ecotones between shrubland and pastureland is
also a robust predictor of territory occurrence (Fernndez et al. 2006, Fernndez et al.
2007). Finally, the common genet is considered to be intermediate, falling between
109

Chapter 2

a typical generalist and a typical specialist species (Virgs et al. 1999). Genets need
bushes and hollow trees as sites for nocturnal and diurnal resting and feed mainly on
small mammals such as the long-tailed field mouse (Apodemus sylvaticus) (Delibes
1974, Palomares 1986, Palomares and Delibes 1988, Palomares and Delibes
1991). We also hypothesized that distance to permanent water resources may be an
important factor for all species particularly during the hottest months when many
surface water sources dry out. Additionally, proximity to humans and infrastructure
derived from their activity as well as traffic index inside the protected area may
be detrimental to all the species because they produce higher mortality, degrade
the original Mediterranean ecosystems and involve a higher risk of predation or
competition with non-native carnivores (i.e., domestic dogs).

Highly correlated predictors (r > 0.6) were never included in the same

model. In addition, we fitted an intercept-only equation in order to test improvement

over the null model of no effect. Fitted models were compared using the Akaike
Information Criterion (AICc) and model weights (Burnham and Anderson 2002).
We ranked models by their AICc values and determined the model averaged
parameter estimates (Burham and Anderson 2002). The relative variable importance
of predictor variable j (wj) was determined as the sum of the wi across all models
where j occurs. Larger wj values indicate a higher relative importance of variable j
compared to other variables (Burnham and Anderson 2002).

Candidate model equations were fitted using Generalized Linear Mixed

Models (GLMM) in SAS 9.2 with logit-link and binomial (for lynxes and genets) or
negative binomial (for badgers, mongooses and foxes) error structure (McCullagh
and Nelder 1989). Observer and quadrant were modelled as random effects,
humidity as an additional explanatory variable (Soto et al. 2012), and the distance
110

Chapter 2

covered per quadrant during track censuses as an offset in all the models.

Additionally, although we selected 4 km2 grids for sampling in order to

diminish the possibility of the same individuals being sampled in neighbouring

grids (several of carnivore species sampled may have home ranges of this size;
Palomares 1994, Palomares and Delibes 1994), track counts at neighbouring
grids can be expected to show spatial autocorrelation. Therefore, we checked for
autocorrelation in our data through inspection of semi-variograms and Morans
I correlograms of non-spatial negative binomial and logistic generalised model
residuals. We performed analyses using an autocovariate (AC) method in the
GLIMMIX procedure (SAS 9.2, Littell et al. 1996) when spatial autocorrelation
in non-spatial generalised model residuals was detected (Hosmer and Lemeshow
2000). The autocovariate method accounts for fine-scale spatial variation in the data
by estimating how much the response at each location reflects response values at
surrounding locations (Dormann 2007). This extra parameter is intended to capture
spatial autocorrelation originating from endogenous processes such as movement
of censused individuals between sampling sites (Smith 1994, Keitt et al. 2002,
Yamaguchi et al. 2003). For every quadrant in our study area, we calculated the
autocovariate in ArcGIS 9.3 (ESRI, Redlands, CA, USA) as:

where yj is the number of tracks at quadrant j and wj is the inverse Euclidean distance
between locations i and j. Hence, an autocovariate at location i is defined as a
weighted sum of observation records y at locations j in a neighbourhood determined
by Ni (Miller et al. 2007). The neighbourhood size may be informed by biological
111

Chapter 2

parameters, such as the species dispersal capacity (Knapp et al. 2003) if the cause
of spatial autocorrelation is known (or at least suspected). We hypothesised that
autocorrelation in our data may partially originated from movement of censused
individuals between sampling sites so we set the neighbourhood size to two quadrants
from each quadrant border to capture the average home range for all species. We
incorporated each autocovariate as an additional explanatory variable in the GLMM
models to account for the variation explained by space while maintaining the same
variable selection procedures as for spatially invariant models. Finally, we tested
autocovariate models for autocorrelation in the Pearson residuals, using Morans
I correlograms and semivariograms of the most parsimonious model. We used
variogram procedure in SAS 9.2 to conduct these tests.

To separate the ecological space filled by each species within an

n-dimensional hypervolume representing environmental variability at DNP, we

used an ordination method called the outlying mean index analysis (OMI) (Doldec
et al. 2000). This method measures the marginality of a species habitat distribution
(niche position), i.e., the distance between the mean habitat conditions used by a
species (species centroid) and the mean habitat conditions across the study area
(origin of the niche hyperspace) as well as the species tolerance (niche breadth), i.e.,
the amplitude in the distribution of each species along the sampled environmental
gradients (Hurlbert 1978). The OMI index measures the niche position of each
species. Species with high values of OMI have marginal niches and are assumed
to be influenced by a subset of the measured environmental variables (occur in
atypical habitats in a region), and those with low values have non-marginal niches
and indicate no specific response of a species to the environmental variables (occur
in typical habitats in a region); such species tend to be more common throughout
112

Chapter 2

the study area. The niche breadth or species tolerance is measured by an additional
variance term provided by this method. Low values of species tolerance mean that
a species is distributed across habitats with a limited range of conditions (specialist
species), while high values imply that a species is distributed across habitats with
widely varying environmental conditions (generalist species). Residual tolerance is
the variation in species occurrence not accounted for by the main gradient. Outlying
mean index is robust to unimodal, linear, or a mixture of species response curves and
is not biased against species-poor or low-abundance sites on the synthetic gradient.
Its interpretations are also robust to multicollinearity among the explanatory
variables (Doldec et al. 2000). We determined significance of the outlying mean
index analysis at =0.05 based upon a Monte Carlo simulation (Metropolis and
Ulam 1949), in which observed marginalities were statistically compared to 10,000
random permutation values of species marginalities or the null hypothesis that
species are distributed equivalently in relation to the environmental variables. The
OMI analysis was performed with the ADE4 library (Thioulouse et al. 1997) in the
R Software (R Development Core Team 2005).

RESULTS

We surveyed 471 km and 8,373 carnivore tracks were found, with foxes,

badgers and mongooses being the most frequent species (Fig. 2, Table 2). For prey,
5,000 tracks were detected on 11.6 km sampled. The most common prey species
were wild ungulates and rabbits (Table 2). The variables Ra and Tot were correlated
(r = 0.669, P < 0.001).

Based on the p-value of Morans I (P>0.05) and the semivariograms of

residuals, habitat use models for badgers, mongooses and foxes showed spatial
113

Chapter 2

Figure 2. Maps of scaled abundances (0-1) (tracks/km) per quadrant of the five carnivore
species for both study years in DNP. (a) Genetta genetta; (b) Lynx pardinus; (c) Herpestes
icheumon; (d) Meles meles and (d) Vulpes vulpes.
114

Chapter 2

autocorrelation while it this was not detected for lynxes and genets. We therefore
fitted spatial negative binomial generalised models (i.e., using an autocovariate)
for badgers, mongooses and foxes and non-spatial logistic generalised models for
lynxes and genets.

The best approximating models (AICc < 2) for mongooses, badgers and

foxes belonged to the set of candidates designed with the hypothesis of vegetation
and landscape structure + prey availability as well as with the hypothesis of prey
availability (Table 3). For mongooses, top models included as predictors of their
relative abundance DV, DW, Tot, Ra, %SB, %B and eBP. Variables DW and DV had
the highest weights and were positively associated with the number of mongoose
tracks (Table 5). Variables Tot and %B were also positive and had relatively high
weights (wj > 0.4) (Table 5). For badgers, models included as predictors Ra, SM,
%SB, %B, %T, DW, V and eBP. All of these predictors were positive except for the
distance to La Vera. Ra was the variable with the highest weight (wj = 0.577) (Table
5) and was positively associated with the number of badger tracks. For the red fox
only one model was supported by data (Table 3). This model included as predictors
%B, %T, DW and SM and were all positively associated with the number of fox
tracks. Only two variables showed high weights; %B (wj = 0.809) and DW (wj =
0.871) (Table 5).

The best approximating models for lynxes and genets belonged to the set

of candidates designed with the hypothesis of vegetation and landscape structure,

vegetation and landscape structure + prey availability as well as with the global
hypothesis (Table 4). For genets, top models included as predictors %B, %T, DW,
SM, DH and T. Four variables included in top models (%B, %T, DW and SM) had
high weights (> 0.812; Table 5). %B, %T and SM were positively associated with the
115

116

61

55

29

12

15

14

69

69

49

43

21

69

69

Vulpes vulpes

Herpestes ichneumon

Meles meles

Genetta genetta

Felis sp.

Canis familiaris

Lutra lutra

TOTAL

Oryctolagus cuniculus

Small mammals*

Alectoris rufa

Domestic ungulates**

Wild ungulates***

TOTAL

65

65

27

44

52

65

65

14

12

21

22

53

50

61

17

2nd

# Study year (1st (2008-2009); 2nd (2009-2010))

No distinction between tracks of Felis silvestris and Felis catus.

***Dama dama, Cervus elaphus

**Bos Taurus, Equus caballus

73 (1.7)

2nd

4312

37 (0.9)

29 (0.7)

35 (0.8)

83 (1.9)

665 (15.4)

617 (14.3)

3.25.8

0.20.7

0.20.8

0.10.5

1.22.5

2.63.7

3.94.6

17.29.1

0.30.9

1st

Average

210 (4.2)

260 (5.2)

233 (4.7)

2.5.2

12.521.6

6.7.4

0.3 0.9

2nd

Average

0.0-5.0

2nd

Range

0-0.09

0-0.3

0-0.2

0-1.1

0.0-42.1

0.0-3.8

0.0-5.9

0.0-3.9

0.0-15.9

0.0-19.5

0.0-21.2

3.2.3

4.0.7

3.6.0

32.844.6

2.64.45

0.20.4

0.20.5

0.20.3

0.40.8

3.33.4

3.13.7

0-0.1

0-0.07

0-0.1

0.007-1

0.0-77.4

0.0-2.2

0.0-3.4

0.0-1.4

0.0-3.1

0.0-17.3

0.0-17.2

2.4-42.1 13.110.2 1.6-77.4

0.0-5.4

1st

Range

KAI

9132

5000

26.927.9

0-1.1

15.417.0

0-1

3720 (40.7) 2165 (43.3) 54.742.6 0.02-0.6 33.323.2 0.020.4

168 (1.8)

848 (9.3)

454 (5.0)

3942 (43.2) 2132 (42.6) 58.059.5

6639

49 (0.7)

38 (0.6)

48 (0.7)

280 (4.2)

786 (11.8)

996 (15.0)

4368 (65.9) 2773 (64.3)

74 (1.1)

1st

N tracks (%)

*Eliomys quercinus, Arvicola sapidus, Rattus rattus, Apodemus sylvaticus, Mus spp.

17

69

Lynx pardinus

#1st

Specie

Positive quadrants

Table 2. Number of positive 2x2 km quadrants (for 69 and 65 sampled in each year), total tracks found (N tracks) and percentage regarding total
tracks found (%), and number of tracks per km (KAI) for each carnivore and their potential prey species censused in Doana National Park during
the wet season for every study year.

Chapter 2

Chapter 2

presence of genet tracks while DW was negatively associated. Finally, for lynxes
top models included as predictors %SB, %B, DW, eBP, V and Ra. Variables included
in the best models showed high weights (wj>0.8) (Table 5). %SB, %B, DW, eBP
and Ra were positively associated with the presence of lynx tracks while DV was
negatively associated.

To illustrate niche separation between species, we used the first two axes

of the outlying mean index analysis, which accounted for 99.95% of the total
explained environmental variability (Table 1, supplementary information). The
overall outlying mean index analysis (i.e. sensitivity of carnivores to environmental
variables) was significant (P < 0.0001).

The position of the species along the first two OMI axes is presented in Fig.

3. Genets and lynxes were clearly separated from mongooses, badgers and foxes.

Figure 3. OMI analysis. Ecological position of the five carnivore species in the n-dimensional
hypervolume representing fine-scale environmental variability in DNP. (a) Lynx pardinus,
(b) Genetta genetta, (c) Herpestes ichneumon, (d) Meles meles and (d) Vulpes vulpes.
117

118

875.03 22.06 0.00

5. DW, DV

862.91

8. Tot

9. DH, T

9.94

6.82
0.00

0.02

885.97 33.00 0.00

859.79

7. SM

Human disturbance

858.83

6. Ra
0.03

878.34 25.37 0.00

4. %SB, %B, DW, DV

5.86

884.54 31.57 0.00

3. %B, %T, DW

Prey availability

880.35 27.38 0.00

918.52 65.55 0.00

16

11

12

15

14

17

Herpestes ichneumon
AICc
i
Wi Ranking

2. %SB, %B, DW, eBP, DV

Vegetation and landscape structure

1. Intercept only

Null model

Model

889.94

861.13

861.61

858.97

890.03

888.92

886.03

885.97

910.64

AICc

30.97

2.16

2.64

0.00

31.06

29.95

27.06

27.00

51.67

0.00

0.11

0.09

0.33

0.00

0.00

0.00

0.00

0.00

Meles meles
i
Wi

14

15

13

12

11

17

Ranking

4.99

4.71

4.16

0.03

0.04

0.05

1253.69 36.63 0.00

1222.05

1221.77

1221.22

1252.62 35.56 0.00

1250.28 33.22 0.00

1248.20 31.14 0.00

1252.19 35.13 0.00

16

14

12

11

13

17

Vulpes vulpes
i
Wi Ranking

1277.53 60.47 0.00

AICc

Table 3. Summary of spatial negative binomial predictive models for Egyptian mongoose, Eurasian badger and red fox abundance and model
selection estimators; -2 log(L)=-2 log-likelihood estimates; AICc = second order Akaikes Information Criterion; i = (AICc)i - (AICc)min; Akaike
Wi = Akaike weights. Top models (AICc2) are shown in bold.

Chapter 2

855.37

852.97

12. %SB, %B, DW, DV, Tot

13. DW, DV, Tot

See Table 1 for model definitions

879.28 26.31 0.00

0.02

17. DW, DV, Tot, DH, T

6.59

859.56

16. %SB, %B, DW, DV, Tot, DH, T

0.05

0.48

0.14

0.02

0.25

863.78 10.81 0.00

4.61

0.00

2.40

5.95

1.33

15. %B, %T, DW, SM, DH, T

14. %SB, %B, DW, eBP, DV, Ra, DH, T 857.58

Global models

858.92

11. %B, %T, DW, SM

Vegetation and landscape structure

+ prey availability
10. %SB, %B, DW, eBP, DV, Ra 854.30

13

10

894.15

867.54

864.74

863.31

865.11

863.36

860.64

859.94

35.18

8.57

5.77

4.34

6.14

4.39

1.67

0.97

0.00

0.00

0.02

0.04

0.02

0.04

0.14

0.21

16

10

2.86

4.33

6.01

3.86

2.05

0.00

3.86

0.10

0.05

0.02

0.06

0.15

0.42

0.06

1253.63 36.57 0.00

1219.92

1221.39

1223.07

1220.92

1219.11

1217.06

1220.92

15

10

Chapter 2

119

Chapter 2
Table 4. Summary of non-spatial logistic predictive models for common genet and Iberian lynx
occurrence, and model selection estimators; -2 log(L)=-2 log-likelihood estimates; AICc = second
order Akaikes Information Criterion; i = (AICc)i - (AICc)min; Akaike Wi = Akaike weights. Top
models (AICc2) are shown in bold.
Genetta genetta
Model

AICc

Wi

Lynx pardinus
Rank

AICc

Wi

Rank

Null model
1. Intercept only

175.23 18.59 0.00

17

151.89 44.15

0.00

15

2. %SB, %B, DW, eBP, V

168.97 12.33 0.00

12

107.74

0.00

0.52

3. %B, %T, DW

163.47

0.02

141.70 33.96

0.00

10

4. %SB, %B, DW, V

170.63 13.99 0.00

13

121.26 13.52

0.00

5. DW, V

172.75 16.11 0.00

14

125.16 17.42

0.00

6. Ra

173.89 17.25 0.00

16

151.71 43.97

0.00

14

7. SM

173.62 16.98 0.00

15

152.87 45.13

0.00

16

8. Tot

163.08

6.44

0.02

148.24 40.50

0.00

13

166.58

9.94

0.00

11

155.17 47.43

0.00

17

10. %SB, %B, DW, eBP, V, Ra 164.69

8.05

0.01

109.53

1.79

0.21

11. %B, %T, DW, SM

157.25

0.61

0.34

142.21 34.47

0.00

11

12. %SB, %B, DW, V, Tot

162.11

5.47

0.03

122.23 14.49

0.00

13. DW, V, Tot

162.67

6.03

0.02

110.99

3.25

0.10

162.32

5.68

0.03

110.02

2.28

0.17

156.64

0.00

0.47

146.19 38.45

0.00

12

16. %SB, %B, DW, V, Tot, DH, T

161.11

4.47

0.05

122.93 15.19

0.00

17. DW, V, Tot, DH, T

164.91

8.27

0.01

10

122.72 14.98

0.00

Vegetation and landscape structure

6.83

Prey availability

Human pressure
9. DH, T
Vegetation and landscape structure
+ prey availability

Global models
14. %SB, %B, DW, eBP, V, Ra,
DH, T
15. %B, %T, DW, SM, DH, T

See Table 1 for model definitions

Genets and lynxes had the highest marginality values and the lowest tolerance
to average habitat conditions of the synthetic gradients (Table 2, supplementary
information). Badgers and foxes had the lowest marginality and the highest
120

Chapter 2

tolerance values whereas mongooses exhibited relatively low marginality and high
tolerance. Although the overall response of mongooses was non significant their
marginality and tolerance levels was between that of genets/lynxes and badgers/
foxes as predicted.

Ordination diagrams on the first two axes of the OMI describe the

environmental gradients that best discriminated the occurrences of the five carnivore
species in DNP (Fig. 1, supplementary information). The first OMI axis was most
influenced by prey availability and shrub cover far from human nuclei and the
proportion of tall shrubs and trees at the opposing end of the gradient. The second
OMI axis was most influenced by distance to the main ecotone between marshland
and scrubland (La Vera) and prey availability and percentage of ecotones between
pastureland and scrubland at the opposite end of the gradient (Table 2, supplementary
information). The presence of genet tracks was negatively associated with axis 1
(Fig. 3), and thus was positively related to patches distant from human nuclei and
with high tree and tall shrub coverage. Lynxes were positively associated with axis
1 and negatively with axis 2 and thus positively related to patches with high prey
availability (mainly rabbits) close to La Vera. Conversely, mongooses, badgers and
foxes were more likely to be found across a wide range of habitat types with varying
environmental conditions.

DISCUSSION

Heterogeneous landscapes offer many choices in habitat selection (i.e.

different axes over which species can differ), thus broadening opportunities for
coexistence. However, in quite homogeneous landscapes where a general habitat
type dominates, the distribution of the resources available for species is limited, and
121

122

autocov

HUM

DH

Tot

SM

Ra

DW

eBP

%T

%SB

%B

Variables

0.22

0.07

0.61

0.16

0.01

2.96

0.02

8.08

0.17

0.04

0.08

0.21

0.08

0.24

-0.27

-0.10

0.12

0.17

1.00

1.00

0.64

0.32

0.93

0.96

0.29

0.44

0.48

Herpestes
ichneumon

SE()
wj

9.71

0.02
3.27

0.01

0.01

0.00

0.02

0.01

0.03

0.00

0.19

0.14

0.10

0.26

SE()

0.05

-0.01

0.25

0.42

0.14

0.33

Meles meles

1.00

1.00

0.21

0.25

0.58

0.26

0.46

0.24

0.29

0.45

wj

1.91

0.01

0.00

0.00

0.29

0.35

0.00

0.02

0.00

0.01

0.22

0.25

SE()

wj

1.00

1.00

0.47

0.51

0.87

0.81

Vulpes vulpes

0.02

-0.16

0.01

0.60

-0.12

0.05

0.10

0.02

0.14

0.04

0.18

0.07

0.01

0.06

SE()

wj

1.00

0.56

0.56

0.47

0.98

0.83

0.95

Genetta genetta

0.02

0.02

-2.56

0.48

1.85

1.69

0.32

0.02

0.00

0.70

0.37

0.69

1.2

0.12

SE()

wj

1.00

0.38

1.00

1.00

0.90

0.90

0.90

Lynx pardinus

Table 5. Variable weight (wj), average model coefficient () with average standard error (SE()) for variables included in top models.
Variables are described in Table 1.

