The EPA Draft Report of Groundwater
Contamination Near Pavillion, Wyoming:
Main Findings and Stakeholder Responses
Peter Folger
Specialist in Energy and Natural Resources Policy
Mary Tiemann
Specialist in Environmental Policy
David M. Bearden
Specialist in Environmental Policy
January 25, 2012
Congressional Research Service
7-5700
www.crs.gov
R42327
CRS Report for Congress
Pr
epared for Members and Committees of Congress
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Summary
On December 8, 2011, the U.S. Environmental Protection Agency (EPA) issued a draft report on
its investigation of groundwater contamination near the town of Pavillion, Wyoming. Under the
Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA), residents
of Pavillion petitioned EPA, asking the agency to investigate whether groundwater contamination
exists, its extent, and possible sources. Following the petition, EPA began its investigation three
years ago. Although the final report may contain revised or more specific conclusions, the draft
report indicated that EPA had identified certain constituents in groundwater above the production
zone of the Pavillion natural gas wells that are consistent with some of the constituents used in
natural gas well operations, including the process of hydraulic fracturing. In its report, EPA
claimed that its approach to the investigation best supports the explanation that inorganic and
organic compounds associated with hydraulic fracturing have contaminated the aquifer at or
below the depths used for domestic water supply in the Pavillion area. EPA also stated that its
approach indicates that gas production activities have likely enhanced the migration of natural gas
in the aquifer and the migration of gas to domestic wells in the area. EPA did not appear to
conclude that there was a definitive link to a release from the production wells, nor to the
constituents found in domestic wells in shallower parts of the aquifer.
Because the draft report linked groundwater contamination in the deeper portions of the Wind
River Formation aquifer to activities related to hydraulic fracturing during natural gas production
in the area, it raised concerns about hydraulic fracturing practices in general. Organizations
representing portions of the natural gas industry and other stakeholders took issue with some of
the findings in the draft report, and questioned the scientific validity of EPA’s contention that “the
explanation best fitting the data for the deep monitoring wells is that constituents associated with
hydraulic fracturing have been released into the Wind River drinking water aquifer at depths
above the current production zone.”
Congressional Research Service
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Contents
Federal Role at Pavillion.................................................................................................................. 1
EPA Investigation Authority...................................................................................................... 1
Related Federal Public Health Study......................................................................................... 1
Primary Findings of the EPA Draft Report ...................................................................................... 2
Background................................................................................................................................ 2
Detecting Contamination in Groundwater................................................................................. 4
Contaminants in Shallow Groundwater—Surface Pits ....................................................... 4
Contaminants in Deeper Groundwater—Natural Gas Operations and Hydraulic
Fracturing? ....................................................................................................................... 5
Enhanced Migration of Natural Gas?.................................................................................. 9
Summary of EPA’s Reasoning................................................................................................. 11
Stakeholder Responses to the EPA Draft Report ........................................................................... 12
Industry Groups ....................................................................................................................... 12
Environmental Advocacy Groups............................................................................................ 13
Discussion...................................................................................................................................... 14
Tight Sand Gas Versus Shale Gas............................................................................................ 15
Hydraulic Fracturing in Deep Versus Shallow Reservoirs ...................................................... 16
Vertical Wells Versus Horizontal Wells ................................................................................... 16
The Hydraulic Fracturing Process ........................................................................................... 17
Next Steps...................................................................................................................................... 18
Figures
Figure 1. Location of the Wind River Basin, Wyoming .................................................................. 3
Figure 2. Well Construction Showing Casing Extending Through an Aquifer.............................. 11
Appendixes
Appendix. EPA Response Authority and Possible Further Actions ............................................... 19
Contacts
Author Contact Information........................................................................................................... 21
Congressional Research Service
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
n December 8, 2011, the U.S. Environmental Protection Agency (EPA) issued a draft
report on its investigation of groundwater contamination near the town of Pavillion,
O Wyoming.1 This CRS report provides a synopsis of the statutory authority for EPA’s
investigation under the Comprehensive Environmental Response, Compensation, and Liability
Act (CERCLA),2 a summary of the primary findings in the EPA Draft Report, and a brief
discussion of issues raised subsequent to the release of the draft report by proponents and
opponents of the use of hydraulic fracturing for natural gas development. Additionally, this report
identifies the next steps EPA may take regarding this investigation.
Although the EPA Draft Report focused on one specific region where hydraulic fracturing was
employed to enhance the production of natural gas, it has raised concerns about hydraulic
fracturing practices in general, and whether EPA’s findings at Pavillion are more broadly
applicable to other regions of the country.
Federal Role at Pavillion
EPA Investigation Authority
In 2008, citizens of Pavillion submitted a public petition asking EPA to conduct an investigation
of possible contamination of the drinking water aquifer underlying the town. Section 105(d) of
CERCLA provides the authority for any person who is, or may be, affected by a release or
threatened release of a hazardous substance, pollutant, or contaminant to petition the President to
assess the potential hazards to public health and the environment.3 Executive Order 12580
delegated this and other response and enforcement authorities under CERCLA to EPA.4 In 2008,
the agency’s Region 8 office responded to the petition and began the investigation of possible
groundwater contamination underlying Pavillion.
Related Federal Public Health Study
To help inform its investigation of the groundwater underlying the Pavillion site, EPA requested
that the Agency for Toxic Substances and Disease Registry (ATSDR), an agency of the U.S.
Department of Health and Human Services, examine the potential health hazards that may be
associated with contaminants found specifically in private residential well water, but not other
portions of the aquifer. Section 104(i)(4) of CERCLA authorizes EPA (or state or local officials)
to request that the ATSDR provide consultations on potential health issues that may be associated
1 U.S. Environmental Protection Agency, Region 8 and Office of Research and Development, National Risk
Management Research Laboratory, (Draft) Investigation of Ground Water Contamination near Pavillion, Wyoming,
EPA 600/R-00/000, December 2011, http://www.epa.gov/region8/superfund/wy/pavillion/
EPA_ReportOnPavillion_Dec-8-2011.pdf. Hereinafter referred to as the EPA Draft Report.
2 42 U.S.C. §9601 et. seq. For a more in-depth discussion of the authorities of CERCLA than presented in this report,
see CRS Report R41039, Comprehensive Environmental Response, Compensation, and Liability Act: A Summary of
Superfund Cleanup Authorities and Related Provisions of the Act, by David M. Bearden.
3 42 U.S.C. §9605(d). EPA is required to complete a preliminary assessment of a site within 12 months of the
submission of a petition, or to provide an explanation of why an assessment may not be appropriate. EPA determined
that an assessment of the groundwater underlying Pavillion was appropriate based on the observations about the water
quality expressed by the petitioners. EPA Draft Report, p. 1.
4 Executive Order 12580, “Superfund Implementation,” 52 Federal Register 2923, January 23, 1987.
Congressional Research Service
1
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
with the release of a hazardous substance at a specific site.5 In response to EPA’s request under
this authority, the ATSDR issued a Health Consultation for Pavillion in August 2010.6
The ATSDR concluded that exposure to some of the contaminants found in the private residential
well water were at levels that could lead to certain health effects, based on the potential for
exposure relative to the health screening criteria that the ATSDR applied, and that some of the
contaminants (such as methane) could present potential explosive hazards in residences under
certain conditions. The ATSDR recommended that residents use alternate or treated water
supplies, and recommended certain other measures to address potential explosive hazards, such as
ventilating bathrooms while showering.7
It should be emphasized that the ATSDR’s study focused specifically on potential hazards
associated with the private residential well water, whereas the scope of EPA’s site investigation
was broader in terms of identifying and characterizing contaminants across the aquifer more
widely and at greater depths. The ATSDR’s finding of the presence of potential hazards was
limited to the private residential well water itself, at shallower depths common to most domestic
wells, and not the greater depths of natural gas production wells. The distinction between
chemical constituents found at shallow depths in the aquifer and those found in deeper portions is
discussed below.
