Reinjection Technology - Process and Applications, Including Disposal of Coal Bed Methane Product Water

by Nikos Warrence and James W. Bauder
Department of Land Resources and Environmental Sciences
Montana State University - Bozeman

SYNOPSIS: Reinjection technology of groundwater pumped to the land surface is currently being used in many different ways, including restoring depleted aquifers, treating contaminated groundwater, and injecting coal bed methane (CBM) product water back into the ground. The physical parameters involving reinjection are first examined. Classifications of injections wells, as well as their potential hazards, are addressed. Some of the legal issues involved in reinjection are also dealt with. A Netherlands study using a unique approach to reinjection, and its applicability to CBM product water injection, is looked at. Issues involving the potential for reinjection of CBM water are discussed. Primarily the engineering obstacles and potential benefits that reinjection would have are dealt with, although some of the legal and political aspects of CBM are also discussed. Our research was unable to reveal definitive validation either in support or in opposition to CBM product water reinjection to the source aquifer. Despite this shortcoming, we are optimistic reinjection can play an important role in dealing with CBM product water in Wyoming and Montana. A brief description of efforts to model where water goes once it is pumped into the ground concludes the paper. Our conclusion is that there are significant uncertainties associated with the feasibility of reinjection of CBM product water to source aquifers and research focus should be directed toward modeling and evaluating CBM reinjection dynamics.

Introduction

     Concern about groundwater has grown rapidly in recent years. Both stricter environmental legislation, largely due to increased knowledge about subsurface waters, and improved technology have led to the increased attention, as has increased demand for usable groundwater (Brooks et al., 2001). Groundwater contamination, aquifer depletion and subsidence, waste disposal, and mining have all come to the forefront in recent years. One of the most far-reaching new ways of dealing with groundwater is reinjection technology. Reinjecting fluids back into the ground has been used to deal with a multitude of groundwater issues. Recharging depleting aquifers, disposing of sewage and industrial waste, dealing with irrigation and stormwater runoff, flood prevention, in situ mining, subsidence control, preventing salt-water intrusion, and remediating contaminated groundwater have all been addressed through reinjection (Bouwer, 1998). Groundwater depletion is another concern, with rapidly increasing populations and the fact that eight percent of food crops globally are currently grown on farms which are using groundwater faster than the aquifers are being replenished (Postel, 2001). As can be imagined, technology to reinject water and other fluids into the ground is quite varied, ranging from shallow sinkholes to complex reinjection wells several thousand feet deep. (Bouwer, 1998).

     One of the technically most challenging sorts of reinjection is that involving wastewater disposal from the coal bed methane extraction process. Coal bed methane (CBM) is defined as natural gas - methane - derived from coal seams saturated with water. To release methane gas adsorbed to coal particle faces and absorbed into water in the coal seam, the pressure in saturated coal seams is reduced by continuously pumping water out of the coal seam containing methane (Rice and Nuccio, 2000). Because of the generally high sodicity and salinity concentration of this water, surface disposal can potentially be harmful to crops and soils. Whether or not water can realistically be reinjected (that is, put back into the original coal seam) or if it must be injected to a shallower or deeper aquifer is a key question currently under debate. In any event, disposal by injection to an aquifer other than the source aquifer is the preferred mode of disposing of CBM wastewater outside the Powder River Basin, and it has potential for use in this area in the future (EPA, 2001). Cost is the main obstacle to reinjection of CBM product water to the source aquifer. The other main issues are the physical constraints of simultaneously withdrawing and reinjecting into the same aquifer, if putting product water back into the producing coal seam is the desired goal.

     It should be noted that reinjection implies putting water back into the original aquifer it came from, i.e. the coal seam in CBM, whereas injection merely implies putting water into the ground. Injection of CBM product clearly works and is being done in the southwest and southeast U.S., but until now there has not been sufficient incentive for industry to look into reinjection into the producing coal seam. Within the Powder River Basin concern about drying up of ephemeral springs and wells, which can potentially still occur even if water were to be injected into a different formation, will hopefully lead to an increased research and hopefully understanding of the physical and economic feasibility of CBM reinjection. In any event, before addressing specific applications of reinjection, the physical process and various sorts of injection wells will be explained.