Chapter 2

Chapter 2

opportunities for coexistence are likely to be the most restrictive. Hence, subtle
patterns of habitat partitioning as well as coexistence between species with different
degree of specialisation may promote ecological separation and therefore sympatry
under this scenario. A parsimony-based strategy for confronting different model
hypotheses allowed us to assess fine-scale landscape attributes linked to five
carnivore species in a protected area where species coexist in a largely similar
habitat type, the Mediterranean scrubland. These species exhibited spatial storage
or separate niches along a gradient of environmental variability according to their
degrees of specialization.

On the basis on OMI analysis, lynxes and genets are marked specialists and

showed the narrowest and most marginal niches throughout DNPs environmental
conditions (Fig. 3). Additionally, lynxes and genets showed niche segregation
as a result of their differences in ecological preferences (Table 2 supplementary
information), which is supported by the fine-scale habitat use results. Results
of fine-scale habitat use models for lynxes and genets were probably related to
the availability of prime resources for both species such as refuges and prey.
Lynxes and genets showed the most restricted habitat use pattern depending on
vegetation, landscape, prey and human-related variables. Lynxes were associated
with zones of high density of bushes likely representing areas of late-successional
Mediterranean communities, an uncommonm microhabitat in the area due to the
human transformation of the autochthonous vegetation (Garca-Novo and MartnCabrera 2005). The density of ecotones between bushes and pastureland, which in
turn favour abundance of the main prey of lynx, the European rabbit (Fernndez
et al. 2006, Fernndez et al. 2007), was also a robust predictor of the presence of
lynx tracks, as well as the distance to the ecotone band between the Mediterranean
123

Chapter 2

scrubland La Vera, where refuge and grass availability favour a high rabbit density
in Doana (Villafuerte and Moreno 1997, Fernndez and Palomares 2000). As
expected, genets showed association to areas with high small mammal abundance
(a community mostly composed of long-tailed field mouse), important prey items
for the species in Mediterranean areas (Palomares and Delibes 1991), as well as
preference for pine forest areas with undergrowth bushes. These results agree with
previous findings for the radio-tracked population of the Iberian lynx (Palomares et
al. 2000, Palomares 2001, Fernndez et al. 2003) and the common genet (Palomares
and Delibes 1988, 1991, 1994, Palomares et al. 1996) in the protected area. The low
abundance and the spatially structured distribution of both species in DNP (Fig.
2) could therefore be explained by the fact that genets and lynxes may be limited
in their home ranges to areas of simultaneous convergence of high small mammal
or rabbit availability as well as a high percentage of tall shrub cover and overall
understorey or a high percentage of tree cover, respectively.

Foxes and badgers showed a non-specific habitat use pattern and a wide

distribution across the environmental conditions sampled in our study area as

shown by the low weights of predictor in the fine-scale habitat models (Table 5)
and the OMI analysis results (Fig. 3) suggested. Rabbits, as well as bush density
and distance to water sources, were the most important predictors associated with
numbers of badger and red fox tracks, respectively. Badgers are trophic generalists
but in Mediterranean areas rabbits may constitute an important prey for the species
(Revilla and Palomares 2002). Foxes exhibited higher relative abundances in areas
with dense bush cover, a microhabitat that could potentially offer higher protection.

Mongooses were positively associated with dense bush cover and high

availability of different prey items as a whole (i.e., rabbits, small mammals and
124

Chapter 2

partridges) as well as with high availability of rabbits. It could be explained by the

fact that in spite of the trophic generalism of the species (Santos et al. 2007), wild
rabbits constitute an important prey item for the species in Mediterranean areas
(Palomares and Delibes 1991).

Foxes, badgers and mongooses overlapped in niche position, but foxes

and badgers had broader niches than mongooses (Fig. 3, Table 1 supplementary
information). In fact, mongooses exhibited and intermediate niche position and
niche breadth between that of genets/lynxes and that of badgers/foxes (Table 2,
supplementary information). These results suggest that mongooses exhibited certain
degree of habitat specialization.

The wider spatial niches exhibited by mongooses, foxes and badgers may

be a response to fluctuations in resource availability characteristic of Mediterranean

environments. Mediterraneity (Virgs and Casanovas 1999) may result in an
opportunistic behaviour in food gathering and habitat use broadening realized niches
of species and allowing coexistence between species with similar requirements by
switching to other temporarily available resources.

In summary, sympatry of carnivore species in DNP appears to be mediated by

fine-scale habitat selection only for the most specialist species (lynxes and genets).
Badgers and foxes meanwhile did not show any clear pattern of specialization at
a fine scale as expected due to their high levels of habitat and trophic generalism.
Although mongooses were more specialised in habitat selection than badgers and
foxes, for areas such as that studied here the three species can coexist with no
apparent differences in habitat use and/or realized niche segregation. Differences in
activity patterns of these three species might help to explain the coexistence among
them. Mongooses are almost exclusively diurnal in the study area (Palomares and
125

Chapter 2

Delibes 1993) while badgers are exclusively nocturnal (Revilla and Palomares
2002) and foxes nocturnal and crepuscular. Therefore, the carnivore community
structure in DNP seems not to be defined only by habitat use partitioning.
Interspecific interactions among predators are another aspect that should
be explored and can also greatly shape the community structure. Mongooses, for
example, were strongly negatively associated with distance to La Vera, which might
be related with the high use that lynxes made of this area (Palomares et al. 1991,
Viota et al. 2012).

Under a landscape analysis scheme that we have defined as coarse-scaled,

where Mediterranean shrubland can be considered as a homogeneous landscape,

we have been able to detect differences in suitability for carnivores in DNP after
examining fine scale habitat variables. Differences in fine-scale habitat use and
ecological separation through segregation along the ecological niche dimensions
have been shown as key mechanisms for some species allowing coexistence. Hence,
the study of spatial heterogeneity at small scales particularly in largely homogeneous
areas is essential to understanding the community structure of coexisting similar
species. To manage conservation areas that protect coexisting carnivore species,
we must understand species-specific habitat use patterns and evaluate the role that
competitors play in determining these patterns.

ACKNOWLEDGEMENTS

This research was funded by the projects CGL2004-00346/BOS (Spanish Ministry of

Education and Science) and 17/2005 (Spanish Ministry of the Environment; National Parks Research
Program). Land-Rover Espaa lent us two vehicles for this work. We are very grateful especially to J.C.
Rivilla and S. Desnia for assistance during fieldwork and to M. Gonzlez for her valuable suggestions
on earlier versions of the manuscript. C. Soto was also supported by a JAE-Predoc grant from the CSIC.
126

Chapter 2

REFERENCES
Amores F (1975) Diet of the red fox (Vulpes vulpes) in the western Sierra Morena (south Spain).

Doana Acta vert 2 (2):221-239.

Balestrieri A, Remonti L, Prigioni C (2004) Diet of the Eurasian badger (Meles meles) in an

agricultural riverine habitat (NW Italy). Hystrix 15 (2):3-12.

Benson JF, Chamberlain MJ (2007) Space use and habitat selection by female Louisiana black

bears in the Tensas River Basin of Louisiana. Journal of Wildlife Management 71

(1):117-126.
Brown WL, Wilson EO (1956) Character displacement. Systematic Zoology 5 (2):49-64.
Burnham KP, David R Anderson (1998) Model Selection and Inference: A Practical Information

Theoretical Approach. New York: Springer-Verlag.

Burnham KP, David R Anderson (2002) Model Selection and Multimodel Inference: A Practical

Information-TheoreticalApproach. 2d ed. New York: Springer-Verlag.

Cavallini P, Lovari S (1991) Environmental-Factors Influencing the use of Habitat in the red fox,

Vulpes vulpes. J Zool 223:323-339.

Carvalho JC, Gomes P (2001) Food habits and trophic niche overlap of the red fox, European wild

cat and common genet in the Peneda-Geres National Park. Galemys, 13 (2): 39-48.

Chesson P (2000) General theory of competitive coexistence in spatially-varying environments.

Theor Popul Biol 58 (3):211-237.

Delibes M (1974) Sobre la alimentacin y biologa de la gineta (Genetta genetta, L.) en Espaa.

Doana Acta Vertebrata, 1 (1): 143-199.

Delibes M, Rodrguez A, Ferreras P (2000) Action plan for the conservation of the Iberian lynx in

Europe (Lynx pardinus). Council of Europe Publishing, Strasbourg.

Delibes-Mateos M, de Simon JF, Villafuerte R, Ferreras P (2008) Feeding responses of the red fox
(Vulpes vulpes) to different wild rabbit (Oryctolagus cuniculus) densities: a regional

approach. Eur J Wildl Res 54 (1):71-78.

Doledec S, Chessel D, Gimaret-Carpentier C (2000) Niche separation in community analysis: A

new method. Ecology 81 (10):2914-2927.

Dormann CF, McPherson JM, Araujo MB, Bivand R, Bolliger J, Carl G, Davies RG, Hirzel A,

Jetz W, Kissling WD, Kuhn I, Ohlemuller R, Peres-Neto PR, Reineking B, Schroder B,

127

Chapter 2
Schurr FM, Wilson R (2007) Methods to account for spatial autocorrelation in the analysis
of species distributional data: a review. Ecography 30 (5):609-628.
Fernndez N, Delibes M, Palomares F, Mladenoff DJ (2003) Identifying breeding habitat for the

Iberian lynx: Inferences from a fine-scale spatial analysis. Ecol Appl 13 (5):1310-1324.

Fernndez N, Delibes M, Palomares F (2007) Habitat-related heterogeneity in breeding in a

metapopulation of the Iberian lynx. Ecography 30: 431-439.

Fernndez N, Delibes M, Palomares F (2006) Landscape evaluation in conservation: molecular

sampling and habitat modeling for the Iberian lynx. Ecological applications 16: 1037-1049.

Fernndez N, Palomares F (2000) The selection of breeding dens by the endangered Iberian lynx
(Lynx pardinus): implications for its conservation. Biological Conservation 94 (1):51-61.
Ferreras P, Beltran JF, Aldama JJ, Delibes M (1997) Spatial organization and land tenure system of

the endangered Iberian lynx (Lynx pardinus). J Zool 243:163- 189.

Hardin G (1960) Competitive exclusion principle. Science 131 (3409):1292-1297.

Garca-novo, F. y C. Martn-Cabrera -2005- Doana: Agua y Biosfera. Doana 2005, Confederacin
hidrogrfica del Guadalquivir, Ministerio de Medio Ambiente. Madrid.
Gause, G.F. (1934). The struggle for existence. Baltimore, MD: Williams & Wilkins.
Hilborn R, Mangel M (1997) The ecological detective. Confronting models with data. Monographs

in Population Biology 28: I-XVII, 1-315.

Hosmer, D.W., Lemeshow, S., 2000. Applied Logistic Regression.Wiley, NewYork.

Hurlbert SH (1978) Measurement of niche overlap and some relatives. Ecology 59 (1):67-77.
Hutchinson GE (1959) Homage to Santa-Rosalia or why are there so many kinds of animals. Am

Nat 93 (870):145-159.

Johnson CJ, Seip DR, Boyce MS (2004) A quantitative approach to conservation planning: using

resource selection functions to map the distribution of mountain caribou at multiple spatial

scales. Journal of Applied Ecology 41 (2):238-251.

Johnson JB, Omland KS (2004) Model selection in ecology and evolution. Trends Ecol Evol 19
(2):101-108.
Keitt TH, Bjornstad ON, Dixon PM, Citron-Pousty S (2002) Accounting for spatial pattern when

modeling organism-environment interactions. Ecography 25 (5):616-625.

Knapp RA, Matthews KR, Preisler HK, Jellison R (2003) Developing probabilistic models to
128

Chapter 2

predict amphibian site occupancy in a patchy landscape. Ecol Appl 13 (4):1069-1082.

Kruuk H, Parish T (1981) Feeding specialization of the european badger Meles meles in scotland.

J Anim Ecol 50 (3):773-788.

Kruuk H, Parish T (1985) Food, food availability and weight of badgers (Meles meles) in relation

to agricultural changes. Journal of Applied Ecology 22 (3):705-715.

Kufner MB (1986) Tamao actividad, densidad relativa y preferencias de hbitat de los pequeos

y medianos mamferos de Doana, como factores condicionantes de su tasa de predacin.

Tesis doctoral. Universidad Autnoma de Madrid.

Kufner MB, Moreno S (1989) Abundancia y amplitud de los desplazamientos de Apodemus

sylvaticus en cuatro bitopes de Doana que difieren en cobertura vegetal. Doana

Acta Vertebrata 16 (1):179-181.

Littell RC, GA Milliken, WW Stroup, RD Wolfinger (1996) SAS System for Mixed Models, Cary,

NC: SAS Institute Inc.

Martin J, Basille M, Van Moorter B, Kindberg J, Allain D, Swenson JE (2010) Coping with

human disturbance: spatial and temporal tactics of the brown bear (Ursus arctos).

Canadian Journal of Zoology 88 (9):875-883.

Martin R, Rodriguez A, Delibes M (1995) Local feeding specialization by badgers (Meles meles)

in a mediterranean environment. Oecologia 101 (1):45-50.

McCullagh, P. and Nelder, J. A. (1989). Generalized Linear Models, 2nd edition. London:

Chapman and Hall.

Metropolis N, Ulam S (1949) The monte carlo method. J Am Stat Assoc 44 (247):335-341.
Miller J, Franklin J, Aspinall R (2007) Incorporating spatial dependence in predictive vegetation
models. Ecol Model 202 (3-4):225-242.
Moreno S, Beltran JF, Cotilla I, Kuffner B, Laffite R, Jordan G, Ayala J, Quintero C, Jimenez A,

Castro F, Cabezas S, Villafuerte R (2007) Long-term decline of the European wild

rabbit (Oryctolagus cuniculus) in south-western Spain. Wildl Res 34 (8):652-658.

Moreno S, Kufner MB (1988) Seasonal patterns in the wood mouse-population in Mediterranean

scrubland. Acta Theriol 33 (1-11):79-85.

Palomares F (1994) Site fidelity and body mass on home-range size of Egyptian mongooses.

Canad J Zool. 72, 465-469.

129

Chapter 2
Palomares F (1986) Ecologa de la gineta y del meloncillo en el Parque Nacional de Doana. Tesis

de licenciatura. Universidad de Granada. 186 pp.

Palomares F (2001) Vegetation structure and prey abundance requirements of the Iberian lynx:

implications for the design of reserves and corridors. Journal of Applied Ecology

38 (1):9-18.

Palomares F, Delibes M (1988) Time and Space Use by Two Common Genets (Genetta genetta) in

the Doana National Park, Spain. J Mammal 69 (3):635-637.

Palomares F, Delibes M (1991) Dieta del meloncillo, Herpestes ichneumon, en Coto del Rey, norte

del Parque Nacional de Doana. Doana Acta Vertebrata, 18: 187-194.

Palomares F, Delibes M (1991). Ecologa comparada de la gineta, Genetta genetta (L.) y el meloncillo,

Herpestes ichneumon, (L.) (Mammalia, Viverridae) en Doana (SO de la Pennsula

Ibrica). Bol. R. Soc. Esp. Hist. Nat. (Secc. Biol.), 87: 257-266.

Palomares F, Delibes M (1993) Key habitats for Egyptian mongooses in Doana National-park,

south-western spain. Journal of Applied Ecology 30 (4):752-758.

Palomares F, Delibes M (1993) Resting ecology and behavior of Egyptian mongooses (Herpestes
ichneumon) in southwestern spain. J Zool 230:557-566
Palomares F, Delibes M (1994) Spatiotemporal ecology and behavior of european genets in

southwestern spain. J Mammal 75 (3):714-724.

Palomares F, Delibes M, Ferreras P, Fedriani JM, Calzada J, Revilla E (2000) Iberian lynx in a

fragmented landscape: Predispersal, dispersal, and postdispersal habitats. Conserv Biol 14

(3):809-818.
Palomares F, Delibes M, Revilla E, Calzada J, Fedriani JM (2001) Spatial ecology of Iberian lynx

and abundance of European rabbits in southwestern Spain. Wildl Monogr (148):1-36.

Palomares F, Ferreras P, Fedriani JM, Delibes M (1996) Spatial relationships between Iberian lynx

and other carnivores in an area of south-western Spain. Journal of Applied Ecology 33

(1):5-13.
Palomares F, Rodriguez A, Laffitte R, Delibes M (1991) The status and distribution of the Iberian

Lynx Felis-Pardina (Temminck) in Coto Doana area, Sw Spain. Biological Conservation

57 (2):159-169.

Revilla E, Palomares F (2002) Does local feeding specialization exist in Eurasian badgers? Can J
130

Chapter 2

Zool-Rev Can Zool 80 (1):83-93.

Romn J, A Barn, E Revilla (2010) Evaluacin de los efectos del trnsito a motor sobre especies

y comunidades de inters en el Espacio Natural de Doana. Consejera de Medio

Ambiente, Junta de Andaluca. Estacin Biolgica de Doana, CSIC. 236 pp.

Roper TJ, Lups P (1995) Diet of badgers (Meles meles) in central switzerland - an analysis of

stomach contents. Z Saugetierkd-Int J Mamm Biol 60 (1):9-19.

Rosalino LM, Loureiro F, MacDonald DW, Santos-Reis M (2005) Dietary shifts of the badger
(Meles meles) in Mediterranean woodlands: an opportunistic forager with seasonal

specialisms. Mamm Biol 70 (1):12-23.

Rosenzweig ML, Abramsky Z, Brand S (1984) Estimating species interactions in heterogeneous

environments. Oikos 43 (3):329-340.

Santos MJ, Pinto BM, Santos-Reis M (2007) Trophic niche partitioning between two native and

two exotic carnivores in SW Portugal. Web Ecology 7:1-62.

Smith PA (1994) Autocorrelation in logistic-regression modeling of species distributions. Glob

Ecol Biogeogr Lett 4 (2):47-61.

Soto C, Desnia S, Palomares F (2012) Nonbiological factors affecting track censuses: implications
for sampling design and reliability.
Thioulouse J, Chessel D, Doldec S, Olivier J (1997) Ade-4: a multivariate analysis and graphical

display software. Statistics and Computing, 7:75-83.

Villafuerte R, Kufner MB, Delibes M, Moreno S (1993) Environmental-factors influencing the

seasonal daily activity of the European rabbit (Oryctolagus cuniculus) in a Mediterranean

area. Mammalia 57 (3):341-347.

Villafuerte R, Lazo A, Moreno S (1997) Influence of food abundance and quality on rabbit

fluctuations: Conservation and management implications in Doana National Park (SW

Spain). Rev Ecol-Terre Vie 52 (4):345-356.

Villafuerte R, Moreno S (1997) Predation risk, cover type and group size in European rabbits in

Doana (SW Spain).Acta Theriologica, 42 (2): 225-230.

Viota M, Lpez-Bao JV, Palomares F, Rodrguez A (2012) Shift in microhabitat use as a mechanism
allowing the coexistence of victim and killer carnivore predators.

