Environmental Safety Assessment of

Drilling Operations in the Marcellus-Shale

Gas Development
Richard Olawoyin, SPE, John Y. Wang, SPE, and Samuel A. Oyewole, Pennsylvania State University
Summary
The process of gas development is intensive and involves risk to the environment. Statistics confirm that 0.5 to 1% of wells drilled result
in a blowout. Causes of these exploration risks are identified as violations of environmental laws enforced by the Pennsylvania
Department of Environmental Protections (DEP), operational pol- lution (accidental spills and leaks), and operators policy. In addressing
this concern, a risk-assessment methodology was used to evaluate all violations by operators in the State of Pennsylvania from January
2008 to November 2010, by use of Statistical Anal- ysis Software (SAS). The most significant causes of environmen- tal damage and
risk were determined by use of the doubly repeated measure analysis of covariance (ANCOVA). The cate- gory effect and
interaction effect were used to prove the useful- ness of the developed model, which helps explain the safety level of the locality. There
were a total of 2,601 violations between
2008 and 2010 committed by 65 different operators in the Marcel- lus Shale, out of which only 27 of the operators showed significance difference based on environmentally damaging violations
(ranked 5 to 10). A statistical comparison was made to understand the difference between the operators based on the 2,601 total violations. The most significant incidents are ranked [on the basis of
Borda count (Saari 1985)] 3, 5, 9, 10, which accounts for 67% of all the violations. These data reflect several environmental con- cerns
that are currently prevalent in the Marcellus-shale area. This
research identifies environmental incidents, causes and effects of exploration risk, and safety impediments in the Marcellus gas
play. It also presents guidelines for feasible options to minimize environmental risks and consequently increase the degree of safety in the
area. Recommendations on how to mitigate these impending
problems are presented.
Introduction
The use of the hydraulic-fracturing method in the stimulation of reservoirs of tight formations in the Marcellus area is gaining more
momentum as energy demand across the region and/or over the world increases. Recent advances in petroleum-prospecting techniques
have provided better access to the formations of inter- est through directional drilling and well-stimulation technologies. The US
Geological Survey estimated that there is a technical pos- sibility of recovering up to 200 trillion ft3 of natural gas from shale-gas
formations (Fig. 1). Toward this end, there has been a huge capital-intensive investment in the Marcellus shale by some
international energy firms in regard to developing productive fields, providing advanced technology, and exploring the natural
resource of interestgas.
These events are new to many towns that never experienced drilling and production activities and are intensified in regions
that previously experienced little petroleum-development activ- ity. Even though the production of natural gas to meet current
energy-consumption requirements is essential to the survival of the human population at large, it is also ironically detrimental to the
general well-being of continued human existence. There has been a growing concern about the treatment used in the stimula- tion of
shale-gas production in the Marcellus shale, ranging from contamination of drinkable well water to excessive water with- drawal from
the watersheds and other related environmental/psy- chosocial effects of both the drilling operations and the disposal of flowback
fluids generated from hydraulic fracturing. Waste water is usually produced alongside the gas, and the waste water contains toxins that
potentially contain human carcinogens, such as naturally occurring radium 226 picked up from the subsurface. Total dissolved solids in
high concentrations are also present in this waste water together with some of the chemical additives added to the hydraulic-fracturing
fluid. Data provided by the DEP confirm the various violations by the operators in the Mar- cellus shale area, and it is evident that,
in some areas in Pennsyl- vania, these fluids (waste water and undiluted fracturing chemicals) were either accidentally spilled into the
waters or, in some cases, discharged into water bodies because of the unavail- ability of proxy treatment plants.
Marcellus Shale
Shale is a rock formed from the compaction of fragments of other materials, usually silt- and clay-sized particles. Shale is a sedi- mentary
rock with fine-grained texture that forms from the com- paction of silt and mud. The geologic age of the Marcellus shale is Devonian
(350415 million years ago). The Devonian period was marked by a substantial deposit of organisms and algae on the sea bottom,
which was devoid of oxygen; these organisms were able to settle and decompose because of low circulation and the prevalent
anaerobic conditions. After a period of time, the decomposed matter was transformed into carbon materials, and once hydrated, these
became hydrocarbons. These hydrocarbons include petroleum and natural gases, such as crude oil and meth- ane gas. The deposition
cycle during the Devonian period that formed the Marcellus shale was basically richer in carbonate than other shale formations. Natural
gas is mostly deposited in the frac- tures of these rocks in the form of unrestricted gas or as adsorbed gas on the clay surfaces and
kerogen inside the shale-rock matrix (Haines 2006). The shale rocks trap the fluids and make it chal- lenging to extract these fluids
commercialy from the rock, except when natural fractures or induced fractures are connected. The Marcellus shale is situated in the
eastern region of the United States and some parts of Canada. The lateral extent of the Marcel- lus shale is approximately 965.6 km (600
miles) (Durham 2008), from New York (which is to the south), it cuts across mostly the western part of Pennsylvania and extends
through West Virginia and the eastern side of Ohio (Durham 2008). The entire coverage of the Marcellus shale has been approximated to
be a bit larger than the state of Florida, with an area of approximately 140 000
km2 (54,000 sq miles) (Mayhood 2008).
Types of Drilling Fluid Used in the Marcellus Shale. The type of drilling fluid used during a drilling process depends mainly on
factors such as pressure, formation type, and design method. In shale-gas drilling, a few drilling-fluid types are used most often:
aqueous or water-based fluids (WBFs), oil-based fluids, synthetic- based fluids, and air and foam fluids. The constituents of the dif2012 SPE Drilling & Completion

