October 29, 2002
Drinking Water Protection Division
U.S. Environmental Protection Agency
Mail Code 4606
Ariel Rios Building
1200 Pennsylvania Avenue NW
Washington DC 20460 by e-mail only
Re: Comments On Study of Impacts of
Hydraulic Fracturing of Coalbed Methane Wells
Dear Ms. Cronkhite: This letter supplements my e-mail of October 28, 2002 that provided comments concerning the Study of Potential Impacts of Hydraulic Fracturing of Coalbed Methane Wells on Underground Sources of Drinking Water.
In my comments of October 28, 2002, I stated:
I have represented coalfield citizens in coal mine-related water loss cases for some twenty years. Based on my experience with mine blasting and vibration-induced and mine subsidence-induced fracturing of coal seams, which often act as the aquatard / aquaclude for the valley bottom stress relief aquifers, I believe that the science does not support the conclusion that hydraulic fracturing of those coal seams that are nearer the surface (and which typically are the USDW's serving local communities) would be of minimal consequence. The impact of additional fracturing of the coal seam, and potential for disruption of the aquifers carried by those seams and the underclays beneath those seams, could be devastating on local aquifers supplied by secondary permeability stress-relief fracture flow groundwater systems - the aquifers and USDW's upon which most folks living in eastern Kentucky outside of the towns rely. I will support this comment with technical literature that will be sent this week, and will further elaborate on the observations. In order to assure proper consideration of the comments, however, the author incorporates by reference the studies of Borchers and Wyrick, Kipp, Dinger and Lawrence concerning the hydrogeology of eastern Kentucky / Appalachian coal fields and the response and vulnerability of the stress-relief aquifers to disruption.
This letter provides that supplemental comment and supporting documentation.
The Kentucky Resources Council, Inc., is a non-profit environmental advocacy organization incorporated under the laws of the Commonwealth of Kentucky and dedicated to prudent use and conservation of the natural resources of the commonwealth. KRC has, since 1984, provided legal and technical assistance without cost to low-income individuals and communities across the state.
During the many years that the Council has been engaged in providing legal representation to individuals and communities, KRC has observed first-hand the significant adverse effect of surface coal mining operations (including both strip and underground coal mining) on the groundwater supplies of eastern Kentucky. It is from this experience and observation of the nature and occurrence of groundwater resources in eastern Kentucky that these comments are submitted.
KRC first became aware that EPA was seeking comment on this issue on October 26, 2002, so that these comments are of necessity more abbreviated that otherwise would be the case.
Before turning to specific comments concerning the study, KRC is extremely concerned that the agency is entertaining the possibility of supporting legislation exempting hydraulic fracturing from the ambit of the Safe Drinking Water Act. The EPA website indicates that the agency has been asked to support legislation which would exempt hydraulic fracturing from the Safe Drinking Water Act, and that the agency will consider comments on the data prior to making further decisions.
KRC is concerned that the agency is creating an improper linkage between the decision to regulate hydraulic fracturing within the ambit of regulated injections under the SDWA or to support efforts to exempt this type of underground injection from the ambit of the law, and the presence of documented cases of health harm from non-regulation. The extension of protection to underground sources of drinking water from contamination under the Safe Drinking Water Act's Underground Injection Control program does not depend on the existence of documented instances of harm, but instead is a program intended to forestall and prevent such harm by isolating the injected fluids from aquifers that are or could be developed as USDW's. To establish proven harm as a regulatory threshold is a distortion of the intent and purposes of the law and places at risk the resources and potential resources that Congress sought to protect.
KRC is concerned that the study understates significantly the potential for contamination of USDW's from hydraulic fracturing of coal seams associated with coalbed methane (CBM) production.
With respect to studying those areas in which CBM development has or is occurring, the study authors apparently equated lack of documented cases of public health harm with lack of public health risk. The report notes that "[I]f risks from hydraulic fracturing of CBM wells were significant, we would expect to find instances of well water contamination from the practice." Unfortunately, in the absence of background qualitative and quantitative information concerning the nature and occurrence of groundwater resources in areas where CBM development has occurred, a review of the available data may in many cases understate the changes occurring in the groundwater resource as a result of injection of water and other fluids to stimulate CBM production. Public health risk is not the appropriate end-point of the law ? for the statute protects the actual use and potential use of an aquifer yielding above a certain threshold of potentially potable water. It is the impact on the resource itself that should drive the regulatory issues, not an assumption that absent proven health harm, regulatory controls are unnecessary.