131

Chapter 2
Virgs E (2002) Are habitat generalists affected by forest fragmentation? A test with Eurasian

badgers (Meles meles) in coarse-grained fragmented landscapes of central Spain. J Zool

258 (3):313-318.

Virgs E, Casanovas JG (1999) Environmental constraints at the edge of a species distribution, the

Eurasian badger (Meles meles L.): a biogeographic approach. Journal of Biogeography 26

(3):559-564.
Virgs E, Llorente M, Cortes Y (1999) Geographical variation in genet (Genetta genetta L.) diet: a

literature review. Mammal Review 29 (2):117-126.

Yamaguchi N, Rushton S, Macdonald DW (2003) Habitat preferences of feral American mink in

the Upper Thames. J Mammal 84 (4):1356-1373.

Zapata SC, Travaini A, Ferreras P, Delibes M (2007) Analysis of trophic structure of two carnivore

132

assemblages by means of guild identification. Eur J Wildl Res 53 (4):276-286.

Chapter 2

Supplementary information
Table 1. Results of the outlying mean index analysis depicting relationships of occurrences of
Carnivora species to the suite of environmental variables measured across sampling sites in DNP.
Inertia = variance or weighted sum of squared distances to the origin of the environmental axes;
OMI = outlying mean index (marginality) or the deviation of a particular species distribution
from the overall mean habitat conditions (origin of outlying mean index axes), described by the
environmental variables; Tol = tolerance index, which is analogous to niche breadth or spatial
variance of an organisms niche across the measured environmental variablesa function of
all sampling sites with which the species is associated; RTol = residual tolerance. Italicized terms
represent the percentages of variability corresponding to a specific statistic. P = frequency based
on number of random permutations (out of 10,000) that yielded a higher value than the observed
outlying mean index (P 0.05 indicates a significant influence of the environmental variables for a
species).
Species

OMI

Tol

5.73

39.30

17.40

43.30

0.00040

1.61

6.54

26.40

31.90

41.70

0.00010

0.35

2.30

7.45

2.70

18.00

79.30

0.09329

11.79

0.25

1.69

9.83

3.00

13.60

83.40

0.06429

10.89

0.04

4.99

9.51

0.30

12.40

87.30

0.63804

Inertia

OMI

Tol

Lynx pardinus

13.22

5.19

1.35

Genetta genetta

15.66

4.13

Herpestes ichneumon

9.39

Meles meles
Vulpes vulpes

Rtol

Rtol

Table 2. Loadings of environmental variables for the first two axes of the outlying mean index
analysis. Values represent the best linear combination that explains occurrences of Carnivora at
DNP. Variables are described in Table 1.
Variable

Axis 1

Axis 2

%SB

0.2014

0.0251

%B

-0.2361

-0.1876

%T

-0.2138

0.1877

DH

-0.5163

-0.1776

Traf

0.1279

0.3091

eBP

0.0237

-0.5020

DW

0.2807

0.2649

DV

-0.0605

0.5255

Ra

0.2431

-0.3693

SM

0.0365

0.0165

Tot

0.2635

-0.2575

Eigenvalue

0.3766

0.0003

Percentage variance explained

0.9986

0.0009

133

Chapter 2
Figure 1. Contribution of environmental, human and prey variables to the first axes produced by
the OMI analysis. The length of the arrow describes the relative importance of each variable in the
analysis, and the direction of the arrow indicates among-variable correlations.

DV

Traf
DW
T
SM

DH

SB

Axis 1

B
Tot
Ra
eBP

134

CHAPTER 3

Species abundances in a community

of sympatric carnivores: a trade-off
between habitat selection and interspecific
interactions?
Abundancia de especies en una comunidad de carnvoros
simptridos: existe un equilibrio entre la seleccin de
hbitat y las interacciones interespecficas?

135

136

Chapter 3

ABSTRACT

In a carnivore guild, the occurrence of larger species can exerts a significant

effect on the relative abundances of smaller guild members, whereas similar-sized

species may potentially not exhibit any density-dependent relationships. When
modelling habitat use, the species relative abundance could be therefore a misleading
proxy of habitat quality if a larger predator may be limiting their abundance in the
potentially best optimal habitat.

We analyzed the potential effect of the co-occurrence of sympatric

carnivores in the habitat use pattern of five medium-sized carnivore species

(Iberian lynx, Eurasian badger, common genet, Egyptian mongoose and red fox) in
a Mediterranean protected area in southwest Spain (Doana National Park; DNP),
by logistic and regression generalized linear mixed models. We carried out track
censuses during 2007-2008 and 2008-2009 within 2 x 2 km2 quadrants and we
found that the relative abundance of foxes, mongooses and genets were negatively
affected by the occurrence of the top predator of the carnivore guild, the Iberian
lynx. However, abundance of genets and mongooses were positively correlated
with the occurrence of mongooses and badgers, respectively, also improving the
explanatory power of the habitat use models of those species. Intraguild predation
as well as coincidence in the use of resources between species could explain the
patterns of co-occurrence found. We suggest that the particular identity of predator
species within a guild strongly affects the relative competitive advantage of each
guild member and we propose that relative abundances of predators at the same
trophic level should not be overlooked and included as covariates when studying
carnivore community structure and habitat use.

137

Chapter 3

RESUMEN

En un gremio de carnvoros, la ocurrencia de las especies ms grandes puede

ejercer un efecto significativo sobre las abundancias relativas de los miembros ms

pequeos de la comunidad, mientras que las especies de tamao similar no deben presentar
aparentemente ninguna relacin denso-dependiente entre ellas. En consecuencia, cuando
se modela el uso del hbitat por una especie, la abundancia relativa de la misma puede
ser un indicador errneo de la calidad de hbitat si un depredador superior est limitando
la abundancia de la especie en su mejor hbitat potencial.
En este estudio analizamos el efecto potencial de la co-ocurrencia de carnvoros
simptridos en el patrn de uso del hbitat de cinco especies de carnvoros de mediano
tamao (lince ibrico, tejn, gineta, meloncillo y zorro) en un rea protegida en el suroeste
de Espaa (Parque Nacional de Doana), a travs de modelos lineares generalizados
mixtos logsticos y de regresin. Llevamos a cabo censos de rastros durante los aos
2007-2008 y 2008-2009 en cuadrculas de 2 x 2 km2 y encontramos que la abundancia
relativa de zorros, meloncillos y ginetas se vio afectada negativamente por la ocurrencia
del carnvoro superior del gremio, el lince ibrico.

Sin embargo, la abundancia de ginetas y meloncillos se correlacion positivamente

con la ocurrencia de meloncillos y tejones, respectivamente, mejorando tambin el poder

explicativo de los modelos de hbitat para dichas especies. La depredacin intragremial
as como la coincidencia en el uso de los recursos entre especies pueden explicar dichos
patrones de co-ocurrencia detectados. Sugerimos que la identidad particular de una especie
de depredador dentro de un gremio afecta en gran medida su ventaja competitiva dentro
de la comunidad y proponemos que la abundancia relativa de predadores pertenecientes
al mismo nivel trfico debe incluirse como variable explicativa en estudios de uso del
hbitat y estructura de las comunidades de mamferos carnvoros.
138

Chapter 3

INTRODUCTION

The statement that density of a species in a given habitat is a direct measure

of the quality of that habitat for the species is a widely accepted assumption in
habitat use studies (but see VanHorne 1983). Nevertheless, some species may not
exhibit the highest densities in their potentially best optimal habitats according to
the species life-history traits, as their relative abundances and pattern of habitat use
can be affected by interspecific interactions with other species at the same trophic
level (Case and Gilpin 1974, Estes and Palmisano 1974, Menge and Sutherland
1987, Durant 1998).

Among mammal carnivores, body size influences the outcomes of

interspecific interactions (Palomares and Caro 1999). With the exception of some
species that prey in packs (e.g., Rogers and Mech 1981, Paquet and Carbyn 1986),
interference competition is inflicted by a larger carnivore species on a smaller
one killing it (termed as intraguild predation (Polis et al. 1989)) or excluding it
from habitat patches or prey carcasses (Palomares and Caro 1999). These highly
asymmetrical interactions mediated by body size are common (e.g. Fedriani et al.
1999, Kamler et al. 2003, Durant et al. 2010) being the smaller species almost
invariably the loser due to their subordinate position (Palomares and Caro 1999,
Prugh et al. 2009, Roemer et al. 2009). Nevertheless, smaller members of a guild
may also kill cubs, young or subadult individuals of the larger species (Palomares
and Caro 1999). In the absence of predator species, small carnivores should be
ideally distributed based on habitat quality and preferred food availability (Van der
Meer and Ens 1997, Roemer et al. 2009), but in most natural communities smaller
species are potentially subject to top-down effects that mediate their ability to use
preferred habitat and limit their access to high-quality foraging areas (Palomares
139

Chapter 3

and Caro 1999, Ritchie and Johnson 2009). Moreover, in some carnivores,
intraguild predation has a considerable impact on small carnivore mortality rates
and consequently a negative effect on the relative abundance of the species in
the presence of the larger predator (Ralls and White 1995, Sovada et al. 1998).
In such cases, due to these potential predation-driven direct effects or fear-driven
indirect effects, the relative abundances of some predators may not be related to
prey availability or vegetation type. Densities of smaller members of a guild may
represent a trade-off between suitable areas that satisfy their basic requirements and
areas where probability of encounters with larger predators are null or lower (i.e., to
diminish the risk of being killed).

In this paper, we examine the potential effect of the co-occurrence of different

sympatric carnivores on the habitat use pattern of five small and medium-sized
carnivore species in a Mediterranean protected area in southwest Spain, namely
Doana National Park (DNP). On previously fitted fine-scale habitat use models for
a guild of five carnivore species we analysed whether the inclusion of the relative
abundances of any carnivore species improves the explanatory power of habitat
models (Johnson and Omland 2004, Burnham and Anderson 2002). The studied
carnivore guild is composed by the Iberian lynx (Lynx pardinus, 9-15 kg), Eurasian
badger (Meles meles, 7-8 kg), the red fox (Vulpes vulpes, 5-7 kg), the Egyptian
mongoose (Herpestes ichneumon, 3 kg) and the common genet (Genetta genetta,
2 kg). In DNP these species form a community of continuum degree of habitat
and trophic specialization predators with different degree of habitat use overlap.
It has been described aggressive interactions between some of these carnivores in
DNP, which always involved lynxes as the top predator. The Iberian lynx uses to
kill genets, mongooses and foxes (Palomares et al. 1996, Fedriani 1997), but no
140

Chapter 3

aggressive interactions between lynx and badger has been reported. On the other
hand, aggressive interactions between badgers, foxes, mongooses and genets have
been unreported, although according to the theory differences in body size among
some of these species could support such interactions (Palomares and Caro 1999,
Fedriani et al. 2000).

Specifically, we tested the hypothesis that in our carnivore guild the spatial

occurrence of larger species exerts a significant effect on the relative abundances

of other smaller guild members after controlling by species-specific habitat use
patterns. Based in our previous knowledge on the reported aggressive interaction
between species in the study area and in the general knowledge of the effect of
body size on carnivore interactions (reviewed in Palomares and Caro 1999), our
predictions were those outline in Figure 1: a) we expected a negative effect of the
lynx presence and/or abundance on the relative densities of foxes, mongooses and
genets; b) abundance of foxes and badgers should negatively affect abundance of
mongooses and genets and c) it should be no effect between badgers and foxes,
badgers and lynxes and mongooses and genets.

METHODS
Study area

DNP is a fully protected and flat sandy area at sea level in southwestern

Spain (550 km2 379N, 626W), located on the west bank of the Guadalquivir
River mouth. The climate is Mediterranean sub-humid, characterized by dry, hot
summers and mild, wet winters. Average annual rainfall is 500-600 mm. There are
three main biotopes in the park: scrubland, dunes, and marsh (Valverde 1958). The
dune area is situated at the western border of the protected area limited by the
141

Chapter 3

Atlantic Ocean and the marsh area at the northern and eastern borders limited by the
Guadalquivir River. The Mediterranean scrubland represents approximately half of
the National Park surface area and is mainly characterised by heterogeneous patches
of xerophytic species such as Halimium sp. and Cistus sp., and hydrophytic species
such as Erica sp., with some patches of Juniperus phoenica and Pistacia lentiscus
shrubs. Interspersed among the scrubland are scattered cork oak trees (Quercus
suber) and wild olive trees (Olea europea), and a few patches of pine Pinus pinea
Figure 1. Sketch of the possible expected interrelationships among the different carnivore
species studied here. Single-headed arrow and straight lines represent aggressive interactions that
have been previously reported in the literature for those species in the study area; single-headed
arrow and dashed lines represent potential aggressive interactions based in the body size of the
involved species but that have been previously unreported; and double-headed arrow and dashed
lines represent potential neutral expected interactions between similar-sized species in the DNP
carnivore guild.

142

Chapter 3

and eucalyptus Eucalyptus sp. plantations.

In addition to the studied species (i.e., red fox, Eurasian badger, Egyptian

mongoose, common genet and Iberian lynx), there are other carnivores such as
least weasel (Mustela nivalis), European polecat (Mustela putorius), Eurasian otter
(Lutra lutra) and wild cat (Felis silvestris), but they were excluded from the study
because of their scarcity in the area.
Sampling protocol

To assess the co-occurrence of carnivore species and to obtain relative

densities for foxes, badgers, lynxes, mongooses and genets we divided the study
area into 69 2 x 2 km quadrants following UTM coordinates (Figure 2). Quadrants
were sampled throughout the wet seasons of 2007-2008 and 2008-2009. In each
quadrant, a 3 km-length survey route was slowly walked searching for carnivore
tracks. The survey routes were located along firebreaks and car roads between 2 and
12 m wide. Once a continuous track that crossed from one side to the other across
the pathway was detected, we georeferenced it using a global positioning system.
For the two sampled seasons altogether we resampled 21 squares (leaving at least 7
days between samplings) a second time until completing 3 km, because there were
insufficient available paths within the square to achieve this distance. Thus, as we
had more censuses than quadrants we averaged track counts to get a single value per
quadrant. We always carried out surveys at least 3 days after any rainfall.

Data about prey availability and vegetation variables included in habitat

models were obtained through track censuses in transects of 25 m in length and

approximately 1.7 m wide and from visually estimates in 25 m diameter circle
situated in the same places where transects for prey (see Soto et al. 2012 for more
details).

143

Chapter 3

Data analysis

To test the effect of the abundance of other species on the abundance of a

given species according this outlined in Figure 1, we first fitted fine-scale habitat
use models for each species (Chapter 2). After, we added the relative abundance
(Kilometric Abundance Index; KAI) or presence of the different carnivore species
to the best approximating habitat model describing abundance of each species
(hereafter null models) and observed whether models improved (i.e. a decrease of
the Akaike Information Criterion corrected for small sample size; AICc). According
to the information-theoretic approach, models with AICc < 2 from the model with
the lowest AICc are considered to be virtually the same plausible models (Burnham
and Anderson 2002). Hence, in spite of the existence of certain model selection
uncertainty in multimodel inference, statistical models that decrease the AICc of
a reference model more than 2 units could be considered as more parsimonious
models and to best fit the data. Thus, for the specific aims of our study, when AICc
< -2 and model coefficient () was negative we considered that there were support
to the predictions about the abundance or presence of a given species affected the
abundance or presence of the target species; when AICc < -2 and model coefficient
() was positive we considered that both species could coincide in habitat resources
used, but no negative interaction occurred between them; and finally when -2 >
AICc < 2 we considered that abundance or presence of a given species did not
affect the abundance or presence of the target species. Note that if our hypothesis
is right, the inclusion of the relative abundance of presence of smaller species in
the model of habitat selection of larger species should have either no effect or if
any (i.e., AICc > 2), should be positive. Hence, we included as a control of
hypothesis all possible carnivore species in null models for each target species to
144

Chapter 3

test if this was the case.

Null models were defined based on previous habitat use studies for each species in
the DNP (Chapter 2). These models included different variables related to landscape
features, vegetation type, prey availability and human disturbance for each species
(see Table 1 for variable description). The general null model for the occurrence of
mongooses (HI), foxes (VV), badgers (MM) and for the presence of genets (GG)
and lynxes (LP) at site j was specified as:

logit(HI j) = HI + 1DW j + 2DV j + 3Tot j + 4Hum j + HI autocovj

logit(VV j) = VV + 1%B j + 2%T j + 3DW j + 4SMj + 5Humj + VV autocovj
logit(MM j) = MM + 1%B j + 2%SB j + 3eBP j + 4DVj + 5DW j + 6Raj + 7Hum j + MM autocovj
logit(GG j) = GG + 1%B j + 2%T j + 3DW j + 4SMj + 5DH j + 6Traf j + 7Hum j
logit(LP j) = LP + 1%B j + 2%SB j + 3eBP j + 4DVj + 5DW j + 6Raj + 7Hum j

where i is the species-specific intercept, the coefficients 1,..., 4, 1,..., 5, 1,...,

7, 1,..., 7 and 1,..., 7 represent effects of the associated explicative variables
(Table 1) on species i, and i is the effect of the autocovariate on species i (see text
below).
Candidate model equations were fitted using Generalized Linear Mixed Models
(GLMM) with binomial (for genets and lynxes) or negative binomial (for mongooses,
badgers and foxes) error structure using the glmmADMB and lmer library (Fournier
et al. 2012, Skaug et al. 2012) respectively in R 2.15.2 (R Development Core Team
2008). The dependent variable was the number of tracks/km. Observer and quadrant
were modelled as random effects, humidity as a covariate (Soto et al. 2012), and the
distance covered per quadrant during track censuses as an offsset. We incorporated
145

Chapter 3

an autocovariate as an additional explanatory variable in the GLMM models to

account for fine-scale spatial variation in the data by estimating the degree to which
the response at each location reflects response values at surrounding locations
(Dormann 2007). This extra parameter is intended to capture spatial autocorrelation
originating from endogenous processes such as movement of censused individuals
between sampling sites (Smith 1994, Keitt et al. 2002, Yamaguchi et al. 2003).
Table 1. Variables used to define fine-scale habitat use null models for badgers, mongooses, foxes,
genets and lynxes.

Variable

Code

Definition
Mean cover of bushes per quadrant calculated by averaging values
obtained at the different sampling points of vegetation censuses
Mean cover of trees per quadrant calculated by averaging values
obtained at the different sampling points of vegetation censuses
Measured in meters using a Euclidean distance-based approach
from the quadrant centre to the nearest permanently flooded natural
or artificial pond (i.e. dug for the cattle at zones were the water table
is higher) in a digitized water sources cover layer of DNP
Measured in meters from the quadrant centre to the ecotone between
the marshland and the Mediterranean scrubland (locally called La
Vera)

Units

tall shrub

%B

trees

%T

distance to
water

DW

distance to
La Vera

DV

small
mammals

SM

Kilometric Abundance Index of small mammals per quadrant

Tracks/
calculated as the mean between prey indexes at the different
km
sampling points of prey censuses per quadrant

rabbits

Ra

Kilometric Abundance Index of rabbits per quadrant calculated as

Tracks/
the mean between prey indexes at the different sampling points of
km
prey censuses per quadrant

distance to
antropic
edge

DH

Measured in meters from the quadrant centre to the nearest protected

area border influenced by humans (i.e., excluding the beach and
marshland edges)

Mean daily traffic (MDT), adjusted for road length (m) and type of
road (i) in the quadrant (= (MDTi * mi)) derived from Romn et
al. 2010

car/m

Traffic
index
Humidity

Hum

Environmental humidity on census day obtained from a

meteorological station located inside DNP. http://icts.ebd.csic.es

Observer

Obs

Observer who carried out censuses

146

%
%
m

%
no
units

Chapter 3

RESULTS

Null models of foxes, mongooses and genets significantly improved (AICc

< -2) after including the occurrence and/or presence of lynxes (Table 2). The
three species were less abundant in areas where lynxes exhibited a higher relative
abundance (Fig. 3a, c and e). Habitat models of foxes, genets (Fig. 3b) and badgers
also improved after including the relative density of mongooses (Table 2) but the
Figure 2. Map of the DNP showing its location in southwest of the Iberian Peninsula and the 22
km2 quadrants where carnivore track censuses were carried out during the wet seasons of 2007-2008
and 2008-2009. The dashed zone represents the marshland area of the DNP.