Primary Findings of the EPA Draft Report
Background
The Pavillion gas field lies within the Wind River Basin, a deep sedimentary basin extending
across a large area of central Wyoming and bounded on the north and southwest by upfolded and
faulted mountain ranges. (See Figure 1.) The Wind River Formation, an accumulation of
sandstone, conglomerate, shale, and mudstone, is the major source of drinking water for domestic
and public-supply uses in the Wind River Basin.8 The Wind River Formation varies in thickness,
and extends from the ground surface to as deep as 3,400 feet in the Pavillion gas field area.9
Natural gas is produced from wells drilled into the Wind River Formation, and from deeper wells
drilled into the Fort Union Formation, which lies directly underneath the Wind River Formation.
The most productive zone of natural gas extraction is from the bottom of the Wind River
5 42 U.S.C. §9604(i)(4).
6 U.S. Department of Health and Human Services, Agency for Toxic Substances and Disease Registry, Health
Consultation: Evaluation of Contaminants in Private Residential Well Water at Pavillion, Wyoming, Fremont County,
August 31, 2010, available on the agency’s website: http://www.atsdr.cdc.gov/hac/PHA/Pavillion/
Pavillion_HC_Well_Water_08312010.pdf.
7 Subsequent to the ATSDR’s findings, the governor of Wyoming directed the Wyoming Water Development
Commission in September 2010 to study public water supply options for Pavillion. The commission completed its
study in October 2011. The study focused on water supply options for the residents of Pavillion to ensure the safety of
the supplies, but did not further investigate the groundwater contamination nor potential sources of contaminants across
the aquifer. The study, Pavillion Area Water Supply Level I Study: Final Report, October 2011, is available on the
Wyoming Water Development Commission’s website: http://wwdc.state.wy.us/agency_publications/
PavillionWaterSupplyLl_2011.pdf.
8 Richard L. Daddow, Water Resources of the Wind River Indian Reservation, Wyoming, U.S. Geological Survey,
Water Resources Investigations Report 95-4223, Cheyenne, WY, 1996, p. 21.
9 EPA Draft Report, p. 2.
Congressional Research Service
2
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Formation, although hydraulic fracturing to enhance gas production has occurred at locations as
shallow as 1,220 feet below ground surface, according to the EPA Draft Report.10
Figure 1. Location of the Wind River Basin, Wyoming
(showing the location of the Pavillion Gas Field)
Source: U.S. Environmental Protection Agency, Region 8 and Office of Research and Development, National
Risk Management Research Laboratory, (Draft) Investigation of Ground Water Contamination near Pavillion, Wyoming,
EPA 600/R-00/000, December 2011, http://www.epa.gov/region8/superfund/wy/pavillion/
EPA_ReportOnPavillion_Dec-8-2011.pdf. Modified by CRS.
The EPA sampled residential wells, stock wells, shallow monitoring wells, and two municipal
wells. The domestic wells range in depth from approximately 20 feet to nearly 800 feet, and the
two municipal wells are 505 and 515 feet deep.11 The shallow monitoring wells were
approximately 15 feet deep. According to the EPA Draft Report, the early phases of the
investigation detected the presence of methane and diesel-range organic chemicals in some of the
10 EPA Draft Report, p. 2.
11 EPA Draft Report, Table A1, pp. A2-A4.
Congressional Research Service
3
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
deeper domestic wells, which prompted EPA to install two deep monitoring wells in June 2010.12
EPA stated that the purpose of installing two deep monitoring wells—one at 785 feet and the
second at 980 feet—was to differentiate potentially deep sources from potentially shallow sources
of contamination.13 Shallow sources of contamination were thought to be related to leakage from
surface pits used for storage and disposal of drilling wastes and produced and flowback water.
Potential deeper sources were thought to be related to gas production, which would include
drilling and hydraulic fracturing, as well as actual gas production.
Detecting and distinguishing between potentially shallow and potentially deep sources of
groundwater contamination lies at the heart of the primary findings in the EPA Draft Report.
Whether the report clearly links groundwater contamination to drilling or hydraulic fracturing
activities at depth has been the source of relatively heated commentary by proponents and
opponents of the use of hydraulic fracturing for natural gas development. The primary findings in
the report and examples of reactions and commentary by stakeholders are discussed below.
Detecting Contamination in Groundwater
Contaminants in Shallow Groundwater—Surface Pits
According to the EPA Draft Report, the objective of the EPA investigation was to determine the
presence of groundwater contamination above the Pavillion gas field, and to the extent possible
identify the source of the contamination.14 The investigation identified a suite of contaminants in
samples from shallow monitoring wells—wells that monitor the upper portions of the Wind River
aquifer. The contaminants identified in the shallow portions of the aquifer included benzene,
xylenes, gasoline-range organics (GROs), and diesel-range organics (DROs).15 According to the
report, at least 33 surface pits were likely sources for the contaminants detected in shallow
groundwater: “detection [of these contaminants] in ground water samples from shallow
monitoring wells near pits indicates that pits are a source of shallow ground water contamination
in the area of investigation.” 16 The pits were used for disposal of drilling cuttings, hydraulic
fracturing flowback, and water produced from the formation.
12 Diesel-range organics (DROs) are a group of compounds similar to, and including, diesel fuel. DROs include, for
example, phenols, phthalate esters, kerosene, and home heating oil.
13 EPA Draft Report, p. 5. The depths of the monitoring wells refer to the bottom of the screened interval for a well.
The screened interval is the portion of the well where the well casing is not solid steel, but consists of a stainless steel
mesh that allows water from a productive layer in the aquifer to flow into the monitoring well. For example, the
screened interval for the shallower monitoring well extends from 765 to 785 feet; and the screened interval for the
deeper well extends from 960 to 980 feet.
14 EPA Draft Report, p. 33.
15 Petroleum fuels and oils are complex mixtures of many hydrocarbon compounds. Testing can be done for specific
chemicals of concern, such as benzene, and for chemically similar compounds to help identify possible sources of
contamination. Gasoline-range organics (GROs) comprise a group of hydrocarbon compounds structurally similar to,
and including, gasoline. Diesel-range organics (DROs), as discussed above, are a group of compounds similar to, and
including, diesel fuel. DROs contain longer carbon chains than GROs, and DROs include, for example, phenols,
phthalate esters, kerosene, and home heating oil. For analytical purposes, test methods are available to identify and
measure the concentration of different compounds within the GRO and DRO ranges.
16 EPA Draft Report, p. 33. Although now banned in many states, unlined pits and lagoons have long been used to
dispose of wastewater associated with oil and gas production.