The second, and more troubling aspect of the study was the superficial manner in which the "theoretical potential for hydraulic fracturing to impact drinking water wells" was approached in the Appalachian coal fields. In Kentucky, in particular, the report suffers from both a lack of data concerning the occurrence and distribution of the CBM resource, and a fundamental lack of understanding of the nature and occurrence of the groundwater resource.
The potential impacts of induced fracturing of coal seams intended to force fluids into the seam to accentuate fracturing and to "liberate" CBM gases, has the potential to cause contamination of USDW's through several mechanisms. By injecting fluids containing contaminants of concern that exceed primary drinking water standards, the resource may cause direct contamination of water that is or could be withdrawn for consumption. By accentuating fracturing, the injection may alter the relationship of bedding planes, joints and other secondary permeability features of the groundwater flow system in a manner that causes loss of water in areas where groundwater had previously flowed, or which causes communication between aquifers and other water-producing zones having undesirable characteristics (fecal contamination from surficial aquifers; saline from deeper water-transmitting zones). Accentuating fracturing may also cause a dewatering of the USDW into underlying mine workings or deeper aquifers. Finally, water injection and then withdrawal may alter on a more permanent basis the groundwater flow, plugging fractures and destroying the productive capacity of the USDW.
The study appears to have given little consideration to the hydrogeology of the Appalachian coal fields and the distinct probability that induced fracturing of near-surface coal seams (within 200 feet of surface) can disrupt valley-bottom aquifers. The project approach assumes that the USDWs are of low permeability (ES-9) and that gas production occurs in areas far removed from groundwater users (ES-10).
In truth, the literature suggests that the groundwater resource in the Appalachian mountain coal-bearing region is one that is dominated by secondary permeability flow through valley stress-relief fractures and bedding planes that are highly permeable and through which disruption or contamination of the resource will be readily transmitted.
Borchers and Wyrick, in their paper Application of Stress Relief Fracturing Concepts for Monitoring the Effects of Surface Mining on Groundwater in Appalachian Plateau Valleys noted that
hydrologic study in Black Fork Valley in southern West Virginia
revealed that the hydraulic properties of the near-surface rocks
are controlled to a great extent by valley stress-relief features.
It has proven difficult to develop groundwater monitoring designs
for contour surface mines because ground-water flow systems in
the Appalachian Plateau coal areas are poorly understood. The
equations that describe ground-water flow were developed for use
in homogenous, isotropic aquifers of infinite areal extent. These
ideal aquifer systems are not often found in the coal-bearing rocks
of the Appalachian Plateaus. Most ground water in this region is
transmitted through secondary-permeability features such as
fractures and solution openings in the rock.
Borchers cautioned that each watershed must be viewed in a site-specific manner for purposes of understanding the groundwater resources and defining measures to protect the resource from damage and to predict hydrologic consequences, since "stress relief fractures are not the only influence on the flow and occurrence of ground water in the Appalachian Plateaus. Other secondary-permeability features, such as joints, faults, coal cleats, fractures associated with anticlines and lineaments, solution openings, and subsidence fractures, may influence the ground-water flow system of a particular area. Id. (1981).
EPA's assumption of low permeability suggests that the authors are looking at the primary permeability of the rock strata rather than the dominant flow mechanism, which is secondary permeability induced by the stress-release fracturing and bedding planes. Failing to understand the rapid flow of water through this system and the relatively low potential for attenuation of contamination leads, as it appears to have in the study, underestimation of the vulnerability of the resource and the potential to adversely affect other users of the resource.
In reviewing the impact of underground mining on Grapevine Creek, Sloan and Warner noted that using primary hydraulic conductivities derived from rock core samples underestimate actual conductivity, since "using thee estimates of hydraulic conductivity, the calculated flow times are in the hundreds of years for any of these wells that were in fact disturbed immediately subsequent to mining."