147

Chapter 3

three species exhibited higher relative densities in areas where mongooses were
also more abundant.

The inclusion of the presence of genets and the relative abundance of badgers

(Fig. 3d) also improved the habitat model of mongooses that were more abundant
in areas where genets were detected or where the occurrence of badgers was higher
(Table 2). The inclusion of any of the smaller members of the guild significantly
improved the habitat model of the Iberian lynx (Table 2).

DISCUSSION

Our results support the hypothesis that in a carnivore guild, the occurrence

of some species exerts a significant effect on the relative abundance of other guild
members after controlling by species-specific habitat use patterns. As we previously
hypothesized, the relative abundance of foxes, mongooses and genets in DNP was
negatively affected by the presence or occurrence of lynxes, the larger reported
intraguild predator for all of them (Palomares and Caro 1999). In fact, it had been
previously described for Doana as foxes, mongooses and genets avoided temporal
or spatially highly used areas by lynxes (Palomares et al. 1996, 1998, Fedriani et
al. 1999, Viota et al. 2012). Similar results have been obtained in other carnivore
communities involving African wild dogs Lycaon pictus and cheetahs Acinonyx
jubatus with lions Panthera leo and spotted hyenas Crocuta crocuta (Laurenson
1995, Creel and Creel 1996, Durant 2000).

Also as predicted, the relative abundance of smaller species did not affect

the relative abundance of larger ones; in our case lynxes, badgers and foxes.
However, although theory predicts that larger species such as foxes and badgers
might negatively interact with genets and mongooses, results did not confirm
148

Chapter 3
Table 2. Results from negative binomial (foxes, badgers and mongooses) and logistic regression
models (genets and lynxes) investigating the effects of the occurrence and/or presence of predator
species on each carnivore species in the DNP. We report the small sample-sizeadjusted Akaikes
information criteria (AICc), the difference in AICc between each model and the null model (AICc),
the model coefficient () with standard error (SE) as well as its P-value of those models that
decreased the null model 2. Variables are described in Table 1.
Species

Models

SE() P-value

AICc

AICc

0. Intercept only
1. Null model
%B + %T + DW + SM + DH
+ Traf + Hum + autocov
2. Null model + KAIl
3. Null model + KAIb
4. Null model + KAIf
5. Null model + KAIIm
6. Null model + Lynx

175.23

12.55

162.68

0.00

159.75
162.07
161.07
153.01
160.62

-2.94
-0.61
-1.61
-9.67
-2.06

-1.14
.
.
0.26
-1.13

0.72
.
.
0.08
0.59

0.05
.
.
0.00
0.08

0. Intercept only
1. Null model
DW + DV + Tot + Hum +
autocov
2. Null model + KAIl
3. Null model + KAIb
4. Null model + KAIf
5. Null model + KAIg
6. Null model + Lynx
7. Null model + Genet

918.52

38.76

879.76

0.00

877.23
874.89
879.74
879.63
881.56
863.83

-2.53
-4.87
-0.02
-0.13
1.80
-15.93

-0.31
0.06
.
.
.
0.82

0.14
0.02
.
.
.
0.19

0.03
0.01
.
.
.
0.00

0. Intercept only

1277.53

24.66

1252.87

0.00

1253.99
1254.09
1250.46
1254.84
1249.69
1250.91

1.12
1.22
-2.41
1.97
-3.18
-1.96

.
.
0.03
.
-0.27
.

.
.
0.01

.
.
0.03
.
0.02
.

Genetta genetta

Herpestes
ichneumon

Vulpes vulpes
1. Null model
%B + %T + DW + SM + Hum
+ autocov
2. Null model + KAIl
3. Null model + KAIb
4. Null model + KAIm
5. Null model + KAIg
6. Null model + Lynx
7. Null model + Genet

0.12
.

149

Chapter 3
Meles meles

910.64

9.29

901.35

0.00

901.98
903.12
890.65
903.04
900.83
902.95

0.63
1.77
-10.70
1.69
-0.52
1.60

.
.
0.11
.
.
.

.
.
0.03
.
.
.

.
.
0.00
.
.
.

0. Intercept only
153.89
1. Null model
%B + %SB + DW + eBP + DV
111.92
+ Ra + Hum
2. Null model + KAIb
112.27
3. Null model + KAIf
111.00
4. Null model + KAIm
113.92
5. Null model + KAIIg
110.43
6. Null model + Genet
112.21

41.97

0.00

0.35
-0.92
2.00
-1.49
0.29

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

0. Intercept only
1. Null model
%SB + %B + eBP + DW + DV
+ Ra + Hum + autocov
2. Null model + KAIl
3. Null model + KAIf
4. Null model + KAIm
5. Null model + KAIg
6. Null model + Lynx
7. Null model + Genet

Lynx pardinus

KAIl; Kilometre Abundance Index of lynx

KAIb; Kilometre Abundance Index of badger
KAIf; Kilometre Abundance Index of fox
KAIm; Kilometre Abundance Index of mongoose
KAIg; Kilometre Abundance Index of genet
Genet: Presence/Absence of Genet (1/0)
Lynx; Presence/Absence of Lynx (1/0)

this prediction. Contrarily, positive correlations between some of these species

(mongooses with badgers and foxes) were found. The lacks of negative interactions
may be explained by the fact that the degree of intra-guild interference is thought
to depend on body-size differences: at small differences, attacks are less likely to
occur (Buskirk 1999, Donadio and Buskirk 2006), and the positive relationships
between species might simply be explained by the fact that they might exploit
the same resources or that the distribution of the resources used by each species,
although different, might overlap.
150

Chapter 3

The same explanation might apply to the positive association found between

mongooses and genets; two similar-sized species. Genets and mongooses partially
overlap in their habitat use in DNP (Chapter 2). Furthermore, previous studies have
also shown that both mongooses and genets use similar microhabitats such as dense
bushes or thickets for activity and resting in Doana (Palomares and Delibes 1993),
and that a large fraction of their diets may coincide (Palomares and Delibes 1991).
On the same way, although mongooses exhibit a certain higher degree of habitat
specialization than foxes and badgers, the niche overlap of all three predators is large
probably due to their trophic and/or habitat generalism (Chapter 2). Additionally,
mustelids are not species with a high aggressive nature (Palomares and Caro 1999),
so it is reasonable to think that they can coexist with smaller predators without any
interference.
Our results demonstrate that the inclusion of the relative abundance of other
carnivore species in habitat models for a given species may significantly improve
them. Models based on species relative densities that excluded other predators
occurrence could be a misleading proxy of habitat quality when studying patterns of
habitat use by species, and this may be particularly important for smaller predators
that very often are intraguild prey in natural communities. Smaller species might be
more abundant at sub-optimal or marginal habitats due to predation-driven direct
or fear-driven indirect effects from larger predators. There are some examples of
other carnivore species that have been reported as exhibiting limited patterns of
distribution and abundance due to interference competition with larger predators
(Berger and Gese 2007, Peterson 1995, Thurber et al. 1992, White and Garrott
1997, Burrows 1995, Durant 2000). However, habitat use models for carnivore
species at the apex of their ecological communities that are based on vegetation
151

Chapter 3

and/or landscape features as well as on prey availability should be enough to design

adequate management plans that aim to restore habitat for the species.

Figure 3. Relationships between the probability of genet presence and the numbers of lynx (a) and
mongoose tracks (b) detected; between the numbers of mongooses tracks and that of lynxes (c) and
badgers (d); and between the number of fox tracks and the presence of lynx (least square means and
their standard errors are represented) (e).

152

Chapter 3

ACKNOWLEDGMENTS
This study was financed by the projects CGL2004-00346/BOS (Spanish Ministry of Education and
Science) and 17/2005 (Spanish Ministry of the Environment; National Parks Research Programme),
and sponsored by Land-Rover Espaa S.A. We are especially thankful to J.C. Rivilla and S. Desnia
for their assistance during fieldwork. C. Soto received a JAE predoctoral grant from CSIC (Spanish
National Research Council).

REFERENCES
Berger KM, Gese EM (2007) Does interference competition with wolves limit the distribution and

abundance of coyotes? Journal of Animal Ecology 76:10751085.

Burnham KP, Anderson DR (2002) Model selection and multimodel inference. A practical

information-theoretic approach. 2nd ed. Springer, New York.

Burrows R (1995) Demographic changes and social consequences in wild dogs, 19641992. In

Serengeti II. Dynamics, management and conservation of an ecosystem: 385400.

Sinclair, A. R. E. & Arcese, P. (Eds). Chicago: University of Chicago Press.

Buskirk SW (1999) Mesocarnivores of Yellowstone. Pages 165187 in T. W. Clark, A. Peyton

Curlee, S. C. Minta, and P. M. Kareiva, eds. Carnivores in ecosystems: the Yellowstone

experience. Yale University Press, London.

Case TJ, Gilpin ME (1974) Interference Competition and Niche Theory. Proc Natl Acad Sci U S A
71 (8):3073-3077.
Creel S, Creel NM (1996) Limitation of African wild dogs by competition with larger carnivores.
Conservation Biology, 10, 526538.
Donadio E, Buskirk SW (2006) Diet, morphology, and interspecific killing in carnivora. Am Nat
167:524536.
Dormann CF, McPherson JM, Araujo MB, Bivand R, Bolliger J, Carl G, Davies RG, Hirzel A, Jetz
W, Kissling WD, Kuhn I, Ohlemuller R, Peres-Neto PR, Reineking B, Schroder B, Schurr
FM, Wilson R (2007) Methods to account for spatial autocorrelation in the analysis of
species distributional data: a review. Ecography 30 (5):609-628.
Durant SM (1998) Competition refuges and coexistence: an example from Serengeti carnivores. J
153

Chapter 3
Anim Ecol 67 (3):370-386.
Durant SM (2000) Living with the enemy: avoidance of hyenas and lions by cheetahs in the
Serengeti. Behavioral Ecology, 11, 624632.
Durant SM, Craft ME, Foley C, Hampson K, Lobora AL, Msuha M, Eblate E, Bukombe J, McHetto
J, Pettorelli N (2010) Does size matter? An investigation of habitat use across a carnivore
assemblage in the Serengeti, Tanzania. J Anim Ecol 79 (5):1012-1022.
Estes JA, Palmisan JF (1974) Sea Otters - Their Role In Structuring Nearshore Communities.
Science 185 (4156):1058-1060.
Fedriani JM (1997) Relaciones interespecficas entre el lince ibrico, Lynx pardina, el zorro,

Vulpes vulpes, y el tejn, Meles meles en el Parque Nacional de Doana. Universidad de

Sevilla. Facultad de Ciencias Biolgicas. 181 Pg.

Fedriani JM, Palomares F, Delibes M (1999) Niche relations among three sympatric Mediterranean
carnivores. Oecologia 121 (1):138-148.
Fedriani JM, Fuller TK, Sauvajot RM, York EC (2000) Competition and intraguild predation among
three sympatric carnivores. Oecologia 125:258270.
Fournier DA, Skaug HJ, Ancheta J, Ianelli J, Magnusson A, Maunder M, Nielsen A and Sibert

J (2012). AD Model Builder: using automatic differentiation for statistical inference of

highly parameterized complex nonlinear models. Optim. Methods Softw., *27*, pp. 233-249.

Johnson JB, Omland KS (2004) Model selection in ecology and evolution. Trends Ecol Evol 19
(2):101-108.
Kamler JF, Ballard WB, Gilliland RL, Mote K (2003) Spatial relationships between swift foxes and
coyotes in northwestern Texas. Can J Zool-Rev Can Zool 81 (1):168-172.
Keitt TH, Bjornstad ON, Dixon PM, Citron-Pousty S (2002) Accounting for spatial pattern when
modeling organism-environment interactions. Ecography 25 (5):616-625.
Laurenson MK (1995). Implications of high offspring mortality for cheetah population dynamics.
Pages 385399 in A. R. E. Sinclair and P. Arcese, eds. Serengeti II: dynamics, management
and conservation of an ecosystem. University of Chicago, Chicago.
Menge BA, Sutherland JP (1987) Comrnunity regulation: variation in disturbance, competition,

and predation in relation to environmental stress and recruitment. Am Nat 130: 730-757.

Palomares F, Caro TM (1999) Interspecific killing among mammalian carnivores. Am Nat 153
154

Chapter 3
(5):492-508.
Palomares F, Delibes M (1991) Ecologa comparada de la gineta, Genetta genetta (L.) y el meloncillo,
Herpestes ichneumon, (L.) (Mammalia, Viverridae) en Doana (SO de la Pennsula Ibrica.
Bol. R. Soc. esp. Hist. Nat. (Secc. Biol.), 87: 257-266.
Palomares F, Delibes M (1993) Key Habitats for Egyptian Mongooses in Doana-National-Park,
South-Western Spain. Journal of Applied Ecology 30 (4):752-758.
Palomares F, Ferreras P, Fedriani JM, Delibes M (1996) Spatial relationships between Iberian lynx and
other carnivores in an area of south-western Spain. Journal of Applied Ecology 33 (1):5-13.
Palomares F, Ferreras P, Travaini A, Delibes M (1998) Co-existence between Iberian lynx and
Egyptian mongooses: estimating interaction strength by structural equation modelling and
testing by an observational study. J Anim Ecol 67 (6):967-978.
Paquet PC, Carbyn LN (1986) Wolves, Canis-Lupus, Killing Denning Black Bears, UrsusAmericanus, in The Riding-Mountain-National-Park Area. Can Field-Nat 100 (3):371-372.
Peterson RO (1995) Wolves as interspecific competitors in canid ecology. Ecology and Conservation
of Wolves (eds L.N. Carbyn, S.H. Frittz & D.R. Seip), pp. 315324. Circumpolar Press,
Edmonton, Alberta.
Polis GA, Myers CA, Holt RD (1989) The Ecology and Evolution of Intraguild Predation - Potential
Competitors that eat each other. Annu Rev Ecol Syst 20:297-330.
Prugh LR, Stoner CJ, Epps CW, Bean WT, Ripple WJ, Laliberte AS, Brashares JS (2009) The Rise
of the Mesopredator. Bioscience 59 (9):779-791.
R Development Core Team (2008). R: A language and environment for statistical computing. R

Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL

http://www.R-project.org.
Ralls K, White PJ (1995) Predation on San-Joaquin Kit Foxes by Larger Canids. J Mammal 76
(3):723-729.
Ritchie EG, Johnson CN (2009) Predator interactions, mesopredator release and biodiversity
conservation. Ecol Lett 12 (9):982-998.
Roemer GW, Gompper ME, Van Valkenburgh B (2009) The Ecological Role of the Mammalian
Mesocarnivore. Bioscience 59 (2):165-173.
Rogers LL, Mech LD (1981) Interactions of Wolves and Black Bears in Northeastern Minnesota. J
155

Chapter 3
Mammal 62 (2):434-436. doi:10.2307/1380735.
Soto C, Desnia S, Palomares F (2012) Nonbiological factors affecting track censuses: implications
for sampling design and reliability.
Skaug H, Fournier D, Nielsen A, Magnusson A and Bolker B (2012). Generalized Linear Mixed

Models using AD Model Builder.

Smith PA (1994) Autocorrelation in logistic regression modelling of species distributions. Global

Ecol. Biodiv. Lett. 4: 4761.

Sovada MA, Roy CG, Bright JB, Gillis JR (1998) Causes and rates of mortality of swift foxes in
western Kansas. Journal of Wildlife Management 62 (4):1300-1306.
Thurber JM, RO Peterson, JD Woolington and JA Vucetich (1992) Coyote coexistence with wolves
on the Kenai Peninsula, Alaska. Canadian Journal of Zoology 70:24942498.
Valverde JA (1958). An ecological sketch of the Coto Doana. British Birds 51:1-23.
VanderMeer J, Ens BJ (1997) Models of interference and their consequences for the spatial
distribution of ideal and free predators. J Anim Ecol 66 (6):846-858.
Van Horne B (1983) Density as a Misleading Indicator of Habitat Quality. Journal of Wildlife
Management 47 (4):893-901.
Viota M, Lpez-Bao JV, Palomares F, Rodrguez A (2012) Shift in microhabitat use as a mechanism
allowing the coexistence of victim and killer carnivore predators.
White PA, RA Garrott (1997) Factors regulating kit fox populations. Canadian Journal of Zoology
75: 19821988.
Yamaguchi N, Rushton S, Macdonald DW (2003) Habitat preferences of feral American mink in the
Upper Thames. J Mammal 84 (4):1356-1373.

156

Chapter 4

CHAPTER 4

Human-related factors regulate dog

presence in protected areas: implications
for conservation and management control
Factores humanos regulan la presencia de perros en las
reas protegidas: implicaciones para la conservacin y
gestin

Soto C, Palomares F (2012) Human-related factors regulate dog presence in

protected areas: implications for conservation and management control. Submitted.
157

158

Chapter 4

ABSTRACT

The presence of domestic species such as dogs Canis familiaris in protected

areas represents a conservation problem due to competition, predation and/or

disease transmission to native species. This introduced species may increase their
ranging activity towards protected areas by the planning of new urban areas and
by the spread of houses and small urban settlements because they are abundant in
those environments. Dogs effects on wildlife in protected areas may depend on
their nature (domestic dogs vs. feral dogs), on where they are found and on the
factors controlling their space use. In order to improve our ability to design effective
control policies, we investigate the factors affecting detection of dog tracks in a
Mediterranean national park which protects the worlds most critically endangered
felid species, the Iberian lynx (Lynx pardinus).

We studied the presence or absence of dogs at 69 2x2 km grids and

analysed the associated environmental and/or human constraints by logistic

regression models. We failed to detect dogs in areas away from anthropogenic
edges (track census effort above 470 km) so according to our results, dogs at DNP
may be domestic dogs incurring occasionally into the protected area from the
surrounding matrix. The detection of dog tracks was dependent on the presence
of people and consequently on the resources they provide. The direct threats of
dogs to wildlife may therefore not spread throughout the entire reserve or along the
entire edge length. Any strategy aimed at reducing domestic dogs impact in areas
of conservation concern where feral dogs populations are not established should
focus on the presence of settlements and their spatial spread, local awareness and
regulation to encourage local people to restrain their dogs movements as well as
control measures on boundaries and areas close to human dwellings.
159

Chapter 4

RESUMEN

La presencia de especies domsticas como los perros Canis familiaris

en las reas protegidas representa un problema para la conservacin debido a

competicin, predacin y/o transmisin de enfermedades a especies nativas. Estas
especies introducidas deben incrementar su actividad en el interior de las reas
protegidas como consecuencia de la planificacin de nuevas reas urbanas as como
de la expansin de zonas urbanizadas, debido a su abundancia en esos ambientes.

Los efectos de los perros en la fauna nativa en las reas protegidas debe

depender de su naturaleza (si se trata de perros domsticos o perros asilvestrados),

de dnde se encuentran y de los factores que determinan su uso del espacio. Para
mejorar nuestra capacidad de diseo de polticas de control efectivas, analizamos
los factores que determinan la deteccin de rastros de perros en un Parque Nacional
que protege la especie de felino ms amenazada a nivel mundial, el lince Ibrico
(Lynx pardinus).