Congressional Research Service
4
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
The Draft Report further noted that EPA is a member of a stakeholder group working with the gas
field operator—Encana Oil & Gas Inc., a subsidiary of the Canadian Encana Corporation—to
“determine the areal and vertical extent of shallow ground water contamination caused by these
pits.”17 EPA added that Encana is currently engaged in investigating and remediating several pit
areas. Encana has contributed to the cost of furnishing alternate supplies of drinking water to
some Pavillion citizens while its investigation continues as part of the stakeholder group.18
Encana acquired the natural gas field and its infrastructure in 2004; however, drilling for natural
gas began in the 1960s and the surface pits were excavated prior to 2004.19
The EPA Draft Report does not discuss the shallow groundwater contamination in much detail,
and it does not indicate that the source of the contaminants in shallow groundwater is anything
other than the surface pits. Reactions to the report and commentary by stakeholders also have not
focused on the shallow groundwater issues, or on the surface pits as likely sources of
contaminants. The focus of the EPA Draft Report and the issues raised by proponents of natural
gas development and hydraulic fracturing concern the detection and source of contaminants in the
deeper portions of the aquifer. Domestic water wells in the Pavillion area generally use
groundwater from the shallower portions of the aquifer.
Contaminants in Deeper Groundwater—Natural Gas Operations and
Hydraulic Fracturing?
The EPA Draft Report acknowledged that “[d]etection of contaminants in ground water from deep
sources of contamination (production wells, hydraulic fracturing) was considerably more
complex than detection of contaminants from pits necessitating a multiple lines of reasoning
approach common to complex scientific investigations.”20 The Draft Report further explained
that, “[w]hile each individual data set or observation represents an important line of reasoning,
taken as a whole, consistent data sets and observations provide compelling evidence to support an
explanation of data.”21 According to the report, this approach led to its primary finding, “that
constituents associated with hydraulic fracturing have been released into the Wind River drinking
water aquifer at depths above the current production zone.”22
The first set of “lines of reasoning” described in the report refers primarily to chemical
constituents detected in the two deep monitoring wells the EPA installed during June 2010.
Monitoring Well 1 (MW01) was screened (open to the aquifer) between 765 and 785 feet below
ground surface; Monitoring Well 2 (MW02) was screened between 960 and 980 feet below
ground surface. For comparison, the domestic wells sampled during the EPA investigation ranged
between 20 and 800 feet deep, and the two municipal wells included in the study were 505 and
515 feet below the ground surface. However, EPA also notes in the report the absence of baseline
17 EPA Draft Report, p. 33.
18 Encana Oil & Gas, News Release, Why Encana Refutes U.S. EPA Pavillion Groundwater Report, December 12,
2011, http://www.encana.com/news-stories/news-releases/details.html?release=632327. The stakeholder group includes
Encana, the Wyoming Department of Environmental Quality, the Wyoming Oil and Gas Conservation Commission,
Wyoming Geological Survey, Wyoming State Engineers Office, and the Department of the Interior’s Bureau of Land
Management.
19 EPA Draft Report, p. 1.
20 EPA Draft Report, p. 33.
21 EPA Draft Report, p. 33.
22 EPA Draft Report, p. 33.
Congressional Research Service
5
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
groundwater monitoring data that could indicate groundwater conditions prior to gas production
in the area.
The EPA Draft Report also provided a second set of “lines of reasoning” for supporting the
agency’s conclusion that “[a]lthough some natural migration of gas would be expected above a
gas field such as Pavillion, data suggest that enhanced migration of gas has occurred to ground
water at depths used for domestic water supply and to domestic wells.”23 These “lines of
reasoning” refer to chemical data from other wells, to the length of casing and the presence or
absence of cement in gas production wells, and to the nature and timing of citizens’ complaints
about taste and odor problems with their drinking water.
A brief description of the “lines of reasoning” that led EPA to its explanation for the contaminants
in deeper groundwater follows.24
High pH Values
The EPA Draft Report cited “unusual and unexpected” pH values measured in both monitoring
wells.25 The pH values ranged from 11.2 to 12.0. (A pH of 12 is unusually high for most natural
waters, and is approaching the caustic or strongly pH range.) 26 The Draft Report noted that pH
values in domestic wells ranged between 6.9 and 10, indicating that groundwater measured in the
deep monitoring wells was between 10 and 100 times more alkaline than the most alkaline
domestic well sampled during the investigation.27 In the report, EPA also cited geochemical
modeling results indicating that the addition of a strong base, such as potassium hydroxide
(KOH), to groundwater of the Pavillion aquifer at depths of 328 feet or more would increase pH
values significantly. The EPA Draft Report noted that KOH was used in fracking operations in the
Pavillion gas field as a cross-linker and in a solvent, and suggested that the addition of a strong
base (such as KOH) was “the causative factor for elevated pH in the deep monitoring wells.”28
Elevated Potassium and Chloride Concentrations
The EPA Draft Report stated that the inorganic chemistry of the groundwater measured from deep
monitoring wells is distinctive from the groundwater in domestic wells sampled in the study and
from the expected composition of groundwater in the Wind River Formation. In particular, the
23 EPA Draft Report, p. 37.
24 The last section of this report discusses several of the arguments raised to date against some of the individual lines of
reasoning and against EPA’s tentative overall conclusion that the presence of petroleum hydrocarbons and other
chemical compounds in the ground water “is consistent with migration from areas of gas production” where hydraulic
fracturing is taking place.
25 EPA Draft Report, p. 20.
26 A pH of less than 7.0 is considered acidic, while a pH of greater than 7.0 is considered basic (alkaline); a pH of 7.0 is
defined as “neutral.” pH is reported on a log scale, so that each pH unit represents a 10-fold change in concentration.
For example, a pH of 10 is 10 times more alkaline than a pH of 9, and 100 times more alkaline than a pH of 8.
27 These values for domestic wells were reported in the EPA Draft Report text on p. 33; however, Table A2a indicates
that domestic wells contained pH values as high at 10.47 (sample PGDW32), and the lowest pH value for a deep
monitoring well was 11.24 (MW01). Using the numbers reported in Table A2a, the pH of MW01 was 5.9 times as
alkaline as sample PGDW32.
28 EPA Draft Report, p. 20. A cross-linker is added to fracking fluids to increase the viscosity of the fluid in order to
transport the proppant, commonly sand, more effectively into the induced fractures. (Proppants hold open the fractures
and allow gas to flow to the well.)
Congressional Research Service
6
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
report cited elevated concentrations of potassium and of chloride. According to the report,
potassium levels in the monitoring wells were between 8.2 and 18.3 times the mean value of
levels observed in domestic wells. Chloride levels in MW02 were 18 times the mean value for
chloride concentrations measured in domestic wells.29 (Chloride values in MW01, however, were
approximately 23 milligrams per liter, less than the mean value for domestic wells of 25.6
milligrams per liter.)30 It is difficult to ascertain from the report whether the higher potassium and
chloride levels represent a range of natural variability in the deeper portions of the aquifer, or
whether they are related to drilling and hydraulic fracturing activities.
The report cited information from well completion reports and material safety data sheets
(MSDSs) for each of the wells indicating the use of chemicals containing potassium and chloride
in fracture fluids. Namely, the report noted the use of potassium chloride, potassium metaborate,
potassium hydroxide, and ammonium chloride in foam jobs and as cross-linkers in fracture
fluids.31 However, the report did not include any information linking the use of these chemicals
with site-specific hydraulic fracturing jobs, nor did it cite specific groundwater pathways from
hydraulic fracturing to the monitoring wells. The report also considered alternate explanations for
elevated potassium and chloride levels, such as contamination by drilling fluids and additives
used in constructing the monitoring wells, contamination from well completion materials, and
contamination from surface soils. But in its description of how the wells were constructed and
how the materials were handled, EPA did not state that these alternative explanations were
responsible for elevated potassium and chloride levels in the monitoring wells.