Sloan and Warner, A Case Study of Groundwater Impact Caused By Underground Mining.
The study mistakenly assumed that in the coal fields, the predominant groundwater resource will be found in sandstones and siltstones, since they have relative primary conductivities than shales and claystones which act as confining layers. As Schubert noted:
If this were solely the case, normal methods could be used to
determine hydraulic properties and mathematically to predict
groundwater system response to mining activities.
However, fracturing due to jointing and faulting can locally
reduce or destroy the confining effect of relatively imper-
meable beds an create much greater hydraulic conductivities
in shales and siltstones.
In zones of relative intense fracturing, hydraulic conductivities are]
often several orders of magnitude higher than in unfractured rocks,
resulting in fracture-dominated flow. This preferred means of flow
creates difficult problems in predicting groundwater movement
and response of a flow system to mining activities.
Schubert, Fracture Flow of Groundwater in Coal-Bearing Strata, (Argonne National Laboratory).
KRC incorporates by reference these studies, and additionally, the following works which speak to the significantly higher permeabilities of valley-wall, stress-release fracture flow systems.
Borchers and Wyrick, Application of Stress-Relief Fracturing Concepts for Monitoring the Effects of Surface Mining on Groundwater, Symposium on Surface Mining Hydrology, Sedimentology and Reclamation, Lexington, Kentucky 1981.
Schubert, Fracture Flow of Groundwater in Coal-Bearing Strata, Lexington Symposium (1980)
Ferguson,Valley Stress Release in the Allegheny Plateau, US Army Corps of Engineers, Engineering Geology, Association of Engineering Consultants (1967)
Kipp, Dinger, Lawrence, A Conceptual Model of Ground-Water Flow In The Eastern Kentucky Coal Field, Lexington Symposium (1983).
Additionally, the effect of subsidence from previous underground coal mining of one or more seams below the seam from which the CBM is proposed to be produced may dramatically affect the nature and occurrence of the groundwater resource and the hydrologic response to the injection and subsequent withdrawal of fluids on groundwater flow regimes. See: Hobba, Effects of Underground Mining and Mine Collapse on the Hydrogeology of Selected Basins in West Virginia, USGS Report of Investigation RI-33 (1981); Harper et al., Unexpected Hydrologic Perturbation in an Abandoned Underground Coal Mine: Response to Surface Reclamation? Environ. Geol. Water Sci., 15:3 (1990).
In short, before any theoretical predictions concerning the "potential for hydraulic fracturing to impact drinking water wells" is made, the authors would do well to better understand the complex groundwater resource of the Appalachian coal fields, in which the idealized concept of a homogenous, isotropic aquifer in which primary porosity and permeability govern groundwater occurrence and transport, gives way to the reality of a heterogeneous, anisotropic groundwater resource in which secondary, fracture-dominated flow provides the dominant groundwater mechanism for the region. It is this resource that provides drinking water for many coalfield residents, and it is often the coal seams below drainage and the underclays that provide the confining layer for the valley-bottom aquifer. Additionally, underground mining has created in some areas a subterranean reservoir fed by fracture flow and providing potable water for communities. Any hydraulic fracturing to induce CBM production from remaining coal seams may disrupt these groundwater resources, and protection against this possibility demands the application of rigorous site-specific review under the UIC program.
Before EPA entertains the prospect of abandoning protection of the water resource for the residents of the coal fields by exempting hydraulic fracturing from the ambit of the UIC program, it must conduct a realistic assessment of the nature and occurrence of the groundwater resource and the vulnerability of that resource. KRC is convinced that, as Kentucky considers development of CBM as a resource, EPA must remain engaged in regulation of CBM-related injections under the UIC program because there is no other program at the state or federal level exists through which the site-specific investigations and precautions needed to protect the resource would be required, and that in the absence of EPA regulation of the practice, there is a distinct probability of contamination, disruption or loss of these vulnerable resources due to induced fracturing of coal seams associated with CBM.
Thank you for your consideration of these concerns.