Estudiamos la presencia o ausencia de perros en 69 cuadrculas de 2 x 2

km2 y analizamos los factores ambientales y/o humanos que pudieran determinar
su presencia a travs de modelos de regresin logsticos. No detectamos perros en
zonas alejadas de los lmites antrpicos del rea protegida a pesar del esfuerzo de
muestreo (ms de 470 Km. de caminos censados). Segn nuestros resultados, los
perros presentes en el Parque Nacional de Doana deben ser perros domsticos
que hacen incursiones de manera ocasional en el rea protegida desde la matriz
antrpica circundante. La deteccin de rastros de perros dependi de la presencia
de humanos y consecuentemente de los recursos que proporcionan. Los efectos
directos de los perros en la fauna nativa no deben por tanto mostrar la misma
intensidad en toda la extensin del rea protegida o a lo largo de toda la longitud
160

Chapter 4

de sus lmites. Cualquier plan de manejo que pretenda disminuir la incidencia de

perros domsticos en reas protegidas en las cuales no existan poblaciones de
perros silvestres establecidos, debe focalizarse en la presencia de asentamientos
humanos y su localizacin espacial, en la concienciacin y regulacin local para
fomentar que los propietarios restrinjan los movimientos de sus mascotas, as como
en la localizacin de las medidas de control en los lmites de las reas protegidas
prximos a las viviendas o zonas habitadas.

161

Chapter 4

INTRODUCTION

The introduction of non-native species into ecosystems represents one of

the causes of native species decline and endangerment (e.g., King 1985, Soul
1990, Williamson 1999, IUCN 2012). Due to human population growth, urban
sprawl and the rapid urbanization of natural landscapes, humans and with them
their companion animals such as domestic dogs Canis familiaris may be closely in
contact with wildlife in many areas (Ordeana et al. 2010) including places where
nature conservation is a priority, such as protected areas and national parks created
to protect populations of vulnerable or threatened species. These non-native species
introduced by human across the globe (Wandeler et al. 1993) may therefore increase
their ranging activity towards the remaining natural landscapes extending within
those areas the deleterious human-associated effects.

In areas of conservation concern presence of companion animals like domestic

dogs may pose distinct threats including competition, interbreeding, predation and
disease. Dogs harass and kill wildlife exhibiting in many cases a surplus killing
behaviour (e.g., Iverson 1978, Kruuk & Snell 1981, Manor & Saltz 2004, Banks
& Bryan 2007), compete with wildlife (e.g., Boitani 1983, Butler & du Toit 2002,
Butler et al. 2004, Vanak et al. 2009) and spread disease like rabies, parvovirus or
canine distemper (e.g., Sillero-Zubiri et al. 1996, Cleaveland et al. 2000, Fiorello
et al. 2004, Fiorello et al. 2006, Vanak & Gompper 2009). Moreover, dogs such
as mid-sized canids, can also exert a top-down influence on smaller carnivores
through interference competition or intraguild predation (e.g., Glen & Dickman
2005, Mitchell & Banks 2005, Vanak & Gompper 2009). Studying where domestic
dogs came from in protected areas is needed in order to manage them and prevent
their potential impacts on native fauna. Dogs at protected areas may exhibit varying
162

Chapter 4

levels of human reliance, from domestic dogs whose needs are satisfied directly
or indirectly by people (i.e., owned dogs, urban or rural free-roaming dogs and/
or village dogs (Vanak & Gompper 2009)) and spend time sporadically in natural
protected areas, to feral dogs living and reproducing freely in protected areas. The
effects of dogs on wildlife may therefore depend on their nature (i.e., domestic dogs
vs. feral dogs), on where they are found and on the factors controlling their numbers
and space use.

As a heavily human-subsidized species, domestic dogs exhibit higher

densities in areas close to human residences or places with a high density of houses
(e.g., Odell & Knight 2001, Ordeana et al. 2010) as well as in areas near natural
reserve borders with agriculture, where rural human residences are nearby. In
contrast, they may exhibit lower densities in areas contiguous to large tracts of
native forest, which may be acting as a buffer to the entrance of dogs (Srbek-Araujo
& Chiarello 2008).

Domestic dogs records decrease from anthropogenic matrices to forest

patch edges (Torres & Prado 2010). Hence, if we consider protected areas as large
patches, presence and therefore direct negative effects of domestic dogs on native
fauna (i.e., predation and/or competition) may decrease from the anthropogenic
matrix to the protected area interior reaching their maximum at the borders. This
could strengthen the negative anthropogenic edge effect associated with these
artificial border areas (Woodroffe & Ginsberg 1998, Revilla et al. 2001) diminishing
therefore the reserves effectiveness in conserving wildlife.
Feral dogs are meanwhile completely wild and independent of human-derived
materials as food sources (Nesbitt 1975, Green & Gipson 1994). They depend
almost exclusively on wild-caught food (e.g., Marsack & Greg 1990, Glen &
163

Chapter 4

Dickman 2005, Mitchell & Banks 2005, Campos et al. 2007, Glen & Dickman
2008) and might not exhibit any human association. The direct threats of feral dogs
to wildlife may therefore spread throughout entire protected areas.

In this context we studied the patterns of detection of dog tracks and the

associated environmental and/or human constraints that could influence their

presence in a fully protected Mediterranean area, Doana National Park (DNP),
with high potential for dog-arrival due to its proximity to urban and rural settlements
and with the potential of dog-settlement due to its size.

Our main research goals were to answer: (1) where are dogs present at

DNP? And (2) what factors predict dog presence? A priori, we hypothesised that
dogs using DNP might come either from a domestic dog population formed by
individuals that incur occasionally from the surrounding matrix, or from a feral
dog population living and reproducing freely. In the first case, we would expect
that dogs are heavily dependent on humans and are more abundant at the edges
of the protected area close to human settlements than far away from these edges.
In the second case, we would expect that dogs avoid human association, are more
evenly distributed throughout the protected area, and their presence or abundance
related to environmental features describing habitat suitability or potential wild
food availability. DNP is optimal for the design of a study of this type since 1) part
of the protected area is surrounded by human settlements and others are not, and 2)
it is large enough to potentially hold a feral dog population in its interior.

164

Chapter 4

METHODS
Study area

The study was carried out at Doana National Park (DNP), a flat sandy area

located in south-western Spain (550 Km2, 379N, 626W) at sea level. We defined
the anthropogenic DNP edge as the northern and western edges in close contact
with human settlements, crop fields and a high-traffic highway, and the natural DNP
edge as the southern edge limiting with the Atlantic Ocean and the eastern edge
limiting with the Guadalquivir River through a 27,000 ha marshland area (Fig. 1).

Population in the main suburban settlement (situated in the western vicinity

and separated from the DNP by a paved road)undergoes greatvariationbetween

winterand summer, as it ismainlya summer resort occupied by about 2,710 people
in summer seasons. The village situated in the north, and separated by the DNP
from the marshland area, is occupied by about 1,635year-round residents, although
duringa spring pilgrimage, thenumberof visitors canreach up toone million
people. There are also private large and medium-sized farms used for agriculture
as well as six visitor centers, hiking and cycling paths, recreation zones and bird
observatories in the nearby area. The DNP is fenced but the fence is permeable to
small and medium-sized animals including dogs. Free access is forbidden to people
and access to the core area and the dirt-road network inside the DNP is restricted to
the park staff and researchers.

The climate is Mediterranean sub-humid, with mild, wet winters, and hot,

dry summers, and an average annual rainfall around 550 mm. There are three main
biotopes in the park: scrubland, dunes, and marsh (Valverde 1958). The dune area
is situated at the western border of the protected area limited by the Atlantic Ocean
and the marsh area at the northern and eastern borders limited by the Guadalquivir
165

Chapter 4

Figure 1. Study area defined by the Doana National Park site and the surrounding anthropogenic
area. Dog tracks detected during track counts in 22 km2 cell grids between November 2007 and
April 2009 are shown below in detail.
166

Chapter 4

River. The Mediterranean scrubland represents approximately half of the National

Park surface area and is mainly characterised by heterogeneous patches of
xerophytic species such as Halimium sp. and Cistus sp., and hydrophytic ones such
as Erica sp., with some patches of Juniperus phoenica and Pistacia lentiscus shrubs.
Interspersed among the scrubland there are scattered cork oak trees (Quercus suber)
and wild olive trees (Olea europea), and a few patches of pine Pinus pinea and
eucalyptus Eucalyptus sp. plantations.

Among larger mammals wild boar (Sus scrofa), red deer (Cervus elaphus)

and fallow deer (Dama dama) are frequent. Wild carnivores include red fox (Vulpes
vulpes), Eurasian badger (Meles meles), Egyptian mongoose (Herpestes ichneumon),
common genet (Genetta genetta), least weasel (Mustela nivalis), European polecat
(Mustela putorius), Eurasian otter (Lutra lutra), wild cat (Felis silvestris) and
Iberian lynx (Lynx pardinus). 14 small-medium sized mammal species have been
recorded in DNP, as well as 397 bird species, approximately half of which are
breeding in the park.

Field methods

To evaluate detection of dog tracks we carried out systematic track surveys

on sandy paths at 69 2x2 km2 grid cells located within the entire scrubland and dune
areas of the protected area during the wet season of 2007-08 and 2008-09.

We sampled for dog tracks in each square by slowly walking (ca. 1.5 km/h)

at least 3 km along available pathways (i.e., sandy roads and firebreaks). Once a
track was detected, we georeferenced it using a GPS. We re-sampled the same path
(leaving at least seven days between samplings) a second time in a few squares until
completing 3 km if during the first sampling there were insufficient available paths
167

Chapter 4

within the square to achieve this distance. We always carried out surveys at least
three days after any rainfall.

Environmental quality of the sampling grids to hold a feral dog population

was assessed by sampling potential prey availability and general habitat structure.
Feral dogs are habitat generalists and opportunistic foragers depending almost
exclusively on a wide variety of wild-caught food (e.g., Boitani et al. 1995, Marsack
& Greg 1990). Potential prey availability was estimated by counting tracks of small
mammals, European rabbits (Oryctolagus cuniculus), red partridges (Alectoris rufa),
domestic cows (Bos taurus) and horses (Equus caballus) and wild ungulates such
as the fallow deer (Dama dama), the red deer (Cervus elaphus) and the wild boar
(Sus scrofa). Prey such as small mammals, rabbits, partridges and young of wild
ungulates might be hunted by feral dogs, but adults of many species are consumed
as carrion (e.g., Sillero-Zubiri & Macdonald 1997, Butler et al. 2004, Aiyadurai &
Jhala 2006). Prey species were surveyed by walking between 7 and 10 25 m-long
transects of approximately 1.7 m wide (the width of a four-wheel-drive car) and
separated by at least 300 m within each 2x2 km grid. During the first year, prey
transects were carried out throughout the wet season, when tracks from dogs were
surveyed, but during the second year, we concentrated samplings within the month
of April to avoid possible intermonthly variations in abundance for some species
(e.g., see Kufner 1986, Palomares et al. 2001 for small mammals and European
rabbits, respectively.

In order to identify main habitats at DNP (dunes, pine reforestation and

Mediterranean scrubland), general habitat structure was recorded for the first year.
We estimated visually the percentage of open ground cover, and the percentage
and modal height of three vegetation categories: short shrub (xerophytic species
168

Chapter 4

such as Halimium sp. and Cistus sp), tall shrub (Erica sp., Juniperus phoenica and
Pistacia lentiscus shrubs) and trees. The estimation of these variables related to
habitat structure were carried out in a circle of 15 m radius around the sampling
point every 300 m on transects walked for dog and prey tracks. For each square
sampled, we averaged the value obtained at the vegetation sampling points.

Data analyses

We examined different environmental and/or human-related factors

explaining the detection of dog tracks within each 2x2 km grid at DNP using
generalised linear models with a binomial error distribution and a logit link function
(logistic procedure in SAS 9.2 (SAS Inst. Inc., Cary, NC)). Non-biological factors
(i.e. methodological and climatic variables) have been previously reported as
potentially affecting results of track censuses (Soto et al. 2012); therefore we also
incorporated in the model fitting these factors to control for their potential effect.

Each grid cell was associated with a set of habitat variables as vegetation

type (dunes (> 60% of open groundcover on average inside the grid), pine forest (>
60% of pine vegetation on average) and Mediterranean shrub (> 60% of shrub (short
or tall) vegetation on average) and prey abundance (kilometric abundance index
of total prey) and with variables describing their location in DNP by calculating
Euclidean distance from the grid cell centre to every infrastructure.

We used a two-step approach to analyze data. First, we assessed which

methodological and/or climatic variables potentially affect dog detectability and

selected the best-fitting model using an information-theoretic approach (Burnharm
& Anderson 2002). Variables considered to be included in models were the
observer who carried out censuses (three and two observers for both study years
169

Chapter 4

respectively) (Obs), relative humidity (%) on census day (Hum), days since last rain
(Rain), year (Year) and the maximum temperature (C) calculated as the average of
the maximum temperature on the census day and the maximum temperature two
consecutive days before the census day (Temp). Climatic data was obtained from a
meteorological station located inside DNP (Latitude: 37 118, Longitude: 6 33
17) http://icts.ebd.csic.es.

Secondly, we used this best-fitting model as a null model to develop a set

of a priori models of dog tracks detectability at DNP based on three groups of

hypotheses stemming from the different variables considered in relation to 1) the
possible human dependence of dog tracks detectability (i.e., dogs being domestic),
2) the possibility of dogs coming from a feral population (i.e., dog tracks
detectability related to environmental and/or prey variables), or 3) dogs coming
from a combination of both domestic and feral populations. Variables included
in models were the minimum distance to nearest single house or visitors centre
(D_HOU), minimum distance to human settlements (D_VIL), distance to nearest
paved road (D_RD), distance to anthropogenic edge of DNP (D_ANT), distance to
natural edge of DNP (D_NAT), Kilometre Abundance Index of total prey (Pt) and
the vegetation category; dunes, pine forest, scrubland (Veg).

Initially, the correlation among variables was explored using Kendalls tau

statistics, in order to eliminate highly correlated variables (tau > 0.4) and among
them; we retained the more ecologically meaningful ones.

We used the Akaike Information Criterion (AIC) corrected for a small sample

size (AICc) and the difference in AICc between each model and the model with the
lowest AICc (AICc) to rank the models according to their capacity to describe
the data parsimoniously (Burnharm & Anderson 2002). The model with the lowest
170

Chapter 4

AICc and those with AICc 2 were considered to be supported. AICc values
were used to compute Akaikes weights (i), which is the weight of evidence that a
model is the best approximating model given the model set (Burnham & Anderson
2002) and is defined as
i =exp(-1/2i)/R r=1exp(-1/2r).

In addition, the relative variable importance of predictor variable j (j) was

determined as the sum of the i across all models where j occurs. Larger j values
indicate a higher relative importance of variable j compared to other variables.
For each hypothesis we used data from both years and we began by fitting all
variables included and then successively removing the terms that decreased the
AIC the most (Crawley 2002).

Finally, we explored the classification accuracy of the selected model(s)

using the nonparametric estimate of the area under the curve (AUC) of receiveroperating characteristic plots (Hosmer & Lemeshow 2000). AUC indices range
from 0.5 to 1, with ranges from 0.5 to 0.7 indicating poor discrimination, from 0.7
to 0.8 acceptable discrimination, from 0.8 to 0.9 good discrimination, and > 0.9
outstanding discrimination. The AUC measure from the ROC curve is considered
useful for comparing the performance of the detection of dog tracks-absence model
in a threshold-independent fashion (Fielding & Bell 1997).

171

Chapter 4

RESULTS

A total of 471 km was walked and 72 dog tracks were found during surveys

(Fig. 1). We detected dog tracks at 16 and 12 grid cells surveyed in each study year,
respectively. Four anthropogenic variables were strongly correlated. In particular,
distance to the anthropogenic edge of the National Park (D_ANT), distance to
nearest village (D_VIL), distance to nearest paved road (D_ROAD) and distance
to the natural edge of DNP (D_NAT). Analyses were focused on distance to the
anthropogenic edge of DNP and distance to the natural edge of DNP.

The best-fitting model explaining detection of dog tracks based on non-

biological factors included humidity (Hum) and days since last rain (Rain). Humidity
was positive but non-significant (odds ratio = 1.035, 2 = 2.159, P=0.142) whereas
days since last rain was negatively and significantly correlated with detection of
dog tracks (odds ratio = 0.931, 2 = 4.286, P=0.035). Both predictors were therefore
included as covariables in further analyses.

The analysis of dog tracks detectability based on human-related, habitat

and prey variables showed that the a priori hypothesis best adjusted to data only
included human-related predictors.

The best model describing the detection of dog tracks at DNP after adjusting

for detection probability variables in the null model included the distance to the
anthropogenic edge of DNP (explaining 27.6 % of the deviance); the next model
included the distance to the anthropogenic edge of DNP and the distance to the
natural edge of DNP (models 1 and 3, Table 1).

Detection of dogs was significantly and negatively associated to the distance

to the anthropogenic edge of DNP (odds ratio = 0.737, 2 = 8.020, P=0.005)

(Fig. 2). Equation for this model (model 1, Table1) is:
172

Chapter 4

logit(P) = -3.32(1.86) - 0.31(0.11)D_ANT + 0.05(0.02)Hum -

0.06(0.04)Rain
where P is the probability of dog occurrence; values within parentheses are standard
errors.

The relative variable importance of this anthropogenic variable determined

as the sum of the i across all models where the variable occurred was j=0.999.
The discriminating ability of the top model was AUC = 0.802 (P < 0.0001).

DISCUSSION

Results show that the detection of dog tracks at DNP was associated with

distance from the park boundary with human presence, a synthetic indicator of
Table 1. Selection results from logistic regression models investigating the effects of anthropogenic,
habitat and a combination of all variables on detection of dog tracks at DNP. For the top models, we
report the small sample-size-adjusted Akaikes information criteria (AICc), the difference in AICc
between each model with the lowest AICc (AICc) and the AICc weight (wi).

Model code

Deviance
139.210

AICc

121.107

AICc

0. Null model

wi

12.939

0.000

1. D_ANT, Hum, Rain

100.170

108.168

0.000

0.276

2. D_NAT, D_ANT, D_HOU, Hum, Rain

97.470

109.469

1.301

0.144

3. D_NAT, D_ANT, Hum, Rain

98.240

108.242

0.074

0.266

4. D_ANT, D_HOU, Hum, Rain

99.750

109.747

1.579

0.125

5. Pt, Hum, Rain

112.970

120.971

12.803

0.000

6. Veg, Hum, Rain

106.720

122.265

14.097

0.000

7. Pt, Veg, Hum, Rain

110.900

122.897

14.729

0.000

8. D_NAT, D_ANT, D_HOU, Pt, Hum, Rain

97.950

111.953

3.785

0.042

9. D_NAT, D_ANT, Pt, Hum, Rain

98.850

110.853

2.685

0.072

10. D_ANT, Pt, Hum, Rain

100.810

110.811

2.643

0.074

Anthropogenic

Habitat

Global

173

Chapter 4

human influence that captures the effect of distance to nearest village and nearest
paved road.

We found a high number of dog records near the reserves borders where

rural and suburban households were closer and we were unable to detect signs of
dogs far away from these anthropogenic DNP edges in spite of our large census
effort. Additionally, dog tracks detectability did not seem to be related to the
environmental variability of DNP such as vegetation type or prey availability.

This lack of association between dogs and variables describing habitat

suitability or potential wild food availability, their dependence on human-related

variables and their higher abundance at the edges of the protected area close to
human settlements compared to areas far away from these edges, support our

Figure 2. Probability of domestic dog tracks detectability as a function of distance to the

anthropogenic edge of DNP during the wet seasons of 2007-08 and 2008-09.
174

Chapter 4

hypothesis that dogs using DNP come from a domestic dog population formed by
individuals that arrive sporadically from the surrounding matrix (i.e., owned dogs,
urban/rural free-roaming dogs or village dogs (Vanak & Gompper 2009)) and not
from a feral dog population living and reproducing freely. The lack of association
between detection of dog tracks and wildlife food resources possibly also reveals
the dependence of dogs on human-derived materials, which is typical for the vast
majority of dog populations for which diet has been studied (Atickem 2003, Butler
et al. 2004, Vanak 2008, Vanak & Gompper 2009).