Physical parameters of reinjection

     Location is the first concern when looking at groundwater injection. Local climate, soil, water quality, and hydrogeology must all be examined (Custer, Steve1, personal correspondence, 2001). If soils are very permeable, i.e. very sandy soils, and aquifers unconfined and relatively shallow, groundwater recharge can be achieved with infiltration basins (Gill, 2001). This is obviously the easiest and least expensive way to get water back into the ground, but will only work under very specific conditions. In addition to very permeable surface layers, vadose zones (areas before the water hits the aquifer) must have little or no clay or other substances which would inhibit flow to the aquifer. And the aquifers must be both transmissive and unconfined. Water must be able to enter the aquifer, and there must be enough space in the aquifer to hold the incoming water (Bouwer, 1988), unless sufficient reinjection pressure is applied to overcome the existing pieziometric head of the groundwater. Obviously, it is cheaper and mechanically easier to inject water if there is sufficient void space within the receiving aquifer.

     It should be noted that in terms of CBM, whether or not discharge water can be realistically, physically and economically, returned to the same aquifer, that is reinjected, or only injected to a different source is an issue under debate. Since the CBM source coal seam remains saturated or near saturated during the mining process, whether or not there is ample void space in the coal seam to receive reinjected water, at least without a great deal of mechanical pressure and expense, is a question still under debate. (Interestingly, Steve Custer1 has stated that he has heard industry personnel comment that dewatering may be a good idea, although the current practice is to keep the coal seams at or slightly below saturation, primarily because of concerns about structural damage to the aquifers if dewatering were to occur.) Because it is mechanically easier to reinject into a shallower or deeper aquifer, and this clearly does not affect methane extraction, there has been little incentive for industry to investigate reinjection into the source coal seam. Only recently, with concern about the impacts of reducing water levels in coal seam aquifers, has their been serious talk about putting water back into the coal seam. Concerns have been expressed that even if water were injected to a lower formation, ephemeral springs and wells would still dry up, and this reinjected water would be effectively lost unless wells were drilled to deeper depths where the water might be after injection to a lower aquifer.

     In some borderline injection cases, infiltration trenches in combination with piping are dug a few feet beneath the surface to aid infiltration. For instance, the Massachusetts Military Reserve uses trenches and piping to get decontaminated groundwater into shallow aquifers (Gill, 2001). Considerations in designing infiltration basins, and determining whether or not they are feasible, include depth of water in the basins (in which head and required infiltration depth need to be taken into account), length of flooding and drying periods, how often basins will need to be cleaned, pre-sedimentation conditions, soil and bedrock conditions down to the final desired destination for the water, and whether stagnant or flowing water conditions are sought. (Stagnant conditions would allow particles to settle, whereas flowing conditions would cause fine sediments to stay suspended (Bouwer, 1988)).

     Water quality also needs to be considered. Chemical and biological effects of the incoming water need to be addressed (Steve Custer1). Additionally, effects of sodium adsorption ratio (SAR) and total dissolved solids (TDS) on flocculation-deflocculation of clays in the bottom layer and vadose zones of the aquifer need to be considered. Changes in water quality as the water moves downward and the potential for leaching of contaminants and other trace elements should also be addressed (Bouwer,1988)

     More often than not, conditions conducive to infiltration basins do not exist, or suitable land is too expensive or too distant from the water source to be used for reinjection. In these instances, some sort of reinjection well is used. Very similar to pumped wells in design, reinjection wells use screened sections (both to define the position where reinjection begins and to filter out sediments, microorganisms and whatever else could clog up the well), gravel packs (in unconsolidated aquifers), and grouting (Bouwer, 1988). Before reinjection, water must be treated to remove almost all suspended material so that the well does not clog up (Hebbink, 1988). Even so, these wells will eventually clog up at the interface between the well and aquifer because of sedimentation. Biofouling is another common problem. Microorganisms will grow on well screens and, as much as or sometimes even more so than sediments, clog up a well. Screens often have to be periodically cleaned in order to clear out these microorganisms (Steve Custer1). To deal with this, periodic pumping and/or redevelopment of the well is needed. Redevelopment constitutes surging or jetting the well, in which sediments are essentially flushed out through mechanically-induced pressure (Bouwer, 1988). Depending on where the water is being injected, particularly if it is going into a public water supply, federal and state regulations may require additional filtering and treatment of the water (EPA, 2001).