Detection of Synthetic Organic Compounds
During its investigation, EPA detected several synthetic organic compounds in water samples
taken from MW01 and MW02. The synthetic organic compounds would not be expected to occur
naturally in groundwater. These compounds included isopropanol, diethylene glycol, and
triethylene glycol. The EPA Draft Report noted that these three compounds were used in
hydraulic fracture fluids, as a foaming agent and in solvents, according to well completion reports
and MSDSs.32 EPA reported that tert-butyl alcohol was also detected in MW02. Tert-butyl alcohol
is a known breakdown product of methyl tert-butyl ether, or MTBE, a gasoline additive used to
raise the oxygen content of the fuel. It is also a breakdown product of tert-butyl hydroperoxide, a
gel breaker used in hydraulic fracturing fluids. Tert-butyl hydroperoxide was not listed on MSDSs
or on well completion logs, according to the EPA Draft Report. However, the report added that
tert-butyl alcohol is not expected to occur naturally in groundwater, and its source in Pavillion
groundwater remains unresolved.
Detection of Petroleum Hydrocarbons
The EPA Draft Report stated that a number of petroleum hydrocarbons were detected in
groundwater in wells MW01 and MW02. These compounds included benzene, toluene,
ethylbenzene, and xylene (BTEX), trimethylbenzenes, GROs, DROs, and napthalene. The report
29 EPA Draft Report, p. 34. As indicated earlier, EPA noted in the report the absence of baseline groundwater
monitoring data that could indicate groundwater conditions prior to gas production in the area.
30 EPA Draft Report, Table A2a.
31 EPA Draft Report, p. 34.
32 EPA Draft Report, Table 4.
Congressional Research Service
7
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
noted that compounds listed on MSDSs that were used in hydraulic fracturing solutions contained
the petroleum hydrocarbon constituents listed above. For example, the report stated, MSDSs
indicate that diesel fuel was used in a guar polymer slurry; an aromatic solvent that was typically
a BTEX mixture was used as a breaker; and other compounds were used in different components
comprising the suite of chemicals that make up a hydraulic fracture fluid.33
Breakdown Products of Organic Compounds
The EPA Draft Report stated that more organic chemicals were detected at higher concentrations
in the deeper monitoring well (MW02), whereas breakdown products of those organic chemicals
were detected at higher concentrations in the shallower well (MW01).34 Examples of breakdown
products found in these wells included acetate and benzoic acid, which can be formed from the
breakdown of BTEX and glycols. The report cited the occurrence of flowing stock wells as
evidence of an upward hydraulic gradient in the study area,35 which the report suggested concurs
with the presence of enriched breakdown products in shallower, downgradient monitoring well
MW01. In other words, the report suggested that groundwater containing organic compounds
such as BTEX and glycols would travel in an upward direction, and during the course of that
travel those compounds would break down, or degrade, into acetate and benzoic acid.
Well Design and Integrity of Gas Production Wells36
The EPA Draft Report stated that the design and integrity of gas production wells were possibly
“one causative factor in deep ground water contamination at this site.”37 The report noted several
components of well design and integrity that could have been involved: (1) the surface casing of
most production wells did not extend below the deepest domestic wells; (2) there was little
vertical separation between the uppermost zones that were hydraulically fractured and the deepest
domestic wells; and (3) there was an absence of cement, or only sporadic bonding between the
cement, well casing, and formation, in several production wells. Typically, cement fills the gap
between the outside of the well casing and the formation to prevent any leakage of fluids along
the outside of the wellbore into an aquifer. The EPA investigation relied on geophysical logs of
the production wells to infer that in many instances cement was lacking along portions of the
33 EPA Draft Report, pp. 35-36.
34 EPA Draft Report, p. 36.
35 A flowing well is also known as an artesian well, in which the groundwater in the aquifer is at a sufficient pressure to
flow naturally to the land surface without requiring pumping. In such cases, the direction of groundwater flow, or
hydraulic gradient, is from the deeper parts of the aquifer towards the shallower parts of the aquifer.
36 Oil and gas production on private and state lands is regulated by the states. The Wyoming Oil and Gas Conservation
Commission has responsibility for administering the oil and gas rules and related permitting, inspection, and
enforcement activities. The state revised its rules effective September 15, 2010. Revisions include requirements for
directional drilling reporting and certification, and expanded requirements for well stimulation (such as hydraulic
fracturing). The well stimulation rules address well integrity, casing setting depths, and casing design and cementing;
protection of utilizable groundwater; disclosure of hydraulic fracturing fluid contents and concentrations; and
management of recovered fluids. The rules now require surface casing to be run to a depth below known or estimated
utilizable groundwater, and to specified depths below water wells. Operators are required to provide detailed
information regarding the fracturing process, including the source of water and/or trade name fluids, type of proppants,
and estimated pump pressures. After a treatment is complete, the operator must provide fracturing data and production
results (Wyo. Rules and Regs. Oil Gen §§3-8, 22, 45, and elsewhere). Also, the state recently revised its rules
governing water well construction.
37 EPA Draft Report, p. 37.
Congressional Research Service
8
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
wellbore or that sporadic bonding existed just above the zones of hydraulic fracturing. The
absence of cement or the sporadic bonding of some portions, inferred by EPA from the
geophysical logs, implies that fluids could have leaked from the fractured intervals up along those
zones to the aquifer above.
Excursion of Fracture Fluids from Sandstone Units and Along the Wellbore
A lithologic barrier, such as a thick layer of impermeable shale, would typically prevent or limit
the amount of natural gas that would seek to migrate from the gas-filled sandstone lenses upward
toward the surface.38 The EPA Draft Report suggested that the absence of a lithologic barrier
above the gas production zone, such as a laterally continuous shale layer, meant that gas might
have migrated upward “in the event of excursion from fractures.”39 Similarly, if fluid leaked
vertically from hydraulically induced fractures in thin sandstone lenses, it could also have
migrated laterally to nearby wellbores, and then travelled vertically upward along the wellbore if
cement were lacking or if the cement was only sporadically bonded to the well casing and
formation, according to the report.
Enhanced Migration of Natural Gas?
In addition to the seven “lines of reasoning” summarized above, the EPA Draft Report also
claimed that “data suggest that enhanced migration of gas has occurred to ground water at depths
used for domestic water supply and to domestic wells.”40 The report noted that some natural
migration of gas would be expected above the gas field at Pavillion. However, the report listed a
second set of “lines of reasoning” to support the interpretation that hydraulic fracturing and gas
development activities allowed gas and other constituents to migrate into the aquifer where they
would not have if gas development had not taken place.
Isotopic Data
Analysis of carbon isotopes can often be used to identify the source of organic compounds. The
EPA Draft Report pointed to analyses of carbon isotopes indicating that the methane found in
monitoring wells is similar to the methane found in production wells. The isotopic data indicate
that the methane gas is “thermogenic,” derived from the thermal breakdown of organic matter
under pressure in deeper source rocks. Thermogenic methane is distinguished from “biogenic”
methane, which is produced by the breakdown of organic material by organisms called
methanogens. Biogenic methane typically occurs close to the earth’s surface (e.g., methane gas in
landfills is biogenic) and is thus distinguished from methane associated with oil and gas
operations. The EPA Draft Report suggested that the patterns indicated by carbon isotope data
support the hypothesis that organic compounds in the study area migrated upward from depth.