Feral dogs survive and reproduce independently of human assistance but

some feral dogs use human garbage for food (Green & Gipson 1994). A feral dog
population established inside DNP could therefore use the edge of the protected
area to access human subsidies. Nevertheless, the primary feature that distinguishes
feral from domestic dogs is the degree of reliance on humans, so if dogs using
DNP come from a feral dog population living and reproducing freely but accessing
human subsidies for food, we would expect dog detectability to be dependent
on habitat suitability and/or wild food availability and marginally dependent on
human-related variables.

In contrast with domestic dogs, feral dogs are highly social living in packs

or groups year round in most cases (Daniels & Bekoff 1989, Green & Gipson 1994).
In our study area, we only detected isolated dogs tracks. Additionally, cameratrapping studies conducted inside DNP during the same period detected only six
dogs, all of which were found near human settlements and identified as domestic
animals based on their external physical appearance (personal observ.).

Additionally, the occurrence of dog tracks restricted to the DNP edge and

the low number of tracks detected far away from these reserves edges supports
175

Chapter 4

the idea that the presence of domestic dogs at Doana may be exacerbating the
anthropogenic edge effect associated with its border areas.
Some previous authors have also reported a higher occurrence of domestic dogs
near the edges of natural reserves compared to their interiors (Butler & du Toit
2004, Srbek-Araujo & Chiarello 2008, Lacerda et al. 2009, Marks & Duncan 2009).
In this sense, domestic dogs could be generally considered as human-derived
edge effect at protected areas. The direct threats of domestic dogs to wildlife may
therefore not spread throughout the entire protected area or along the entire edge
length reaching its maximum near the anthropogenic border areas.

Edge effects can be important in the dynamics of populations living in

fragmented landscapes because they may affect key population parameters, such
as survival and reproduction (Murcia 1995, Noss & Csuti 1997). The peripheries
of reserves thus function as population sinks (Revilla et al. 2001) and the resulting
edge effect can cause the decline or the extinction of protected carnivore populations
(Woodroffe & Ginsberg 1998). The higher occurrence of dogs in these border areas
due to human presence in the surrounding matrix might exacerbate this effect.

Nevertheless, although domestic dogs rarely leave the vicinity of human

dwellings (Vanak & Gompper 2009, Butler & du Toit 2002) and appear to exhibit
a range mainly limited to reserves borders, their daily activity pattern may involve
free-ranging that can bring them into contact with wildlife, especially when their
movements are not confined to a proscribed outdoor area (Butler et al. 2004, Vanak
2008).

Diseases from dogs that affect wild species (e.g., Woodroffe & Ginsberg

1999, Cleaveland et al. 2000) could be transmitted across the border of reserves
worsen therefore the direct edge effects represented by domestic dogs in protected
176

Chapter 4

areas (i.e., the eventual predation and displacement of some species near the park
border). In consequence, domestic dogs maintaining highly virulent, multi-host
pathogens can induce mortality to wild carnivores owing to interspecific disease
transmission between susceptible wild carnivores in the community even when the
wildlife population of interest may never have contact with them.

DNP houses one of the last metapopulation of the most globally endangered

felid species, the Iberian lynx (Lynx pardinus) (Palomares et al. 2011) so dogs
living or spending time inside the area come into contact with this wild endangered
carnivore posing a serious risk for its conservation (Ferreras et al. 1992, Meli et al.
2009 and 2010, Milln et al. 2009a and 2009b).

Domestic dogs are present in large numbers in urban, suburban and rural

areas, so, due to their high numbers, they can have a substantial impact on wildlife,
even when they do not need to hunt to survive.

This situation forms a complicated scenario for conservation biologists

especially in conservation areas with endangered endemic carnivore populations

and/or those where reserve size is too small in relation to the scale of the species
movement. There is a huge demand for more knowledge about and experience with
this type of situation, in order to help prevent or diminish the impacts on native
fauna in natural reserves.

Conclusions and management implications

Domestic dogs in DNP and in the surrounding matrix can considerably

diminish the reserves effectiveness in conserving wildlife. The rapid urbanization

process close to conservation units and the growth of domestic dog populations is
an increasing worldwide conservation. Thus, transborder conservation measures
177

Chapter 4

must be implemented in the protected areas.

In the specific case of DNP, controlling the dog populations is urgent and key

for the wildlife protection in this National Park. The detection of dog tracks at DNP
was dependent on the presence of people and consequently on the resources they
provide such that the potential direct effects of dogs on wildlife may be stronger on
these anthropogenic boundaries. Management of domestic dogs in protected areas
where feral dogs populations are not established may therefore be focused at borders
and neighbourhoods close to human dwellings. We suggest that control measures
must include the restriction of domestic dogs free-roaming activity through local
public awareness focusing on responsible ownership and biodiversity conservation,
the removal of un-owned dogs through systematic campaigns on reserve boundaries
near human settlements, as well as strengthening of pet policies.

ACKNOWLEDGMENTS
This study was financed by the projects CGL2004-00346/BOS (Spanish Ministry of Education and
Science) and 17/2005 (Spanish Ministry of the Environment; National Parks Research Programme),
and sponsored by Land-Rover Espaa S.A. CS received a JAE predoctoral grant from CSIC
(Spanish National Research Council). We are especially thankful to J.C. Rivilla and S. Desnia for
their assistance during fieldwork, to N. Fernndez for his useful inputs for analysis of the data and
valuable suggestions, and to C. Dickman and P. Ferreras for their comments on earlier versions of
the manuscript.

REFERENCES
Acevedo-Whitehouse, K. (2009). The importance of disease management programmes for wildlife
conservation. Anim. Conserv. 12, 185186.
Aguirre, A.A., Starkey, E.E., Hansen, D.E. (1995). Wildlife diseases in national park ecosystems.

178

Wildlife Soc. B. 23, 415419.

Chapter 4
Aiyadurai, A., Jhala, Y.V. (2006). Foraging and habitat use by golden jackals (Canis aureus) in the

Bhal Region, Gujarat, India. J. Bombay Nat. Hist.Soc. 103, 512.

Atickem, A. (2003). The ecology of domestic dog (Canis familiaris) and its effect on the survival

of the Ethopian wolf (Canis simensis) in the Web Valley of the Bale Mountains National

Park, Ethopia. MS thesis, Addis Ababa University, Addis Ababa, Ethiopia.

Banks, P., Bryant, J. (2007). Four-Legged Friend or Foe? Dog Walking Displaces Native Birds.Biol.
Lett. 3, 611613.
Boitani, L. (1983). Wolf and dog competition in Italy. Acta Zool. Fennica. 174, 259264.
Boitani, L., Francisci, F., Ciucci, P., Andreoli, G. (1995). Population biology and ecology of feral

dogs in central Italy. In The Domestic Dog. Its Evolution, Behaviour, and Interactions

with People: 217244. Serpell, J. (Ed.). Cambridge, UK: Cambridge University Press.

Burnharm, K.P., Anderson, D.R. (2002). Model selection and multimodel inference. A practical

information-theoretic approach. 2nd ed. New York: Springer.

Butler, J.R.A., du Toit, J.T. (2002). Diet of free-ranging Domestic Dogs (Canis familiaris) in rural

Zimbabwe: Implications for wild scavengers on the periphery of wildlife reserves.

Anim. Conserv. 5, 2937.

Butler, J.R.A., du Toit, J.T., Bingham, J. (2004). Free-ranging domestic dogs (Canis familiaris) as

predators and prey in rural Zimbabwe: threats of competition and disease to large wild

carnivores. Biol. Conserv. 115, 369378.

Campos, C.B., Esteves, C.F., Ferraz, K., Crawshaw, P.G., Verdade, L.M. (2007). Diet of free-ranging

cats and dogs in a suburban and rural environment, south-eastern Brazil. J. Zool. 273, 1420.

Clark, T.W., Curlee, A.P., Reading, R.P. (1996). Crafting effective solutions to the large carnivore

conservation problem. Conserv. Biol. 10, 940948.

Cleaveland, S., Appel, M.G.J., Chalmers, W.S.K., Chillingworth, C., Kaare, M., Dye, C. (2000).

Serological and demographic evidence for domestic dogs as a source of canine distemper

virus infection for Serengeti wildlife. Vet. Microbiol. 72, 217227.

Crawley, M.J. (2002). Statistical computing: An introduction to data analysis using S-Plus.

Chichester, West Sussex, UK, Wiley.

Crooks, K.R. (2002). Relative sensitivities of mammalian carnivores to habitat fragmentation.

Conserv. Biol. 16, 488502.

179

Chapter 4
Ferreira, J.P., Leita, I., Santos-Reis, M., Revilla, E. (2011). Human-related factors regulate the

spatial ecology of domestic cats in sensitive areas for conservation. PlosOne 6, e25970.

Ferreras, P., Aldama, J.J, Beltrn, J.F., Delibes, M. (1992). Rates and causes of mortality in a

fragmented population of lberian lynx Felis pardina Temminck, 1824. Biol. Conserv. 61,197202.

Fielding, A.H., Bell, J.F. (1997). A review of methods for the assessment of prediction errors in

conservation presence/absence models. Environ. Conserv. 24, 3849.

Fiorello, C.V., Noss, A.J., Deem, S.L. (2006). Demography, hunting ecology, and pathogen exposure

of domestic dogs in the Isoso of Bolivia. Conserv. Biol. 20, 762771.

Glen, A.S., Dickman, C.R. (2005). Complex interactions among mammalian carnivores in Australia,

and their implications for wildlife management. Biol. Rev. 80, 387401.

Glen, A.S., Dickman, C.R. (2008). Niche overlap between marsupial and eutherian carnivores: does

competition threaten the endangered spotted-tailed quoll?. J.Appl. Ecol. 45, 700707.

Green, J.S., Gipson, P.S. (1994). Feral dogs. In The Handbook: Prevention and Control of Wildlife
Damage :17. Hygnstrom, S. E., Timm R. M and Larson G. E. (Ed.). University of

Nebraska, Lincoln, NE, USA.

IUCN 2012.TheIUCN Red List of Threatened Species. Version 2012.1.

<http://www.iucnredlist.org>. Downloaded on 25 June 2012.
Iverson, J.B. (1978). The impact of feral cats and dogs on populations of theWest Indian rock iguana,

Cyclura carinata. Biol. Conserv. 14, 6373.

King, W.B. (1985). Island birds: will the future repeat the past?. In Conservation of island birds:

315. Moors, P. J. (Ed.). ICBP Technical Publication Number 33.

Kruuk, H., Snell, H. (1981). Prey selection by feral dogs from a population of marine iguanas

Amblyrhynchus cristatus. J. Appl. Ecol. 18, 197204.

Kufner, M.B. (1986). Tamao actividad, densidad relativa y preferencias de hbitat de los pequeos

y medianos mamferos de Doana, como factores condicionantes de su tasa de predacin.

Thesis, Universidad Autnoma, Madrid, Spain.

Lacerda, A.C.R., Tomas, W.M., Marinho-Filho, J. (2009). Domestic dogs as an edge effect in the

Braslia National Park, Brazil: interactions with native mammals. Anim. Conserv. 12, 477487.

Manor, R., Saltz, D. (2004). The impact of free-ranging dogs on gazelle kid/female ratio in a

180

fragmented area. Biol. Conserv. 119, 231236.

Chapter 4
Marks, B.K., Duncan, R.S. (2009). Use of Forest Edges by Free-ranging Cats and Dogs in an

Urban Forest Fragment. South. Nat. 8, 427436.

Marsack, P., Greg, C. (1990). Feeding behavior and diet of dingoes in the Nullarbor region, Western

Australia. Aust. Wildlife Res. 17, 349357.

McCallum, H. (1996). Immunocontraception for wildlife population control. Trends Ecol. Evol. 11,
491493.
Meli, M.L., Cattoril, V., Martnez, F., Lpez, G., Vargas, A., Simn, M.A., Zorrilla, I., Muoz A.,

Palomares F., Lpez-Bao, J.V., Pastor, J., Tandom, R., Willi, B., Hofmann-Lehmann R.,

Lutz, H. (2009). Feline leukaemia virus and other pathogens as important threats to the

survival of the critically endangered Iberian Lynx (Lynx pardinus). PLoS ONE. 4, e4744.

Meli, M.L., Cattori, V., Martnez, F., Lpez, G., Vargas, A., Palomares, F., Lpez-Bao, J.V., Hofmann

Lehmann, R., Lutz, H. (2010). Feline leukemia virus infection: a threat for the survival of the

critically endangered Iberian lynx (Lynx pardinus). Vet. Immunol.Immunop. 134, 6167.

Milln, J., Candela, M.G., Palomares, F., Cubero, M.J., Rodrguez, A., Barral, M., de la Fuente,

J., Almera, S., Len-Vizcano, L (2009a). Disease threats to the endangered Iberian lynx

(Lynx pardinus). Vet. J. 182, 114124.

Milln, J., Cabezn, O., Pabn, M., Dubey, J.P., Almera, S. (2009b). Seroprevalence of Toxoplasma

gondii and Neospora caninum in feral cats (Felis silvestris catus) in Majorca, Balearic

Islands, Spain. Vet. Parasitol. 165, 323326.

Mitchell, B.D., Banks, P.B. (2005). Do wild dogs exclude foxes? Evidence for competition from

dietary and spatial overlaps. Austral Ecol. 30, 581591.

Murcia, C. (1995). Edge effects in fragmented forests: implications for conservation. Trends Ecol.
Evol. 10, 5862.
Nesbitt, W.H. (1975). Ecology of a feral dog pack on a wildlife refuge. In The Wild Canids : 391

395. Fox, M. W. (Ed.). New York, USA: Van Nostrand Reinhold Company.

Neronov, V.M. Khlyap, L.A. Bobrov, V.V., Warshavsky, A.A. (2008). Alien species of mammals

and their impact on natural ecosystems in the biosphere reserves of Russia. Integrative

Zool. 3, 8394.
Noss, R.F., Csuti B. (1997). Habitat fragmentation. In Principles of conservation biology: 269

304. Meffe, G. K. and Carroll, C. R. (Eds.). Sunderland, Massachusetts: Sinauer

181

Chapter 4

Associates, Inc., Publishers.

Nowell, K., Jackson, P. (1996). Wild Cats: Status Survey and Conservation Action Plan. IUCN/

SSC, Gland, Switzerland.

Odell, E.A., Knight, R.L. (2001). Songbird and medium-sized mammal communities associated

with exurban development in Pitkin County, Colorado. Conserv. Biol. 15, 11431150.

Ordeana, M.A., Crooks, K.R., Boydston, E.E., Fisher, R.N., Lyren, L.M., Siudyla, S., Haas,

C.D., Harris, S., Hathaway, S.A., Turschak, G.M., Miles, A.K., Van Vuren, D.H. (2010).

Effects of urbanization on carnivore species distribution and richness. J. Mammal. 91,

13221331.
Palomares, F., Delibes, M., Revilla E., Calzada, J., Fedriani, J.M. (2001). Spatial ecology of Iberian

lynx and abundance of European rabbits in Southwestern Spain. Wildlife Monogr.148,136.

Reed, S.E., Merenlender, A.M. (2011). Effects of Management of Domestic Dogs and Recreation on

Carnivores in Protected Areas in Northern California. Conserv. Biol. 25, 504513.

Revilla, E., Palomares, F., Delibes, M. (2001). Edge-core effects and the effectiveness of traditional

reserves in conservation: Eurasian Badgers in Doana National Park. Conserv. Biol. 15,

148158.
Ryser-Degiorgis, M.P., Hofmann-Lehmann, R., Leutenegger, C.M., af Segerstad, C.H., Morner, T.,

Mattsson, R., Lutz, H. (2005). Epizootiologic investigations of selected infectious disease

agents in freeranging Eurasian lynx from Sweden. J. Wildlife Dis. 41, 5866.

Schonewald-Cox, C., Azari, R., Blume, S. (1991). Scale, variable density and conservation planning

for mammalian carnivores. Conserv. Biol. 5, 491495.

Sillero-Zubiri, C., King, A.A., MacDonald, D.W. (1996). Rabies and mortality in Ethiopian wolves

(Canis simensis). J. Wildlife Dis. 32, 8086.

Sillero-Zubiri, C., Macdonald, D.W. (1997). The Ethiopian Wolf: Status Survey and Conservation Action
Plan. IUCN Canid Specialist Group, Gland, Switzerland, and Cambridge, UK.
Sleeman, J.M., Keane, J.M., Johnson, J.S., Brown, R.J., Woude, S.V. (2001). Feline leukemia virus

in a captive bobcat. J. Wildlife Dis. 37, 194200.

Soto, C.A., Desnica, S., Palomares, F. (2010). Non-biological factors affecting track censuses;

implications for sampling design and reliability. Eur. J. Wildlife Res. 58, 117126.

Soul, M.E. (1990). The onslaught of alien species, and other challenges in the coming decades.
182

Conserv. Biol. 4, 233239.

Srbek-Araujo, A.C., Chiarello, A.G. (2008). Domestic dogs in Atlantic forest preserves of south

eastern Brazil: a camera-trapping study on patterns of entrance and site occupancy rates.

Braz. J. Biol. 68, 771779.

Torres, P.C., Prado, P.I. (2010). Domestic dogs in a fragmented landscape in the Brazilian Atlantic

Forest: abundance, habitat use and caring by owners. Braz. J. Biol. 70, 987994.

Tyndale-Biscoe, C.H. (1994). Virus-vectored immunocontraception of feral mammals. Reprod.

Fert. Develop. 6, 281287.

Valverde, J. A. (1958). An ecological sketch of the Coto Doana. British Birds. 51, 1 23.
Vanak, A.T. (2008). Intraguild interactions between native and domestic carnivores in central India.

PhD Dissertation, University of Missouri, Columbia, MO, USA.

Vanak, A.T., Gompper, M. E. (2009a). Dogs Canis familiaris as carnivores: their role and function

in intraguild competition. Mammal Rev. 39, 265283.

Vanak, A.T., Gompper, M.E. (2009b). Dietary niche separation between sympatric free-ranging

domestic dogs and Indian foxes in central India. J. Mammal. 90, 10581065.

Vanak, A.T., Thaker, M., Gompper, M.E. (2009). Experimental examination of behavioural interactions

between free-ranging wild and domestic canids. Behav. Ecol. Sociobiol. 64, 279287.

Wandeler, A.I., Matter, H.C., Kappeler, A., Budde, A. (1993). The Ecology of Dogs and Canine

Rabies: A Selective Review. Rev. Sci. Tech. O.I.E. 12, 5171.

Whiteman, C.W., Matushima, E.R., Confalonieri, U.E.C., Palha, M.D.C., Silva, A.S.L., Monteiro,

V.C. (2007). Human and domestic animal populations as a potential threat to wild carnivore

conservation in a fragmented landscape from the Eastern Brazilian Amazon. Biol. Conserv.

138, 290296.

Williamson, M. (1999). Invasions. Ecography. 22, 512.

Woodroffe, R., Ginsberg, J.R. (1998). Edge effects and the extinction of populations inside protected
areas.
Science.280, 21262128.
Woodroffe, R., Ginsberg, J.R. (1999). Conserving the African wild dog Lycaon pictus. I.Diagnosing

and treating causes of decline. Oryx, 33,132-142.

183

Chapter 4

ANNEX II

Domestic dogs sighted within Doana National Park and photo-trapped during the study period.