ferent types of drilling fluids vary considerably. For WBFs, the constituents include brine/water (76%), barite (14%), clay/poly- mer
(6%), and other (4%). The nonaqueous fluids are composed of nonaqueous fluid (46%), barite (33%), brine (18%), emulsifiers (2%),
and gellants/other (1%) (Clark 1949).
Hydraulic Fracturing
Hydraulic-fracturing technology is required for formation stimula- tion and for optimal recovery. From the time the technology was
3
developed, approximately 70 years ago, more than 7 billion bbl of crude oil and 600 trillion ft of natural gas have been produced
(American Petroleum Institute 2010).
The method enhances the ability of fluid to move more freely within the rock pores by creating cracks in the formation that
serve as pathways for the oil or gas to reach the wellbore and travel to the surface. Fracturing fluids are made up mostly of
water and other chemical additives (added to stall the growth of scale and bacteria around the wellbore) that are pumped at high pressures
into the formation through a designed section of the well casing.
Rick and Safety Issues in Marcellus Gas
Development
Various environmental risk and safety concerns in the Marcellus- shale play have been pointed out with regard to the potential impact to
the environment by the development of natural gas in this area by operators. This paper discusses the severity of envi- ronmental
violations committed by operators. In identifying these problems, it is beneficial for operators to attempt to responsibly guarantee the
effective treatment and disposal of drilling and pro- duction wastes, protect the water supplies, ensure the preservation of quality air, and
tackle all safety concerns within the premises of their operations (Ground Water Protection Council and ALL Consulting 2010). Also
discussed are both the failure of these pe- troleum operators to comply with the regulations that could con- sequently lead to possible
environmental considerations and challenges (US Department of Energy 2009) and the potential threat to human health and the
degradation of the environment, resulting from natural-gas extraction in the Marcellus Shale.
Water Availability, Pollution, and Disposal. The quantity of water needed to drill and/or hydraulically fracture a horizontal gas well
in the Marcellus Shale is huge, between 2 and 5 million gal that mostly come from surface water bodies such as lakes, rivers, and
streams. However, it sometimes comes from different sources, such as underground water, municipal and private water, and recycled
produced water (Ground Water Protection Council 2009). The depth of groundwater in the different US shale formations is shown in
Fig.
2. In addition to the contamination of underground water sources, the surface water is also at risk of uncontrolled drilling fluids, additives, fracturing fluids, flowback water, and produced water.
The design of a fracturing job would include taking into consid- eration wastewater- and drilling-fluid-spill incidents and the storage of these fluids. The waste water is composed mainly of the flowback and produced water that result from the well-stimulation
process and production, respectively. These fluids are retrieved and stored in containment tanks that are then treated for reuse in some
cases. Other operators transport the waste water to nearby treatment facilities with tanks or through wastewater pipelines, and then it is
discharged into a river. Other options of disposal include dumping
the flowback and produced water into deep disposal wells, but strict regulations do not favor this option (Arthur et al. 2008).
Impact Prediction Methodology. Assessment of the potential environmental impacts was undertaken by means of environmental-,
social-, and health-assessment tools. The process included project activities and environmental sensitivities interaction, and impact
identification, description, and rating (a term that includes the pre- diction of magnitude, consequence, and significance of impacts).
The Environmental Impact Assessment process considers not only interactions between impacts of the various project activities and the
sensitivities but also the interactions among the sensitivities.
Impact of Drilling. The rig is one of the first and most conspicu- ous pieces of equipment for the development on site. A proficient
engineer ensures proper measurement of background noise on drilling sites and examines the topography with the objective of
determining the appropriate noise-reduction measures to imple- ment, with consideration for the residents near the site areas