38 Sandstone lenses refer to the intervals of sandstone that are discontinuous in the horizontal direction (i.e., are not
long, continuous layers of sandstone). The lenses of sandstone are interbedded with other lithologies, such as shale, in
both a vertical and horizontal direction.
39 EPA Draft Report, p. 37.
40 EPA Draft Report, p. 37.
Congressional Research Service
9
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Proximity of Methane in Domestic Wells to Production Wells
The EPA Draft Report stated that levels of dissolved methane in domestic wells generally
increase in wells closest to gas production wells in the Pavillion study area.41 The report said that
methane was not detected in domestic water wells that had two or fewer production wells within
approximately 2,000 feet (with the exception of two domestic wells where methane was
detected).
Methane Concentrations Highest Near MW01
The EPA Draft Report observed that methane concentrations were highest in samples in an area
encompassing MW01 and two domestic wells labeled PGDW30 and PGDW05 (shown in Figure
5 on p. 6 of the EPA Draft Report). The report noted that high levels of methane were found in
well PGDW30 at a depth of 260 feet, much shallower than MW01 at 784 feet. The report also
stated that a blowout occurred during gas drilling in 2005 at a depth of 520 feet in a well adjacent
to well PGDW05. The report cited data from a mud-gas log conducted in 1980—prior to most of
the gas production activities—in a well nearly 1,000 feet from where the blowout occurred that
did not indicate the presence of natural gas. From that log, EPA inferred that natural gas was not
present at depths shallower than 1,000 feet in the area where the blowout occurred prior to natural
gas development.
Shallow Surface Casing, Lack of Cement, Sporadic Bonding
The EPA Draft Report noted that surface casing of gas production wells does not extend deeper
than the maximum depth of domestic wells in the Pavillion study area (with the exception of two
production wells). In other words, portions of nearly all the production wells were uncased at the
same depth in the aquifer where the deepest domestic wells obtained their water. EPA asserted
that the shallow surface casing, combined with data suggesting lack of cement or sporadic cement
bonding between production casing and the formation (discussed above), would facilitate upward
migration of natural gas from deeper gas production zones toward shallower domestic wells.
Figure 2 is a diagram of a well showing a typical array of casing types extending from the ground
surface downwards. It shows the casing extending through and beneath the aquifer. According to
the EPA Draft Report, most gas wells in the Pavillion field were not constructed with casing
extending completely through the deepest portion of the aquifer.
41 EPA Draft Report, p. 38.
Congressional Research Service
10
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Figure 2. Well Construction Showing Casing Extending Through an Aquifer
Source: Adapted from the American Petroleum Institute, Hydraulic Fracturing, http://www.api.org/policy/
exploration/hydraulicfracturing/upload/HYDRAULIC_FRACT_ILLUSTRATION_121609.pdf . Modified by CRS.
Notes: Not to scale. Shown for illustration purposes, not representative of wells in the Pavillion field.
Citizen Complaints
Last, the EPA Draft Report stated that citizen complaints about odor and taste problems with their
well water that began concurrently with or after hydraulic fracturing were “internally consistent,”
but no baseline data for domestic wells are available for comparison. Baseline data would help
determine past levels of gas flux to domestic wells. Nevertheless, the report stated that “[c]itizens
complaints often serve as the first indication of subsurface contamination and cannot be
dismissed without further evaluation, particularly in the absence of routine ground water
monitoring prior to and during gas production.”42 Furthermore, Section 105(d) of CERCLA
obligated EPA to perform the site investigation once potentially affected citizens submitted the
petition, unless the agency had determined that the investigation was inappropriate and had
provided the citizens with an explanation for such a determination.
Summary of EPA’s Reasoning
In summary, EPA claimed that its “lines of reasoning” approach best supports the explanation that
inorganic and organic compounds associated with hydraulic fracturing have contaminated the
aquifer at or below the depths used for domestic water supply in the Pavillion area. EPA also
stated that its approach indicates that gas production activities have likely enhanced the migration
of natural gas in the aquifer and the migration of gas to domestic wells in the area.
42 EPA Draft Report, p. 39.
Congressional Research Service
11
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Stakeholder Responses to the EPA Draft Report
Industry Groups
Encana Oil & Gas, Inc.
On December 12, 2011, Encana Oil & Gas (USA) Inc. issued a press release in which the
company disagreed with the preliminary conclusions of the EPA Draft Report.43 (Encana Oil &
Gas Inc. acquired the Pavillion gas field in 2004 and drilled 44 wells between 2004 and 2007.) In
the press release, Encana asserted that EPA’s data align with previous testing done by Encana and
do not show any impacts to domestic wells from oil and gas development. Encana further asserted
that EPA’s findings that compounds used in hydraulic fracturing have contaminated Pavillion
groundwater “are conjecture, not factual and only serve to trigger undue alarm.”
Encana’s press release raised several issues that the company felt cast doubt on the conclusions of
the EPA Draft Report:
• The Pavillion area has a “unique geology and hydrology.”
• Previous reports have indicated poor water quality in the Pavillion aquifer.
• EPA’s two deep monitoring wells were drilled into a natural gas reservoir and
detected components of natural gas, which is not unexpected, according to the
company.
• The chemical results from the deep monitoring wells are “radically different than
those in domestic water wells ... thereby showing no connection.”
• Several of the manmade chemicals detected in the two deep monitoring wells
were not detected in other wells sampled, but some were detected in quality
control samples. In the press release, Encana states that this indicates problems
with EPA’s methodology in drilling and sampling.
• The press release stated that EPA’s results from the investigation do not exceed
state or federal drinking water quality standards for any constituent related to oil
and gas development.
In its press release, Encana called on EPA and other government officials to subject its data to
independent third-party review. In announcing the opportunity for public comment, EPA had
stated its intention to convene an independent panel of scientific experts for external peer review
in addition to the review of any comments that may be submitted by members of the public.44
43 Encana Oil & Gas, Inc., press release, “Why Encana Refutes U.S. EPA Pavillion Groundwater Report,” December
12, 2011, http://www.encana.com/news-stories/news-releases/details.html?release=632327.
44 Environmental Protection Agency, “Draft Research Report: Investigation of Ground Water Contamination Near
Pavillion, Wyoming,” 76 Federal Register 77829, December 14, 2011.
Congressional Research Service
12
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Energy in Depth
Energy in Depth (EID) is an outreach campaign started by the Independent Petroleum Association
of America in 2009 to promote the development of U.S. onshore energy resources.45 In December
2011, EID released a set of questions about the EPA Draft Report, some of which echo concerns
voiced by Encana Oil & Gas, such as why were chemical results from the deep monitoring wells
different from those found in the domestic water wells.46 The questions touched on whether
chemicals used by EPA in drilling its monitoring wells may have affected the results of sampling
the deep groundwater. The group also raised the issue that high levels of potassium and chloride
have been found previously in the Pavillion area, and that high levels found in the monitoring
wells may reflect background water quality and natural variations in groundwater flow or
composition.
The Petroleum Association of Wyoming
On December 9, 2011, the Petroleum Association of Wyoming issued a press release also raising
concerns with the EPA Draft Report.47 The press release stated concerns similar to those raised by
Encana Oil & Gas, Inc. and by EID about the deep monitoring wells being drilled into gas-
bearing zones, the differences between compounds found in the deep monitoring wells and in
domestic water wells, and quality assurance issues with EPA’s drilling and testing.