184

CHAPTER 5

Surprising low abundance of European

wildcats in a protected area of southwestern
Spain
Abundancia sorprendentemente baja de gato monts en un
rea protegida del suroeste de Espaa

185

186

Chapter 5

ABSTRACT

The wildcat Felis silvestris is a protected species in Europe but the lack

of information on its status in many areas of its distribution range is an obstacle

to conservation initiatives. To assess the status of the species over a 550 km2
Mediterranean protected area in south-western Spain; Doana National Park, we
carried out track censuses during the wet season of 2007-08 and 2008-09 in 2 x 2
km2 quadrants and set camera traps from June 2008 to October 2010 in quadrants or
nearby quadrants where cat tracks were detected. We detected a total of 52 cat tracks
for both study years and identified six different individuals using morphological
criteria from 28 photographs taken at 12 out of 166 trapping stations. We hypothesized
that the a priori surprising low abundance of the species in the protected area might
be likely multifold and could be explained by the decrease of rabbit population
in the DNP during the last decades, the isolation of DNP from the nearest natural
areas that could have slowed the recovery of wildcat populations after the species
declining at the beginning of the 20th century, an increased mortality rate over time
due to potential disease transmission from domestic cats and/or the competitive
exclusion of the species by the Iberian lynx (Lynx pardinus) in the Doana area.

187

Chapter 5

RESUMEN

El gato monts Felis silvestris es una especie protegida en Europa pero la

escasa informacin sobre su estatus de conservacin en muchas zonas de su rea de

distribucin, es un obstculo para la conservacin de la especie.

Para determinar el estado de conservacin de la especie en un rea

Mediterrnea protegida situada en el suroeste de Espaa; el Parque Nacional de

Doana, realizamos censos de rastros durante la estacin hmeda de los aos 20072008 y 2008-2009 en cuadrculas de 2 x 2 km2, y colocamos cmaras de fototrampeo entre junio del 2008 y octubre del 2010 en cuadrculas en las cuales se
detectaron rastros de gatos o en cuadrculas cercanas. Detectamos un total de 52
rastros de gato en ambos aos de muestreo e identificamos 6 individuos a travs
de criterios morfolgicos a partir de 28 fotografas obtenidos en 12 de las 166
estaciones de foto-trampeo colocadas. Sugerimos que la a priori sorprendentemente
baja abundancia de la especie en el rea protegida debe estar originada por causas
mltiples como el descenso de las poblaciones de conejos en Doana durante las
ltimas dcadas, el aislamiento de las zonas naturales ms cercanas que podra
haber ralentizado la recuperacin del gato monts despus del declive de la especie
a inicios del siglo XX, una mayor tasa de mortalidad a lo largo del tiempo debido
a la potencial transmisin de enfermedades a travs de gatos domsticos y/o una
exclusin competitiva por el lince ibrico (Lynx pardinus) en el rea.

188

Chapter 5

INTRODUCTION

The European wildcat Felis silvestris is one of few wild felids in Europe,

but its conservation status is somewhat paradoxical. The wildcat is listed as of

Least Concern by the IUCN (2008) due to its wide distribution, ranging from the
Iberian Peninsula to Eastern Europe (Nowell and Jackson 1996, IUCN 2007).
Nevertheless, human-mediated habitat disturbance and large-scale hunting in the
early 20th century have led to severe local declines and extirpations in Europe
(Stahl and Lger 1992, Sunquist and Sunquist 2002), resulting in a fragmented
distribution (Stahl and Artois 1991, Nowell and Jackson 1996, Peichocki 2001).
Subsequent legal protection, under the Bern Convention (Appendix II 1979) and the
European Habitat Directive 92/43/EEC (EUROP 1992), has reduced the causes of
this decline and has led to a spontaneous recovery of European wildcat populations
in some parts of Europe (Stahl and Artois 1991). But despite this legal protection,
the wildcat continues to face number of threats throughout its range (Lozano 2009),
with human persecution (predator control) and habitat alteration (Lozano et al. 2007,
Virgs and Travaini 2005) likely the most important. Hence, the European wildcat
is considered to be Near-Threatened in the 25 member states of the European
Union (Temple and Terry 2007) including Spain, where wildcat subpopulations are
suspected to have decreased at a rate of >30% over three generations (Palomo and
Gisbert 2002).

In order to develop action plans for the conservation of the wildcat and

define areas where conservation of wildcats should be priority, it is necessary to

evaluate its distribution, abundance, ecological requirements and population status.
The aim of this study was to assess the presence of European wildcat in a protected
Mediterranean area, the Doana National Park (DNP), where wildcats and other
189

Chapter 5

wildlife have been protected for more than five decades (that is, a priori low human
pressure and persecution), and where habitat and prey availability should be
potentially favourable to wildcat.

Although previous information is anecdotal and sporadic, it seems that the

abundance of the species has been very low long ago. Fifty years ago Valverde
(1967) considered the species as low abundant in the area. More recent available
information refers to a litter of three kittens found within the Doana National
Park in 1997 when looking for Iberian lynx litters (N. Fernndez, pers. comm.),
three individuals captured in 1999, 30 pictures of wildcats photo-trapped between
2000 and 2007 and a total of 52 direct sightings of the apparently wildcats between
1989 and 2000 collected by all personal working in the Doana National Park
(Centro International de Estudios y Convenciones Ecolgicas y Medioambientales
(CIECEM)) (see Annex III). Note that for the same period, other carnivores such as
foxes or lynxes were sighted in 1211 and 669 occasions, respectively.

The main objective of this study was to assess the wildcats current status

in the Doana area and to discuss about the factors explaining its abundance.
Given how little is known about wildcat biology and the vulnerability of Iberian
populations, we aimed to provide baseline data to promote further research and
conservation of the wildcat in southern Spain.

METHODS
Study area

DNP is approximately a 550 km2 area in southwestern Spain bordered to

the south and west by the Atlantic Ocean and to the east by the Guadalquivir River
mouth. The area is flat and mostly near sea level; soils are predominantly sandy and
190

Chapter 5

of marine origin. The climate is Mediterranean subhumid and has marked seasons:
winters are mild and wet, and summers are hot and dry. Mean annual precipitation is
approximately 550 mm. Its situation between the Atlantic Ocean and Mediterranean
Sea and in southern Europe promotes some of the highest biological diversity on
the continent, particularly of vertebrate animals and vascular plants (FernndezDelgado 1997). There are three main biotopes in the park: scrubland, dunes, and
marsh (Valverde 1958). The dune area is situated at the western border of the
protected area where it is limited by the Atlantic Ocean and the marsh area lies at the
northern and eastern borders limited by the Guadalquivir River. The Mediterranean
scrubland represents approximately half of the National Park surface area and
is mainly characterised by heterogeneous patches of xerophytic species such as
Halimium sp. and Cistus sp., and hydrophytic ones such as Erica sp., with some
patches of Juniperus phoenica and Pistacia lentiscus shrubs. Interspersed among
the scrubland are scattered cork oak trees (Quercus suber) and wild olive trees (Olea
europea), and a few patches of pine Pinus pinea and eucalyptus Eucalyptus sp.
plantations. Vegetation in bare sand dunes is scarce and dune hollows are colonized
by pines Pinus pinea and varied scrubland species.

DNP is fully protected and access to the core area and the dirt-road network

inside DNP is restricted to the park staff and researchers. The northern and western
edges of DNP are in close contact with human settlements, crop fields and a high
use paved road (Fig. 1). These surroundings support intense human activity, with
private large- and medium-sized farms used for agriculture as well as six visitor
centers, hiking and cycling paths, recreation zones and bird observatories in the
nearby area.

191

Chapter 5

Field methods

We used track surveys as a first approximation and photo-trapping studies

as a more directed methodology (since tracks of wildcats are not distinguishable

from that of domestic cats) to analyze the presence and/or occurrence of wildcats in
DNP. We carried out systematic track surveys at 69 and 67 2 x 2 km2 quadrant cells
located within the entire scrubland and dune areas of the DNP during the wet season
of 2007-08 and 2008-09, respectively. Marshland area was not sampled as its clay
soils make it unsuitable for track censuses. We sampled for cat tracks in each square
by slowly walking (ca. 1.5 km/h) at least 3 km along sandy roads and firebreaks.
Once a track was detected, we georeferenced it using a GPS. We re-sampled the
same path (leaving at least seven days between samplings) a second time in a few
squares until completing 3 km if during the first sampling there were insufficient
available paths within the square to achieve this distance. We always carried out
surveys at least three days after any rainfall. Potential prey availability for wildcats
was also estimated by counting tracks of European rabbits (Oryctolagus cuniculus),
red partridges (Alectoris rufa) and small mammals (probably mostly long-tailed
field mouse (Apodemus sylvaticus) according to Kufner and Moreno 1989) in 25 m
long and approximately 1.7 m wide transects separated by at least 300 m (see Soto
et al. 2012). We also visually estimated variables related to vegetation type and
structure along transects in a circle of 15 m radius around the sampling point (see
Table 1 for variable description).

To asses the occurrence of wildcats and to determine if tracks detected

belonged to the species, we used camera-trapping techniques from June 2008 to

November 2010. Thirteen camera trap surveys with an average duration of 23 days
were conducted. To set the cameras, we selected quadrants and nearby quadrants
192

Chapter 5
Table 1. Variables used to compare means between quadrants with no evidence of cat presence,
quadrants where wildcats were photographed and quadrants where cat tracks were detected during
track censuses by non-parametric tests.
Variable
Vegetation

Code

Definition

Units

Short shrub
Short shrub
height
Tall shrub

%S

Mean cover of short scrubland per quadrant*

S_h

Average height of short scrubland per quadrant*

%B

Mean cover of bushes per quadrant*

Tall shrub height

B_h

Average height of bushes per quadrant*

Trees

%T

Mean cover of trees per quadrant*

Tree height

T_h

Average height of trees per quadrant*

Landscape

Distance to water DW

Distance to La
Vera

DV

Ecotones between
pastureland and
eBP
scrubland

Measured in meters using a Euclidean distance-based

approach from the quadrant centre to the nearest
permanently flooded natural or artificial pond (i.e., dug
for the cattle at zones were the water table is higher) in a
digitized water sources cover layer of DNP
Measured in meters from the quadrant centre to the
ecotone between the marshland and the Mediterranean
scrubland (locally called La Vera)
Linear measure of the density of the ecotone between
patches with bush cover >50% and patches with pasture
cover >50% defined from a reclassified fine-scale 1:10
000 vegetation map for the year 1996-2006 obtained from
the Sistema de Informacin Ambiental de Andaluca

m/ha

Prey availability
Rabbits

Ra

Small mammals

SM

Total prey

Tot

Kilometric Abundance Index of rabbits per quadrant *

Tracks/km
Kilometric Abundance Index of small mammals per
Tracks/km
quadrant *
Kilometric Abundance Index of rabbits + partridges +
Tracks/km
small mammals + ungulates per quadrant *

Human disturbance
Distance to
antropic edge

DH

Measured in meters from the quadrant centre to the

nearest protected area border influenced by humans (i.e.
excluding the beach and marshland edges)

Predators occurrence
Kilometre
abundance index
of domestic dogs
Kilometre
abundance index
of lynxes

KAId Number of dog tracks detected per km per quadrant

KAIl

Number of lynx tracks detected per km per quadrant

m/ha
tracks/km

* calculated by averaging values obtained at the different sampling points within quadrants

193

Chapter 5

where cat tracks were detected during track censuses. Camera traps were set
in the borders of car tracks and firebreaks, or in the edges of patches of dense
Mediterranean shrub with pasturelands (habitat potentially favourable for wildcats
(Lozano et al. 2003). In order to maximize the detection of wildcats we used scent
lures of valerian, catnip and canned sardines sprayed on a piece of cotton attached
to a wooden stake at a 3050 cm height, wet fish (sardines), live prey (reported as
the most efficient lure for sampling some felid species (Guilt et al. 2010, Garrote et
al. 2012)) such as rock pigeons (Columba livia) and rabbits (Oryctolagus cuniculus)
in wire cages inaccessible to wildcats, as well as no attractants. Cages of live prey
were approximately 100 x 50 x 50 cm and supplied with ample food and water
at least twice a week. On average, we set up 15 digital camera traps with passive
infrared motion sensors and automatic flash (Cuddeback Digital Scouting Model
Expert) per quadrant. Cameras were placed 20 cm above ground, at a distance
of 2-4 m from the lure with 300-400 m between them. We set camera traps with a
delay time of 1 min between successive photos, and checked at least twice per week
to replace attractant lures and twice a month for battery replacement.

Differentiation between wildcats and domestic cats was based on the general

physical appearance and on the pelage pattern of the individuals. Studies on the
European wildcat have indicated that camera trapping can be used to some extent
to determine the presence and abundance of this species and that individuals are
identifiable based on their morphology (Ragni and Possenti 1996, Monterroso et al.
2005, Anile et al. 2007, Karanth et al. 2004). We considered photos as independent
events when taken more than 4 hours apart for the same individuals or if different
individuals could be identified (OBrien et al. 2003).

194

Chapter 5

Figure 1. Locations of cat tracks detected during track censuses in 2 x 2 km2 quadrants during
2007-2009 (a); locations of the 166 trapping stations set during 2008-2010 (b), and the camera trap
stations that provided pictures of wildcats and domestic cats, respectively, as well as the home range
of the wildcat radio-tracked in DNP (c) are shown.
195

Chapter 5

We performed a KruskalWallis test to determine whether different measures of

landscape structure, vegetation type, prey availability, human disturbance and the
occurrence of other carnivore species (i.e., domestic dogs (Canis familiaris) and
Iberian lynx (Lynx pardinus) as potentially negatively affecting wildcats (Palomares
and Caro 1999) (Table 1) differed across three categories of quadrants with different
wildcat presence: (0) quadrants with no evidence of wildcat presence (i.e., neither
photographs nor tracks), (1) quadrants where wildcats were photographed and
(2) quadrants where cat tracks were detected. Multiple post hoc comparisons of
mean ranks were also estimated to detect statistically significant differences for all
pairs of groups (Siegel and Castellan 1988). KruskalWallis analyses and post hoc
comparisons were conducted in SPSS software.

We also trapped a wildcat in a box-trap in December 2008 under a broader

project that aimed to study the effectiveness of red fox control actions within DNP.
The captured animal was chemically immobilized with a 0.75 ml dose (100mg*ml-1)
of tiletamine- zolazepam (Zoletil, Virbac, Spain), measured, weighted, checked
for any sanitary disorders and sexed. Genetic analysis from blood samples collected
revealed that the individual was pure or without any indication of parental
domestic heritage (Alves, P.C., pers. comm.). After handling, the individual was
maintained in the dark and returned to the capture location for release after complete
recovery of reflexes (13 h). The individual was fitted with a radio-collar (Wildlife
Materials, Inc., Carbondale, Illinois, USA), radio-tracked between December 2009
and March 2010 and located on average twice per week between 9:00 am and 2:00
pm. We used triangulation to determine the position of the individual (White and
Garrott 1990) and the minimum convex polygon method to estimate the home
range size based on all available locations (Mohr 1947, White and Garrott 1990)
196

Chapter 5

with Hawths Analysis Tools (Beyer 2004) in ArcGIS (ESRI, Redlands, CA). We
determine the main vegetation types included within the individual home range
from a 1:10 000 fine-scale vegetation map for the year 1996-2006 obtained from the
Sistema de Informacin Ambiental de Andaluca (http://www.juntadeandalucia.es/
medioambiente/site/rediam/)

RESULTS

We detected a total of 25 and 27 cat tracks in 8 and 17 of 69 and 67 quadrants

censused in each year, respectively (Fig. 1). We set cameras at 166 different points
in 25 quadrants. Camera effort was 5761 trap-days with an average of 24.4 day/
cameras (SD= 8.02, range=1-33). We obtained a total of 2173 photographs in
which we identified mammals (n = 2050) or birds (n = 123). The red fox was the
most common species photographed followed by the Egyptian mongoose and the
common genet (Table 2). Thirty pictures of cats were taken at 12 of the 166 points
where we set camera trap stations. Twenty-eight of these pictures were of wildcats
and two pictures were of domestic cats. These camera traps were baited with
pigeons (n = 8) and wet sardines (n = 6) (Table 2). Wildcats were photographed
between 20:00 and 07:00 hours. Six different wildcats were identified (Fig. 2) from
12 camera trap records (Table 3) and all were photographed only once. None of
the individuals could be sexed, nor could the presence of more than one individual
per trap-site be ascertained. Two-hundred six camera-trap days were required on
average to document the presence of an individual.

Radio-tracking effort produced 24 locations and a home range size of

approximately 24 km2 (Fig. 1). More than 60% of the individual home range
included areas of Mediterranean short scrubland (i.e., species such as Halimium sp.
197

Chapter 5

and Cistus sp.) and approximately 18% pine woodlands with understorey vegetation
(i.e., short shrubs and tall shrubs such as Erica sp., Juniperus phoenica and Pistacia
lentiscus).

We found differences between quadrants where wildcats were photographed,

quadrants with only cat tracks and quadrants with no evidence of cat presence for the
Table 2. Total number and percentage of pictures taken during camera-trapping surveys at DNP.
Specie

Felis silvestris

28

1.3

Felis catus

0.1

Lynx pardinus

0.1

Genetta genetta

60

2.8

Herpestes ichneumon

152

7.0

26

1.2

Meles meles
Vulpes vulpes

1284 59.1

Canis familiaris
Others*

0.3

611

28.1

*other mammals and birds

Table 3. UTM coordinates (29S) of the camera-trap positions where wildcat were photographed.
The period of time the camera was active and the number of wildcat pictures taken and baits used
are also provided.

Trap station

198

Pictures

Trap-days

Bait

185585

4106480

22

Pigeon

186451

4106914

22

Pigeon

185780

4100419

10

Pigeon

186990

4099547

29

Wet sardine

184710

4101010

29

Wet sardine

185476

4107561

29

Wet sardine

185967

4108394

29

Pigeon

185895

4108264

29

Wet sardine

186503

4109315

29

Wet sardine

10

183961

4117788

20

Pigeon

11

190451

4105106

22

Pigeon

12

189275

4105888

30

Pigeon

Chapter 5
Figure 2. Individual wildcats distinguished on the basis of their external aspect in the Doana
National Park between 2008 and 2010.

variable distance to the anthropic edge of the protected area (Table 4). Mann-Whitney
U post-hoc tests revealed that quadrants where wildcats were photographed were
closer to the anthropic edge of the protected area than quadrants with no evidence of
cats and than quadrants where tracks were detected (Table 5). No differences were
found for any of the remaining variables analysed.

199

Chapter 5

DISCUSSION

This study provides the first systematic information on the occurrence of

wildcats throughout the DNP and reveals a priori surprising low abundance of the
species in the protected area. Wildcats like autochthonous Mediterranean scrubland
areas with scrubpastureland mosaics (Lozano et al. 2003), and may rely for
feeding on rabbits (Gil-Snchez et al. 1999, Lozano et al. 2006)) or small mammals
(Sarmento 1996, Molen and Gil-Snchez 2003, Carvalho and Gomes 2004). These
habitats and prey are common in many parts of the DNP, so one would expect that
wildcats were more abundant in the area than what we have found. Furthermore,
protection of the DNP for more than 5 decades has provided a safe place for wildcats.
Therefore, DNP should hold one of the largest wildcat populations in south western
Spain and be one of the most important areas for conservation of the species.