2012 SPE Drilling & Completion

(Chesapeake Energy 2009). Other environmental factors consid- ered include low emission capacity [thereby meeting the terms of the
Clean Air Act (US Environmental Protection Agency 1990)], speed, accuracy (which reduces the occurrence and release of natu- rally
occurring radioactive materials), and safety. The environmen- tal impact of the drilling fluid used for the shale-development practices is
an important aspect of the process. Thus, many factors are considered before choosing drilling fluids. Drilling is a regu- lated practice
managed at the state level. Whereas state gas agen- cies have the ability to require operators to vary from standard practices, the agencies
typically do so only when it is essential to safeguard gas resources and the environment. Water impound- ments also have been used to
contain water for hydraulic-fracturing needs and are usually lined to curtail the loss of water resulting from infiltration. The use of pits
as water-storage facilities is becoming an important tool in the shale-gas industry because the drilling and hydraulic fracturing of these
wells often require signifi- cant volumes of water as the base fluid for both purposes.
Economic Losses. Several economic consequences to environ- mental problems are generated as a result of shale-gas development.
However, three major areas come into focus when analyzing these economic impacts: claims from personal injury or accident, cost of
drilling fluid, and the cost of environmental remediation. It is im- portant not only to examine the contributing factors to these incidents or accidents before any monetary value is attributed to them but also to analyze the associated expenses of the claims.
Pollution Sources and Environmental Risks. Both the materi- als used for drilling operations and the waste generated during and
after the well development are potential contaminants. Mate- rialssuch as the drilling fluids, additives, and chemicals (as well as the
waste materials, such as produced water and waste water)often find their way into the environment through inten- tional,
unintentional, and sometimes permitted releases. The pol- lutants and sources include the following.
Drilling Wastes. These consist of fluid and solid types that are very potent, and they degrade the environment in various ways. The
estimated volume of drilling waste generated from each well during development is 1000 to 5000 m3, and when a number of wells are
present, the waste generated also increases exponentially (Stanislav and Cleveland 2008).
Produced Water. Produced water is derived during the pro- duction of a gas. The mineralized water is highly saline, approxi- mately
300 g/L, which is much higher than seawater. The extent of salinity may depend on the formation type. In addition, the heavy- metal
content is very high (Neff et al. 1987; Stanislav 2008).
Air-Pollution Potential. Natural deposits contain hydrogen-sul- fide gas and some volatile organic compounds that, when released to
the environment, are toxic and can harm the health of local residents, workers, and organisms that come in contact with these toxins.
Blowout. A blowout incident is an environmental disaster because the contamination of the aquifer, surface water, soil and air is
likely to occur, and even explosions can, and the impact of these occurrences includes project economics, remediation costs, liability
costs, equipment repairs, and lost time because of shut-in and other causes (Clark 2002). Other sources of pollution include pipeline
leakages and mechanical failures, and accidents due to infrastructural damages to petroleum installations.
Technology SolutionsEnvironmental Sustainability and Profitability. Several pollution-prevention technologies have been
developed to effectively reduce or eliminate environmental pollutions in the oil and gas exploration and production (E&P) industry.
Newly developed engineering and operational techni- ques in E&P have emphasized the following:
Smarter: intelligent and fast computing; referred to as smart wells
Farther-deviated wells: such as horizontal wells or multiple laterals that are laterally extensive
Deeper wells: deep subsalt E&P
Cleaner operations: downhole-separation technology, gas-to- liquids conversion
Smaller footprint: minimal effect on the environment, use of microhole technology
Because of these advancements, fewer wells are drilled, smaller
well pads are used, and fewer roads are affected, whereas more gas is produced, resulting in less waste generated. These novel
extraction techniques are changing the landscape of drilling, and they have fewer impacts on the ecosystem than conventional techniques. The economic importance of this advancement is huge
because wildlife, aquifer life, vegetation, surface and underground water sources, and air quality are minimally affected during developmental activities.
Analysis, Result, and Discussion
Project Description. Data from the Pennsylvania DEP on viola- tions of operators in the Marcellus Shale from 2008 to 2010 were
gathered and analyzed. The total number of violations for each year is summarized (Table 1).
Marcellus-Shale Violations. The Borda count method that is described by Saari (1985) is an iteration-process tool that is useful in
decision making. The technique was used to rank the different
violations by operators during 20082010. Table 1 presents the violations ranked into different subgroups.
According to the Pennsylvania DEP, there are 1,450 horizontal wells in the Marcellus-shale play and a total of 551 vertical wells,
making the total number of spudded (initial-drilled) wells in the
Pennsylvania Marcellus play to be 2,001.
In the ranking list in Table 2, 1 categorizes the lowest risk level or least hazardous incident and 5 is the intermediate risk level.
From left to right, the risk level increases until it gets to 10the highest risk level and most unsafe violation. Weights were assigned on
the basis of priority and category. All violations in the Marcellus Shale were categorized on the basis of severity and potential impact to
the environment. For example, failure to post a permit number at the drilling site was ranked 1, whereas violations that directly
compromise environmental safety were ranked higher (e.g., the discharge of industrial waste into surface water) and were given a
ranking number of 10. The violation severity trends from 20082010 in the Marcellus-shale area are presented in Fig. 3.
There is a 342% increase in the incidence of potentially risky vio- lations from 2008 to 2010, compared with the total violations
com- mitted between 2008 and 2010, which showed a 110% increase. It was identified from the data made available by the
Pennsylvania DEP that operators in the Marcellus shale violated the states oil and gas rules and regulations a total of 2,601 times from 1
January 2008 to 30 November 2010. Of these, 2,083 violations (or 80% of the total violations committed) were potential threats to the
environment.
2012 SPE Drilling & Completion