Environmental Advocacy Groups
Natural Resources Defense Council
A commentator from the Natural Resources Defense Council (NRDC)48 pointed to the EPA Draft
Report’s findings to underscore the NRDC advocacy position that
wells that will be hydraulically fractured be located in a geologically suitable location such
that a suitable confining zone is present, any potential contamination pathways—including
improperly constructed or abandoned wells—must be identified and remediated, and
properly constructed wells, baseline testing, and site characterization are crucial to
preventing contamination of USDWs [underground sources of drinking water].49
Another NRDC commentator also cited the results of the report to support the claim that many
factors are at play in hydraulic fracturing, any one of which “can go wrong.”50 The commentator
45 See “What’s EID?” at http://www.energyindepth.org/whats-eid/.
46 Energy in Depth, “*Update VI* Six—Actually, Seven—Questions for EPA on Pavillion,”
http://www.energyindepth.org/six-questions-for-epa-on-pavillion/.
47 The Petroleum Association of Wyoming, press release, Petroleum Association of Wyoming States Serious Concerns
with EPA’s Unsubstantiated and Reckless Claims, http://www.pawyo.org/PAW_News%20Release_12082011.pdf.
48 The Natural Resources Defense Council is a not-for-profit, tax-exempt environmental advocacy organization. See
http://www.nrdc.org/about/.
49 Natural Resources Defense Council, Briana Mordick’s Blog, Groundwater in Pavillion, WY Contaminated by
Hydraulic Fracturing Through Multiple Subsurface Pathways, December 9, 2011, http://switchboard.nrdc.org/blogs/
bmordick/groundwater_in_pavillion_wy_co.html.
50 Natural Resources Defense Council, Amy Mall’s Blog, “New EPA Report Ties Hydraulic Fracturing to Groundwater
Contamination,” December 8, 2011, http://switchboard.nrdc.org/blogs/amall/new_epa_report_ties_hydraulic.html.
Congressional Research Service
13
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
stated that much stronger rules are needed and that is why NRDC supports federal regulation of
fracking under the Safe Drinking Water Act.51
Environmental Defense Fund
A commentator from the Environmental Defense Fund (EDF) echoed remarks in the NRDC
critique that the “draft report is Exhibit A on why stronger regulation and enforcement is
necessary if the general public is EVER going to believe that shale gas development is a safe
source of natural gas.”52
Pro Publica
An article published by Pro Publica, an independent nonprofit news service, stated that findings
from the EPA Draft Report “could be a turning point in the heated national debate about whether
contamination from fracking is happening, and are likely to shape how the country regulates and
develops natural gas resources in the Marcellus Shale and across Appalachian states.” The article
also stated that some of the findings in the report contradict what the drilling industry has argued
about why fracking is safe. The article said that those industry arguments are “that hydrologic
pressure would naturally force fluids down, not up; that deep geologic barriers provide a
watertight barrier preventing the movement of chemicals towards the surface; and that the
problems with the cement and steel barriers around gas wells aren’t connected to fracking.” 53
Discussion
On December 14, 2011, EPA began a 45-day public comment period for the Draft Report with a
closing date of January 27, 2012.54 On January 18, EPA announced that the agency would accept
comments through March 12, 2012. 55 Additionally, the report will be peer-reviewed by a panel of
independent scientists. On January 17, EPA published a 30-day notice inviting public nominations
of scientific experts to be considered as peer reviewers for the external review of the Draft
51 The Safe Drinking Water Act (SDWA) establishes the national program for protecting “underground sources of
drinking water” (USDWs) by limiting, through regulation, underground injection that could contaminate usable
aquifers. SDWA §1421 directs the EPA Administrator to issue regulations for state programs, and mandates that the
EPA rules “contain minimum requirements for programs to prevent underground injection that endangers drinking
water sources.” UIC provisions, as amended, are contained in SDWA Part C, §§1421-1426; 42 U.S.C. §§300h-300h-5.
The Energy Policy Act (EPAct) of 2005 (P.L. 109-58, §322), amended the SDWA to exempt from the definition of
underground injection the injection of fluids or propping agents (other than diesel fuel) for hydraulic fracturing
purposes (42 U.S.C. §300h(d)). For a discussion of hydraulic fracturing regulatory proposals and issues, see CRS
Report R41760, Hydraulic Fracturing and Safe Drinking Water Act Issues, by Mary Tiemann and Adam Vann.
52 Environmental Defense Fund, Mark Brownstein, EPA’s Pavillion, “WY Groundwater Contamination Study A Wake
Up Call,” December 8, 2011, http://blogs.edf.org/energyexchange/2011/12/08/epas-pavillion-wy-groundwater-
contamination-study-a-wake-up-call/.
53 Abrahm Lustgarten and Nicholas Kusnetz, “Feds Link Water Contamination to Fracking for the First Time,” Pro
Publica, December 8, 2011, http://www.propublica.org/article/feds-link-water-contamination-to-fracking-for-first-time.
54 Environmental Protection Agency, “Draft Research Report: Investigation of Ground Water,” 76 Federal Register
77829-77830, December 14, 2011.
55 See http://www.epa.gov/region8/superfund/wy/pavillion/#1.
Congressional Research Service
14
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Report.56 EPA intends to convene a peer review panel in March or April, and to issue a final
report in 2012.57
Although the final report may contain revised or more specific conclusions, the Draft Report
indicates that EPA identified certain constituents above the production zone of the natural gas
wells that are consistent with some of the constituents used in the well operations. EPA did not
appear to conclude that there was a definitive link to a release from the production wells, nor
to the constituents found in the domestic wells in the shallower portion of the aquifer. Absent
such a link, EPA also did not conclude in its Draft Report that the constituents found in the
aquifer were caused by a specific release that may pose a threat to human health or the
environment at the Pavillion site.58
Judging by a preliminary scan of public comments made by stakeholders, some of which are
described above, it is likely that proponents and opponents of hydraulic fracturing will continue
to disagree over the EPA Draft Report’s main conclusions linking hydraulic fracturing chemicals,
and perhaps the hydraulic fracturing process specifically, with groundwater contamination in the
Pavillion area. Also, there will likely be continued efforts by critics of hydraulic fracturing to
generalize the EPA Draft Report’s findings to regions where hydraulic fracturing is used to
develop other natural gas resources, such as the Marcellus Shale in the Northeast, the Barnett
Formation in Texas, and the Bakken Formation in North Dakota. However, the geology and
hydrology of each region differs. The differences in geology and hydrology could make a
significant difference in the likelihood of contaminating drinking water aquifers from hydraulic
fracturing and from other natural gas development activities. The overall process of hydraulic
fracturing and of exploration and production of natural gas, however, is broadly similar
irrespective of region. A few of the important similarities and differences between the Pavillion
region and other gas-producing regions are described below, with the intention of providing some
context for evaluating future arguments for and against generalizing results from the EPA Draft
Report more broadly.
Tight Sand Gas Versus Shale Gas
The Pavillion field is known as a tight sand gas field. Natural gas is extracted from sandstone
lenses in the Wind River Formation and in the underlying Fort Union Formation. The sandstone
lenses are interbedded with less permeable rocks, such as shales and mudstones. The natural gas
did not originate in the sandstone lenses, but was likely formed in deeper and older rocks and then
migrated into the sandstone lenses. The sandstone lenses, therefore, constitute the reservoir for
natural gas, but not the source. The gas remains trapped in the sandstone reservoirs because the
surrounding rocks are relatively impermeable to flow and keep the gas within the sandstone
lenses.