Wildcat scarcity in DNP might be due to several factors. First, during the

last decades wild rabbit population has decreased everywhere and in the DNP due to
diseases such as myxomatosis and Rabbit Hemorrhagic Disease (RHD) (Thompson
and King 1994, Villafuerte et al. 1995) and to changes in scrubland management
(Moreno and Villafuerte 1995). Abundance of the wildcat population might have
diminished for this reason. In fact, the large home range size of the radio-tracked
individual in our study area suggests low food abundance. However, there are some
areas where rabbits are abundant within the park, and wildcats can also consume
other alternative prey species such as small mammals (Lozano et al. 2003, Malo et
al. 2004), so this reason not fully explain the low abundance of wildcats in the area.
Secondly, the isolation of DNP from the nearest natural areas (Sierra Morena and
Cdiz) due to human settlements, widespread field crops or the Guadalquivir River
may also contribute to the low abundance of the species. Wildcats disappeared
200

Chapter 5

from many regions across its range and reached minimum levels at the beginning
of the 20th century (McOrist and Kitchener 1994). The recovery of the species
in several places was possible in the 1990s when anthropic pressure on wildcat
populations and their habitat was reduced (e.g., Parent 1975, Easterbee et al. 1991)
but the isolation and fragmented distribution of the species in Doana may have
prevented the recovery in spite of the reduction in potential threats. Additionally,

Table 4. Results of the Kruskal Wallis Test to test for differences in the means of several variables
between (0) quadrants with no evidence of cat presence, (1) quadrants where wildcats were
photographed and (2) quadrants where cat tracks were detected. Significant variables are represented
in bold.
Mean rank
Variable

Kruskal Wallis Test

2

Chi-square P-value

Vegetation
%SB

37.03 39.58 29.54

2.574

0.276

%S

33.72 43.17 33.58

1.265

0.531

S_h

38.70 34.92 28.18

4.229

0.121

%B

37.77 35.00 29.54

2.590

0.274

B_h

35.36 27.67 34.86

0.796

0.672

%T

33.66 29.67 36.90

0.796

0.672

T_h

35.35 32.00 33.84

0.193

0.908

38.41 16.33 33.08

6.637

0.036

eBP

38.24 32.83 29.36

3.065

0.216

DW

31.43 34.17 39.12

2.257

0.324

DV

30.81 50.00 36.24

5.168

0.075

KAId

32.43 40.67 36.08

2.437

0.296

KAIl

36.65 34.50 31.32

1.731

0.421

32.34 27.42 34.00

0.606

0.739

SM

29.25 25.00 38.98

4.963

0.084

Tot

32.40 28.83 33.56

0.312

0.856

Human disturbance
DH
Landscape

Predators

Prey

201

Chapter 5

an increased mortality rate over time due to potential disease transmission from
domestic cats (e.g., McOrist et al. 1991) could be another causal factor explaining
wildcat scarcity in Doana. The domestic cat population in the Doana area might
have increased over the last decades due to the urbanization of the surrounding
neighbourhoods of the protected area. Hence, as in other natural areas (e.g., Ferreira
et al. 2011), feral and/or domestic may spend time sporadically within DNP during
their free-ranging activity potentially transmitting diseases to their wild relatives.
In fact our results confirm the presence of domestic cats within the DNP close to
human settlements (Fig. 1, Table 4). Finally, although our results do not confirm
that the relative abundance of the Iberian lynx can negatively affect the detection
of wildcats, interspecific interactions between wildcats and the Iberian lynx may
have also led to the competitive exclusion of the species as carnivores persisting
at low population densities have been suggested to experience an increase in the
effect ofintraguildpredation (Creel and Creel 1998, Creel 2001, Creel et al. 2001,
Creel and Creel 2002). The European wildcat and the Iberian lynx may potentially
overlap in habitat use of Mediterranean scrubland but intraguild interaction with
the Iberian lynx may have resulted in habitat partitioning in that wildcats will avoid
habitat patches of high lynx densities to the detriment of its own success in prey
acquisition and access to the most suitable habitats. In turn, wildcats may have
been forced to enlarge their home ranges to continue hunting and may even roam
in human-occupied areas increasing their mortality risk. In fact our results revealed
that wildcats were detected more frequently near the anthropic edge of the area
suggesting that individuals could be ranging outside of DNP. Nevertheless, in spite
of the continued decrease of the Iberian lynx population in the Doana area in
recent decades (Palomares et al. 2012) wildcats have not increased its abundance
202

Chapter 5

in the area confirming the hypothesis that the low occurrence of the species in DNP
can not be only attributed to Iberian lynx decrease.

In summary, although the DNP is optimal for a large wildcat population the

potential threats explained above may shed light on the low occurrence of the species
in the area. Nevertheless, many threats to the species may remain unidentified in
Doana, so although its low population density makes field studies and direct
observation difficult, more research and detailed information on occurrence as well
as on wildcat habitat requirements based on radio-tracking efforts are necessary to
provide guidelines for management and conservation of the species in the area.

ACKNOWLEDGEMENTS

This research was funded by the projects CGL2004-00346/BOS (Spanish Ministry of

Education and Science) and 17/2005 (Spanish Ministry of the Environment; National Parks Research
Program). Land-Rover Espaa lent us two vehicles for this work. We are very grateful especially to
J.C. Rivilla and S. Desnica for assistance during fieldwork. C. Soto was also supported by a JAE
Predoc grant from the CSIC.

REFERENCES
Anile S, Bizzarri L, Ragni B (2007) Experiences obtained from camera trapping the wildcat in Sicily

(Italy). Hystrix Italian Journal of Mammalogy Supplement V European Congress of

Mammalogy, 294.

Beyer HL (2004). Hawths Analysis Tools for ArcGIS. Available at www.spatialecology.com/htools.

Carvalho JC, Gomes P (2004) Feeding resource partitioning among four sympatric carnivores in

the Peneda-Gers National Park (Portugal). Journal of Zoology 263: 275283.

Creel SPG, Creel N (2001)Interspecificcompetition and the population biology of extinction

prone-carnivores.In: Carnivore Conservation. Eds. Gitleman, JL, Funk, SM

MacDonald, D Wayne RK. pp. 35-60. Cambridge University Press.

Creel S (2001) Four factors modifying the effect of competition on carnivore population dynamics
203

Chapter 5

as illustrated by African wild dogs. Conserv Biol 15 (1):271-274.

Creel S, Creel NM (1998) Six ecological factors that may limit African wild dogs, Lycaon pictus.

Animal Conservation 1 (1):1-9.

Creel S, Creel NM (2002) The African wild dog: behavior, ecology, and conservation. The African

wild dog: behavior, ecology, and conservation. Princeton University Press.

Daniels MJ, Balharry D, Hirst D, Kitchener AC, Aspinall RJ (1998) Morphological and pelage

characteristics of wild living cats in Scotland: implications for defining the wildcat. J

Zool 244:231-247.

Duarte J, Vargas JM (2001) Son selectivos los controles de predadores en los cotos de caza?

Galemys 13 (special number): 19.

Easterbee N, Hepburn LV, Jefferies DJ (1991) Survey of the status and distribution of the wildcat

in Scotland, 1983-1987. Survey of the status and distribution of the wildcat in Scotland,

1983-1987. Nature Conservancy Council for Scotland, Edinburgh.

Garrote G, Gil-Sanchez JM, McCain EB, de Lillo S, Telleria JL, Simon MA (2012) The effect of

attractant lures in camera trapping: a case study of population estimates for the Iberian lynx

(Lynx pardinus). Eur J Wildl Res 58 (5):881-884.

Gil-Sanchez JM, Valenzuela G, Sanchez JF (1999) Iberian wild cat Felis silvestris tartessia predation on

rabbit Oryctolagus cuniculus: functional response and age selection. Acta Theriol 44 (4):421-428

Guilt F, Agudin S, El-Khadir N, Fernandez-Olalla M, Figueredo J, Dominguez FG, Garzon P,

Gonzalez G, Munoz-Igualada J, Oria J, Silvestre F (2010) Factors conditioning the camera-

trapping efficiency for the Iberian lynx (Lynx pardinus). Eur J Wildl Res 56 (4):633-640.

Ferreira JP, Leito I, Santos-Reis M, Revilla E (2011) Human-Related factors regulate the spatial ecology

of domestic cats in sensitive areas for conservation. PLoS ONE 6(10): e25970 (2011).

Ferrer M. 1993. El guila imperial. Quercus, Madrid, Spain.

Fernndez-Delgado C (1997) Conservation management of an European Natural Area: Doana

National Park, Spain. En: Groom, M.J., Meffe, G.K., Carroll, C.R. (Eds.), Principles of

Conservation Biology, pp. 536-543. Third Edition. Sinauer Associates, Inc. Publishers,

Sunderland, Massachusetts, USA.

IUCN (2007)European Mammal Assessment. IUCN, Gland, Switzerland and Cambridge, UK.
Jackson R.M., Roe J.D., Wangchuk R., Hunter D.O. 2005. Surveying Snow Leopard populations
204

Chapter 5

with emphasis on camera trapping. A handbook. Sonoma, California, The Snow Leopard

Conservancy: 73.

Karanth KU, Chundawat RS, Nichol JD, Kumar NS (2004) Estimation of tiger densities in the

tropical dry forests of Panna, Central India, using photographic capture-recapture sampling.

Animal Conservation 7:285-290.

Lozano J, Moleon M, Virgos E (2006) Biogeographical patterns in the diet of the wildcat, Felis
silvestris Schreber, in Eurasia: factors affecting the trophic diversity. Journal of

Biogeography 33 (6):1076-1085.

Lozano J, Virgos E, Cabezas-Diaz S, Mangas JG (2007) Increase of large game species in

Mediterranean areas: Is the European wildcat (Felis silvestris) facing a new threat?

Biological Conservation 138 (3-4):321-329.

Lozano J, Virgs E, Malo AF, Huertas DL, Casanovas JG (2003) Importance of scrubpastureland

mosaics for wild-living cats occurrence in a Mediterranean area: implications for the

conservation of the wildcat (Felis silvestris). Biodivers Conserv 12 (5):921-935.

Lozano 2009. Gato montsFelis silvestris. In: Enciclopedia Virtual de los Vertebrados Espaoles.

(L. M. Carrascal and A. Salvador, Eds.). Museo Nacional de Ciencias Naturales, Madrid.

http://www.vertebradosibericos.org/

Malo AF, Lozano J, Huertas DL, Virgos E (2004) A change of diet from rodents to rabbits
(Oryctolagus cuniculus). Is the wildcat (Felis silvestris) a specialist predator? J Zool
263:401-407.
McOrist S, Kitchener AC (1994) Current Threats to the European Wildcat, Felis silvestris in

Scotland. Ambio 23 (4-5):243-245.

Mohr CO (1947) Table of equivalent populations of North American Small Mammals. Am Midl

Nat 37 (1):223-249.

Moleon M, Gil-Sanchez JM (2003) Food habits of the wildcat (Felis silvestris) in a peculiar habitat:

the Mediterranean high mountain. J Zool 260:17-22.

Monterroso, P., Sarmento, P., Ferreras, P. and Alves, P. C. 2005. Spatial distribution of the European

wildcat (Felis silvestris) in Vale do Guadiana Natural Park, South Portugal. Symposium:

Biology and Conservation of the European Wildcat (Felis silvestris) (ed M. Herrmann), pp.

17. NABU, Germany, January 21st 23rd 2005.

205

Chapter 5
McOrist, S., Boid, R., Jones, T.W., Hubbard, A.L., Easterbee, N., Jarret, O., 1991.Some viral and

protozoal diseases of the European wildcat Felis silvestris. Journal of Wildlife Diseases 27,

693696.
Moreno S, Villafuerte R (1995) Traditional management of scrubland for the conservation of
rabbits Oryctolagus cuniculus and their predators in Doana National Park, Spain.

Biological Conservation 73:81-85.

OBrien TG, Kinnaird MF, Wibisono HT (2003) Crouching tigers, hidden prey: Sumatran tiger

and prey populations in a tropical forest landscape. Animal Conservation 6:131-139.

Palomo LJ and Gisbert J (2002) Atlas de los Mamferos terrestres de Espaa. Direccin General de

Conservacin de la Naturaleza-SECEM-SECEMU, Madrid.

Parent GM (1975) La migration recente, a caractere invasionnel, du chat sauvage, Felis silvestris

Schreber, en Lorraine Belge. Mammalia 39 (2):251-288.

Peichocki, R. 2001. Lebensrume. Die Verbreitung der Wildkatze in Europa. In: H. Grabe and G.

Worel (eds),Die Wildkatze. Zurck auf leisen Pfoten, pp. 14-27. Buch and Kunstverlag

Oberpfalz, Amberg.

Ragni B. 2006. Il gatto selvatico. Salvati dallArca. A. Perdisa Ed., Bologna.

Rodriguez A, Delibes M (1992) Current range and status of the Iberian Lynx Felis-Pardina

Temminck, 1824 in Spain. Biological Conservation 61 (3):189-196.

Ruiz-Olmo J (1986) Dades sobre les causes de mortalitat dels carnvors (Mammalia) als massissos

del Montseny i del Montnegre i les seves rodalies. In: II trobada destudiosos del Montseny.

Diputacin de Barcelona, Servei de Parcs Naturals, Barcelona, pp. 2123.

Sarmento P (1996) Feeding ecology of the European wildcat Felis silvestris in Portugal. Acta

Theriologica 41: 409-414

Stahl P, Artois M (1991) Status and Conservation of the wild cat (Felis silvestris) in Europe and

around the Mediterranean rim. Council of Europe, Strasbourg.

Stahl P, Leger F (1992)Le chat sauvage (Felis silvestris, Schreber, 1777).En: Artois, M. y Maurin,

H. (Eds). Encyclopdie des Carnivores de France. Socit Franaise pour lEtude etla

Protectiondes Mammifres (S.F.E.P.M.), Bohallard, Puceul.

Stahl P,Artois M, Aubert MFA (1988) Organisation spatiale et dplacements des chats forestiers

206

adultes (Felis silvestris, Schreber, 1777) enLorraine.Revue dcologie (Terre et

Chapter 5

Vie), 43: 113-131.

Sunquist M, Sunquist F (2002)Wild Cats of the World. The University of Chicago Press, Chicago.
Temple HJ, Terry A (2007) The status and distribution of European mammals. Office for Official

Publicationsof the European Communities, Luxembourg.

Villafuerte R, Calvete C, Blanco JC, Lucientes J (1995) Incidence of viral hemorrhagic disease in

wild rabbit populations in Spain. Mammalia 59 (4):651-659.

Valverde JA(1967) Estructura de una Comunidad de Vertebrados Terrestres. C.S.I.C., Madrid.

Virgos E, Travaini A (2005) Relationship between small-game hunting and carnivore diversity in

central Spain. Biodivers Conserv 14 (14):3475-3486.

White GC, Garrott RA (1990) Analysis of wildlife radio-tracking data. Analysis of wildlife radio-

tracking data. Academic Press Inc., San Diego, New York etc.

Wild cats: status survey and conservation action plan (1996). Wild cats: status survey and

conservation action plan. IUCN.

Wilson GJ, Delahay RJ (2001) A review of methods to estimate the abundance of terrestrial

carnivores using field signs and observation. Wildl Res 28 (2):151-164.

207

Chapter 5

ANNEX III

Pictures of sporadic sightings of individual wildcats within the DNP (a and

b), wildcat radio-tracked in the study area between December 2009 and March
2010 (c) and wildcat photographed in March 2011 by the personal working in the
Doana National Park (Centro International de Estudios y Convenciones Ecolgicas
y Medioambientales (CIECEM)) (d). Credit pictures: CIECEM (a, b and d) and C.
Soto (c).

208

Conclusiones

209

Conclusiones

1. La abundancia relativa de especies obtenida a travs de censos de rastros

depende de la abundancia en s de las especies censadas pero tambin de
factores no biolgicos condicionados por las condiciones climticas los das
anteriores al censo y de factores relacionados con el mtodo de censado.

2. Las variables que incrementan la calidad del sustrato sobre el que se

realizan los censos (elevada humedad ambiental, ausencia de viento fuerte,
o un nmero bajo de das transcurridos desde las ltimas precipitaciones)
permiten una mayor deteccin de rastros en sustratos arenosos. Adems,
en funcin del tamao de las especies censadas, variables metodolgicas
como la distancia del transecto sobre el que se realiza el censo al borde de
la vegetacin ms cercana o el observador, tambin afectan al nmero de
rastros encontrados.

3. Para disminuir la variabilidad en el nmero de rastros detectados y aumentar

la fiabilidad de los datos obtenidos se recomienda restringir los censos a
ciertas condiciones climticas y/o metodolgicas, y si no es posible, incluir
en los modelos estadsticos las variables relacionadas con el clima y el
mtodo que potencialmente pueden afectar el nmero de rastros encontrados
durante los censos.

4. De las cinco especies de carnvoros censadas en toda la zona de matorral del

Parque Nacional de Doana, el zorro fue la especie ms comn y ubicua,
encontrndose en el 98.5% de las cuadrculas muestreadas. Le sigui el
tejn (97.1%), el meloncillo (94.2%), la gineta (59.4%) y el lince que ocup
algo ms de un tercio del rea muestreada (37.6%).
210

Conclusiones

5. Modelos de uso del hbitat a escala fina indicaron que la presencia de lince
estuvo asociada a zonas con una alta cobertura de matorral alto y densidad
de ecotonos entre matorral alto y pastizal, mientras que la presencia de
ginetas se asoci a zonas de pinares con vegetacin densa de sotobosque
as como con una alta abundancia de micromamferos. Los zorros, tejones
y meloncillos mostraron un patrn de seleccin de hbitat menos marcado,
siendo la abundancia de presas totales y de conejo de monte, o la cobertura
de matorral alto algunos de los predictores de su abundancia y distribucin.

6. Anlisis de nicho indicaron que ginetas y linces son las especies ms

especializadas con nichos ecolgicos ms estrechos, segregados y
marginales. Los zorros y tejones sin embargo, mostraron un generalismo
manifiesto, situndose espacialmente su nicho ecolgico entre el de zorros
y tejones por un lado y el de linces y ginetas por otro.

7. Cuando a los modelos generales de uso del hbitat se le aadieron las

variables de abundancia relativa o presencia de las otras especies de
carnvoros, se encontr una relacin negativa entre la presencia del lince
y la abundancia relativa de zorros, meloncillos y ginetas. Sin embargo, la
abundancia de ginetas y meloncillos estuvo positivamente asociada con la
abundancia relativa de meloncillos y tejones, respectivamente.

8. Las asociaciones encontradas entre especies se enmarcan dentro de las

relaciones interespecficas conocidas dentro de la comunidad de carnvoros
del Parque Nacional, donde el lince acta como depredador intragremial
de especies mas pequeas como zorros, meloncillos y ginetas, y sugieren
211

Conclusiones

la importancia de incluir la presencia y/o abundancia de los grandes

depredadores en los estudios de uso y seleccin del hbitat de depredadores
pequeos y medianos.

9. Rastros de perros fueron detectados en el 17-23% de la superficie de

matorral del Parque Nacional de Doana. Hubo mayor probabilidad de
encontrar perros en las reas cercanas a los lmites del Parque Nacional, y
no se encontraron rastros en zonas del interior del Parque, lo que sugiere
que se trata de perros procedentes de los ncleos urbanos y que no hay una
poblacin silvestre establecida en el mismo.

10. Para un control efectivo de perros que usan el interior del Parque Nacional
debera actuarse sobre los ncleos urbanos circundantes, as como intensificar
las medidas de control en los lmites del Parque Nacional.

11. La abundancia de gato monts en el Parque Nacional de Doana fue

sorprendentemente baja a pesar del nivel de proteccin que ofrece el rea de
estudio. Durante dos aos de estudio se detectaron solo 52 rastros de gato en
algo ms de un tercio de las cuadrculas muestreadas, y solo se identificaron
6 individuos diferentes de gato monts para lo que fue necesario un esfuerzo
de muestreo de 24.4 cmaras/da para detectar cada uno de ellos.

12. La baja abundancia de gato monts en el Parque Nacional podra explicarse por uno
o ms factores tales como el descenso de las poblaciones de conejo, el aislamiento
de la poblacin, una posible alta incidencia de enfermedades debido al contacto
con gatos domsticos, y la competencia por interferencia con el lince ibrico.
212