The original data were processed and grouped by year of inci- dents (20082010); each year has 13 variables. The first variable is
Operators in the Marcellus Shale 65 that is referred to as company in the code. The second to 11th variables, called Severity of
Violations, are the numbers of violations in each rank 1 through 10. These are denoted by r1 to r10 in the SAS code. They are also
called category in the created model or fac- tor determination. The last two variables in the data are Total and Weight. These are
recalculated by use of the SAS code.
Weighted Average of Violations. The violation averages were calculated by use of the weights that are ranked 1 to 10 to separate the
companies into four groups. The average of violations and not the sum of violations is used so that the severity of different viola- tions is
taken into account. For example, 10 violations with rank- ing number 1 are different from one violation with ranking number 10.
The weighted averages were calculated by use of SAS, as follows:
Statistical Objectives of Study. Classify operators on the basis of severity rate (i.e., severity of the violations on the basis of significant values). Develop a statistical code by use of SAS with ANCOVA for predictive modeling and real visualization of the situation.
Previous Studies vs. This Study. In previous studies, the envi- ronmental impacts of the drilling activities were described by use of
violation per well (incident rate), which did not account for the magnitude of impact of the incidents on the Marcellus-shale environment; therefore, the safety levels analyzed were not accurate. In some cases, operators were found to have violated several
times, but the violations were neutral to the Marcellus-shale envi- ronment. This current study highlights the severity of gas-development activities and also corrects misconceptions concerning drilling activities in the Marcellus-shale environment. The main variables
are presented in the following.
Analytical Summary
The total number of violations for the 3 years analyzed was 2,601. From the analysis, the response variables were the number of violations of operators each year, and the factor is the category rank
1 through 10; the covariates were the years (2008 to 2010), and the ANCOVA repeated measure was used for year and
category. There were 65 independent companies for the 3 years, in category rank 1 to 10. By use of the mixed procedure, the total
number of observations read and used was 1,290; the convergence criteria were met after the fourth iteration. The null-model likeli- hood
2
ratio test came up with 6 degrees of freedom, v of 490.85, and p < 0.0001. After the power transformation, the F-test statistic value was
found to be 4.62 for the significant factors and interac- tions, whereas 4.61 was found for the main effects of category and year
interactions. The t values for Category rank 4 to 5 had
the highest t values of 10.71, compared with other higher signifi- cant interactions such as Category rank 7 through 10, t values of
5.75; Category rank 2 through 5, t values of 6.25; Category rank 1

2012 SPE Drilling & Completion

through 5, t values of 5.47; Category rank 7 through 9, t values of

5.04; and Category rank 6 through 10, t values of 4.48.
The importance of this study is emphasized in the need to identify the operations/operators that severely affect the environment. The severity rates (SRs) of violations that specifically relate
to environmental quality were analyzed, and results were com- pared with incident rates (IRs) (Figs. 5 through 7). The SR value is a
competent measure of the intensity of environmental effect in
the area of operation. This study has shown that in 2008, SR was more than IR for 24 operators; in 2009, SR was more than IR for
30 operators; and in 2010, SR was more than IR for 44 operators (illustrated in Figs. 5 through 7). Because IR takes into account
multiple factors that may or may not be related to environment,