56 U.S. Environmental Protection Agency, “Request for Nominations for Peer Reviewers for the Draft Research Report
Entitled, ‘Investigation of Ground Water Contamination Near Pavillion, WY,’” 77 Federal Register 2292, January 17,
2012.
57 Telephone conversation with Pamela Janifer, EPA, Office of Congressional and Intergovernmental Relations,
January 18, 2012.
58 As such, EPA would not appear to be required under Section 105(d) of CERCLA at this juncture to evaluate the site
to determine its eligibility for listing on the NPL under the Hazard Ranking System and whether cleanup may be
warranted under the Superfund program.
Congressional Research Service
15
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Tight gas sandstones generally are defined as unconventional gas deposits because they generally
have lower permeability than other types of sandstones in conventional deposits. Unconventional
gas deposits require enhanced recovery techniques to produce the gas, such as hydraulic
fracturing. Conventional gas deposits, by contrast, can produce gas to the surface via a well under
the natural pressure and permeability of the reservoir (at least, until the natural pressure is
depleted).
The crucial geologic difference between tight sand gas formations and shale gas formations is that
shale gas formations are both the source rock and the reservoir rock. The natural gas is formed
within the shale layers, but because shale is virtually impermeable to flow, the gas remains
trapped and bound to the matrix of organic matter in the shale. Shale gas formations are also
deemed unconventional gas deposits.
The distinction between tight gas and shale gas is important in the Pavillion area because in the
upper 1,000 feet of the Wind River Formation, the sandstone lenses are also part of the aquifer
used for water supply. The sandstone, in contrast to shale, has enough permeability to transmit
groundwater to water wells in the region. In a sense, the sandstone lenses can act as a reservoir
for both natural gas and for groundwater. Shale formations, such as the Marcellus Shale, are not
permeable enough to transmit water and are generally not considered aquifers. In fact, thick
layers of shale are considered to be barriers to groundwater flow. The issue at Pavillion, where
hydraulic fracturing and gas production are occurring only slightly deeper than the deepest water
wells, would likely not be an issue for most shale gas plays.
Hydraulic Fracturing in Deep Versus Shallow Reservoirs
As noted above, the uppermost region of hydraulic fracturing in the Pavillion field is within a few
hundred feet of the deepest water wells. The close vertical proximity of natural gas development
activities and the bottom of the drinking water aquifer means that injected fluids would not have
to travel far to reach the aquifer, provided the fluids had a suitable pathway. Put another way, at
Pavillion there is less rock between the gas development activities and the aquifer. In contrast,
deeper shale gas reservoirs, such as the Marcellus Shale in the northeast United States, are
separated from overlying drinking water aquifers by thousands of feet of rock in areas under
active development.
In addition, if the intervening interval contains layers of rock relatively impermeable to flow, such
as other shale formations, then the chances of upward migration of injected fluids are reduced. In
such cases, the only pathways for fluid migration from a deep shale gas reservoir would be along
leaky old wells or poorly constructed production wells. Those types of wells would provide
possible routes for fluids to migrate upward because the wells pierce the intervening rock layers
and could connect the drinking water aquifer to the deeper, hydraulically fractured gas shale
reservoir.
Vertical Wells Versus Horizontal Wells
Vertical wells were drilled in the Pavillion field to hydraulically fracture and produce natural gas.
In tight sand reservoirs, such as the Pavillion field, often more wells are required to efficiently
Congressional Research Service
16
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
produce the gas from a given section of the reservoir than from conventional sand reservoirs.59 In
other words, one well in a tight gas reservoir will produce less gas over time than what would be
expected from a well in a conventional sand reservoir. That means that the well spacing for tight
gas sands could be much denser than for conventional sand gas fields. According to one source,
well spacing in a conventional sand reservoir is generally 160 to 320 acres per well, but in a tight
sand reservoir the well spacing can be as little as 10 acres per well.60 The greater number of wells
required to produce gas in tight sands also increases the number of potential vertical pathways
from the fracture or production zone to the surface, or to a drinking water aquifer if some wells
are improperly constructed or leak over time.
Well spacing for vertical wells in other unconventional gas reservoirs, such as the Marcellus
Shale, would also be more dense as compared to conventional gas reservoirs. However,
horizontal drilling is increasingly used to both hydraulically fracture and produce gas from shale
gas reservoirs. According to one source, shale gas development could require only one horizontal
well instead of four vertical wells to produce the same amount of gas.61 Also, one drill pad is
required for each vertical well drilled, while multiple horizontal wells could be drilled from the
same drill pad. If four horizontal wells were drilled from a single drill pad, that would be the
equivalent of drilling 16 vertical wells.62 For shale gas fields where horizontal wells are chiefly
used, the number of potential vertical pathways per land area that could transport leaked
contaminants to overlying drinking water aquifers likely would be far fewer than for tight gas
sand fields such as at Pavillion.63
The Hydraulic Fracturing Process
Although there would likely be some differences in the exact composition of hydraulic fracturing
fluids used and the volumes of fluid injected, the overall hydraulic fracturing process used at the
Pavillion field was probably generally similar to hydraulic fracture processes for other
unconventional gas fields. Horizontal wells used for hydraulically fracturing shale gas fields, such
as the Marcellus Shale, probably require a greater overall volume of fluid per well than is
required for vertical wells drilled into tight gas sands, such as Pavillion. The requirement for
greater volumes of water in shale gas fields would present different challenges regarding water
supply and water disposal than for tight gas sand fields, such as Pavillion. In addition to greater
volumes injected into the subsurface, greater volumes of fracture fluid would need to be stored at
the surface during a hydraulic fracturing operation, which could also increase the likelihood of
surface spills. Surface spills could infiltrate into shallow drinking water aquifers and pose a threat
to nearby water wells.
59 Stephen A. Holditch, “Tight Gas Sands,” Journal of Petroleum Technology, Distinguished Author Series, June 2006,
http://www.spe.org/jpt/print/archives/2006/06/JPT2006_06_DA_series.pdf.
60 Industry Technology Facilitator (ITF), Understanding Hydraulic Fracturing and Tight Gas Sands, July 4, 2011,
http://www.oil-itf.com/index/news-app/story.104/title.understanding-hydraulic-fracturing-and-tight-gas-sands.
61 J. Daniel Arthur, Brian Bohm, and Mark Layne, “Hydraulic Fracturing Considerations for Natural Gas Wells of the
Marcellus Shale,” presented at the Ground Water Protection Council 2008 Annual Forum, Cincinnati, OH, September
21-24, 2008, p. 8, http://www.dec.ny.gov/docs/materials_minerals_pdf/GWPCMarcellus.pdf.
62 Arthur et al., 2008, p.8.
63 Arthur et al., 2008, Table 1, shows a range of well spacings for different shale gas fields. The table indicates that well
spacing varies from as few as 40 acres per well in the Marcellus and Haynesville formations to as many as 640 acres
per well in the Woodford Formation.