health, or safety (EHS), SR depicts the direct effect of every oper- ators violation affecting EHS. Some operators with high IRs
were found to have lower SR; this is significant because those operators had violations unrelated to EHS. The high level of significance of these SRs in the violations is an indication that it is possible to track the pertinent short- and long-term effects of dril- ling
activities on the environment of operation, and this informa- tion can be applied in finding the most-suitable mitigation and control
methods.
Mitigation Measures
Some of the mitigations and control measures are presented in Table 3. The measures as suggested in Table 3, if considered, are useful
plans in minimizing environmental risks, during the devel- opment of natural gas from the shale formation.
Recommendations and Applications
This research work demonstrates how risk-assessment techniques (by use of statistics such as doubly repeated ANCOVA) are effec- tive
in the determination of the severity of impact and incidents
on the basis of the comparative responses (incidents, suggested by violations in this case) of the significant categories. It also shows the
extent of violation contributions to environmental damage and identification of severity rates because of gas-development activ- ities in
the Marcellus-shale area. The DEP could use this model to validate claims by several interest groups on the extent of degrada- tion in
these active areas. If the most significant incidents (ranks 3,
5, 9, 10) that account for 67% of all violations are addressed by appropriate mitigation and control measures, this will minimize the
footprint of E&P activities in the Marcellus-shale gas play. This will also create a positive outlook for operators (minimal damages
to the environment will build more public trust and host commun- ities will be more accommodating to exploration and developmen- tal
activities), a cost benefit (if proper precautions are taken as
suggested in the mitigation plan, operators will not be liable to increasing settlement payments and violation fees, and operations
will not be interrupted) and a minimal risk outcome (if operators are careful enough to avoid violating the DEP s laws, this would
decrease the risk to human health and the ecosystem where they
operate).
The DEP, energy policy makers, safety personnel, administra- tive decision makers, supervisors, and managers could use the
analysis derived in this study to create an operative safety-monitoring program from the inception of the job until abandonment, which would ultimately provide enormous cost savings to opera- tors, and
result in the environment being appropriately preserved. The statistical modeling of operational activities with regard to pollution control
and mitigation would provide an efficient means of managing safety systems on the basis of exploratory-data meth- odology. The ability
of stakeholders (e.g., the DEP, operators, and environmental concern groups) to use quantitative appraisal and quantity and verification
approaches to improve their existing safety plans would help in creating an adequate, effective, and functioning safety culture that
would consequently reduce inci- dents at the workplace and minimize the environmental footprints from fossil-fuel development. Several
operational and administra- tive costs (e.g., pipeline repairs, remediation cost, equipment repair, liability, and downtime costs) could be
curtailed if there is an acute knowledge of the causes, effects, and long-term conse- quences of risk-related activities and there are
appropriate inter- ventions in place to reduce the severity of such. The social license to operate responsibly is important in the overall
interest of opera- tors with regard to keeping the environment in which they operate safe and this is also important for the worlds
economy that cur- rently mainly depends on oil and gas for sustenance. A safety- conscious operator improves its corporate image and
preserves it

2012 SPE Drilling & Completion

sreputation while doing business profitably with an increasing pub- lic trust evident in the dominant shareholder value of the firm.
Conclusions
The Marcellus shale has become an economically important region in the world because of the huge natural-gas deposits sit- ting
some 8,000 ft below the surface. Shale-gas development has brought several changes to the region, including population increase,
infrastructural development, increase in job opportuni- ties, and provision of substantial revenue. However, the environ- mental hazards
presented by this challenging situation of natural- gas extraction from the Marcellus-shale area are enormous. By use of the riskevaluation technique in analyzing the prevailing situation, the major causes of the safety threat in the environment were identified
accurately. There was a 342% increase in inci- dents of potentially risky violations during 20092010, compared with the total violations
committed during the same period. There were 2,601 violations between 1 January 2008 and 30 November
2010; 80% of the total violations committed were potential threats to the environment. Of the 64 companies currently involved in exploration activities in the area, only 27 were identified as significant contributors to environmental risk because of the high values in the risk-ranking numbers of 3, 5, 9, and 10 (Table 2). Because of the
high t values of these categories, it is essential for more attention to be paid to these contributing effects as shown in the level of significance in the violations by operators in a 3-year period. In a
broader perspective, the economic importance of adequate inter- vention in these categories is essential. If these highly significant
categories are effectively taken care of, this would certainly
enhance the usefulness of the safety model developed and conse- quently minimize the environmental footprints from gas-development activities in the Marcellus-shale area. Toward this end, a synergy is achievable between gas production in the Marcellus shale and
keeping the environment safe from pollution, through the
effective adoption of best-management practices (BMPs).
The risk analysis elaborated the severity rate with a total num- ber of 1,344 violations in 2010, including direct discharge of substances into water bodies. This raises concern on the potential impact of the exploration activities on the health and safety of the
residents in the Marcellus Shale area. It is important for operators to consider effective mitigation plans which will reduce these
risks. In view of the recent trends in the extraction of gas in the
Marcellus shale, it is imperative for operators to comply with and readily and adequately implement the BMPs in accordance with
all slated local and state regulations in the Marcellus-Shale area. This fit will not only diminish the risk associated with natural- resource
exploration but also will create a means of safety
whereby the needed energy for human survival is available along- side a pristine environment. The economic importance of the Marcellus-shale drilling is huge and must be balanced out. Enormous attention is required in water management in this area to avoid carrying
out drilling and hydraulic fracturing for short-term benefits at the detriment of environmental preservation in the long run.

2012 SPE Drilling & Completion