Congressional Research Service
17
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Next Steps
In sum, the EPA Draft Report has raised many issues, questions, and concerns among potentially
affected stakeholders, including the oil and natural gas industry, environmental advocacy
organizations, and other citizens. On January 20, 2012, 11 members of the Senate Environment
and Public Works Committee sent a letter to EPA Administrator Lisa Jackson asking that the EPA
investigation be considered a “highly influential scientific assessment and that any related,
generated report is subject to the most rigorous, independent, and thorough external peer review
process.”64 The extent to which EPA may revise its findings in response to public comments and
the forthcoming external scientific review is unclear and will not be known until the agency
finalizes its report. What further actions the agency may take under CERCLA, or possibly other
applicable authorities, once the report is finalized also are uncertain at this time, as EPA had not
drawn definitive conclusions about any potential risks to human health and the environment
linked to a specific release. (The Appendix reviews EPA’s response authorities under CERCLA.)
Regardless of these outcomes, the potential applicability of EPA’s findings at the Pavillion site to
other sites where similar hydraulic fracturing operations are conducted would depend heavily
upon the extent to which the geology and hydrogeology are similar, as well as other site-specific
factors.
64 Letter from 11 members of the Senate Environment and Public Works Committee to Administrator Lisa Jackson,
January 20, 2012, http://epw.senate.gov/public/index.cfm?FuseAction=Files.View&FileStore_id=04ae8926-3ed7-
427a-9ef9-488a4b9b58be .
Congressional Research Service
18
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Appendix. EPA Response Authority and Possible
Further Actions
The EPA Draft Report constitutes an early stage of the standard site-specific process under
CERCLA, which has focused first on characterizing the groundwater to identify potential
contamination and potential sources of contaminants to discern whether a release of hazardous
substances may have occurred that warrants further action under CERCLA. If EPA were to
determine that cleanup was warranted, certain exclusions or limitations on the authorities of
CERCLA could constrain the actions that EPA could pursue under that statute, such as exclusions
for releases of natural gas65 and naturally occurring substances,66 and exemptions from liability
for federally permitted releases which may include state permitted releases for underground
injection,67 as defined in CERCLA.
Although this type of site investigation is funded and performed under EPA’s Superfund program,
it does not constitute the placement of the site on the National Priorities List (NPL). Rather, such
an investigation is the initial—and in most cases only—stage of the site-specific process under
CERCLA.68 Relatively few potentially contaminated sites reported to EPA result in an NPL
designation. A total of 49,909 sites have been reported to EPA over time. Of this total site
universe, 21,090 sites (42%) have been the subject of site inspections similar to that conducted at
Pavillion, of which 1,652 sites (3%) have been listed on the NPL to date.69
Whether EPA may pursue further action at a site under investigation depends on the findings. In
its draft report, EPA did not reach a conclusion definitively linking contaminants found in the
groundwater to a specific release that may present a risk to human health or the environment.
Accordingly, the agency also has not determined whether cleanup actions may be warranted, nor
has the agency identified any potentially responsible parties as being liable for the contamination
under Section 107 of CERCLA.70 The source of the contamination first would have to be
confirmed and the potential risks further examined, before any determinations could be made as
to whether cleanup may be warranted and whether any potentially responsible parties are
identified who may be liable for the cleanup.
65 42 U.S.C. §9601(14) and 42 U.S.C. §9601(33). For the purposes of CERCLA, natural gas is excluded from the
statutory definition of a hazardous substance, and pollutant or contaminant, respectively.
66 42 U.S.C. §9604(a). EPA generally is prohibited from responding to a release of a hazardous substance that is
naturally occurring, unless EPA determines that the release constitutes a public health or environmental emergency,
and that no other person with the authority and capability to respond to the emergency will do so in a timely manner.
67 42 U.S.C. §9607(j) and 42 U.S.C. §9601(10). Entities conducting site operations performed under certain applicable
federal permits, or state permits (specifically for underground injection involved in oil or natural gas production) are
excluded from liability under CERCLA for a release allowed within the confines of such permits, unless the release
were to violate permit requirements and therefore not be a permitted release in that sense.
68 Information on the stages of the site-specific process under CERCLA is available on EPA’s Superfund program
website: http://www.epa.gov/superfund/cleanup/index.htm.
69 Site numbers are based on search results generated from EPA’s Superfund Site Information Database on January 24,
2012, available at http://cumulis.epa.gov/supercpad/cursites/srchsites.cfm. The total site universe includes archived
sites at which no further federal action is planned. The total of 1,652 sites listed on the NPL includes 355 sites that EPA
later deleted, based on the agency’s determination that the cleanup objectives had been met at those sites.
70 42 U.S.C. §9607. Categories of potentially responsible parties who are financially liable for the costs of response
actions taken under CERCLA include past and current owners and operators of facilities, generators of waste sent to
facilities for disposal, and transporters of waste who selected the facility for disposal.
Congressional Research Service
19
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
If EPA were to find that a release or threatened release of a hazardous substance may present a
threat to human health or the environment, EPA would be required to evaluate the potential
hazards according to the criteria established under Section 105(a)(8)(A) of CERCLA to determine
whether the site may be eligible for listing on the NPL.71 These criteria and how to apply them are
outlined in the Hazard Ranking System (HRS).72 Although EPA is required to evaluate the
potential health hazards at a site in such instances, whether EPA may list the site on the NPL to
elevate its priority for cleanup at the federal level would depend not only on the outcome of the
health hazard evaluation, but also on numerous other statutory and regulatory criteria, including
the criteria under Section 105(h) for deferring a site to the state in which the site is located instead
of listing it on the NPL.73
If a site is not listed on the NPL and is not deferred to the state, EPA still may take certain actions
at the federal level to address potential health and environmental risks, including the performance
of emergency “removal” actions if warranted. Under CERCLA, removal actions generally are
measures intended to address more immediate risks of exposure,74 whereas “remedial” actions
generally are measures intended to provide a more permanent solution to address long-term
risks.75 Although a site must be listed on the NPL to be eligible for Superfund appropriations to
perform remedial actions,76 removal actions are eligible for such federal funds regardless of a
site’s listing status.
EPA also may pursue mechanisms to enforce cleanup liability under CERLCA if the source of
contamination is confirmed, the release that caused the contamination falls under the authorities
of the statute, and the potentially responsible parties who can be held liable under the statute can
be identified. These mechanisms include cleanup orders under Section 10677 and cleanup
agreements under Section 122,78 neither of which hinges on whether a site is listed on the NPL.
At this juncture, EPA has reached no such decisions at the Pavillion site. Rather, the EPA Draft
Report has identified constituents in certain portions of the aquifer that the agency has
characterized as being consistent with, or similar to, some substances used in the natural gas
production operations, but has not definitively concluded the source of the constituents or any
potential risks that may warrant cleanup.
71 42 U.S.C. §9605(a)(8)(A).
72 40 C.F.R. Part 300, Appendix A. Additional information on the Hazard Ranking System is available on EPA’s
Superfund program website: http://www.epa.gov/superfund/programs/npl_hrs/hrsint.htm.
73 42 U.S.C. §9605(h).
74 42 U.S.C. §9601(23).
75 42 U.S.C. §9601(24).
76 40 C.F.R. §300.425.
77 42 U.S.C. §9606.
78 42 U.S.C. §9622.
Congressional Research Service
20
The EPA Draft Report of Groundwater Contamination Near Pavillion, Wyoming
Author Contact Information
Peter Folger
David M. Bearden
Specialist in Energy and Natural Resources Policy
Specialist in Environmental Policy
pfolger@crs.loc.gov, 7-1517
dbearden@crs.loc.gov, 7-2390
Mary Tiemann
Specialist in Environmental Policy
mtiemann@crs.loc.gov, 7-5937
Congressional Research Service
21