     Because of clogging, specific capacity for injection wells in unconsolidated aquifers is generally half that of pumping wells over time. Wells going into areas with higher hydraulic conductivity, i.e. well-developed secondary porosity, have injection rates much closer to pumping wells. Aquifers with fractured rock structure or limestone with solution channels allow for much better infiltration (Bouwer, 1988). This is particularly relevant to reinjection in CBM. Other than expense, the greatest obstacle to reinjecting CBM waste water back into the producing coal seam is that of putting water back into a saturated, or near saturated, system, which makes it extremely difficult (Bauder, Jim3, personal correspondence, 2001). So any additional space where water can go is helpful, be it pore spaces or depleted aquifers (Steve Custer1).

Legal Constraints

     Prior to 1974, at least within the U.S., little attention was payed to groundwater quality or subsurface disposal methods. The Safe Drinking Water Act (SDWA) of that year changed that. Underground Injection Control (UIC) was part of that act. The idea behind this was to prevent contamination of Underground Sources of Drinking Water (USDW) by injection wells (EPA, 2001). The USDW is defined as an aquifer or part of an aquifer which supplies any public water system, or has enough groundwater to potentially supply a public water system, has less than 10,000 mg/L total dissolved solids and is not an exempted aquifer, i.e. an aquifer whose water quality or conditions make it unsuitable for use as a public water source. Title 40 of the Code of Federal Regulations defines a well as "a dug hole or a bored, drilled or driven shaft whose depth is greater than its largest surface dimension." Injection is defined as "the subsurface emplacement of fluids in a well where a fluid is any material that flows or moves whether it is semisolid, liquid, sludge, or gas" (Tennessee Department of Environment and Conservation, 2001). The regulatory agency having jurisdiction must authorize any injection before it occurs. States which have primacy to implement the UIC program have state agencies to authorize injection. In other states, the USEPA would be the regulatory agency (Tennessee Department of Environment and Conservation, 2001). In Montana the regulatory authority is the Montana Oil and Gas Commission.

Classification of Injection Wells

     The UIC divides injection wells into five main categories:
Class I wells   Injection of hazardous wastes or the disposal of industrial and/or municipal fluids beneath the lowermost USDW.
Class II wells   Injection of fluids connected to oil and natural gas mining or fluids or compounds used to enhance hydrocarbon recovery. Usually these wells inject beneath the lowermost USDW except where the USDW is hydrocarbon producing.
Class III wells   Injection of fluids solely for the extraction of minerals, in other words in situ mining.
Class IV wells   These wells have been banned, but were used to dispose of hazardous or radioactive wastes into or above the USDW.
Class V wells   Disposal and aquifer recharge, as is the case with CBM injection wells, rather than assisting in the mining process.

     Injection wells in the southwest U.S., which often inject below the USDW, are considered Class V because their purpose is aquifer recharge and disposal, not to assist in the mining process. Class V wells, into which most reinjection wells fall, are wells not included in the other classes, and generally inject nonhazardous fluid into or above a USDW. Class V wells can be used for disposal, aquifer recharge, fossil fuel and mineral recovery (EPA, 2001). Wells injecting CBM product water would fall into this class.

     Class V wells include 30 different types of wells, ranging from simple cesspools barely deeper than they are wide to complex, and expensive reinjection wells several thousand feet deep. Class V wells fall into two basic categories, "high-tech" and "low-tech." "Low-tech" wells have simple casing designs and equipment and generally inject into shallow formations, either by gravity flow or simple low volume pumps. "High-tech" wells usually have multiple casing strings, complex well head equipment designed to control and measure pressure, and have capacity to inject large volumes of fluid, often into deep saline solutions that are separated from aquifers by a confining layer which is impermeable (Bouwer, 1988).

     A major concern associated with any reinjection is the potential for contaminating other groundwater reserves (Lee, 1996). This depends on where injection occurs relative to the aquifer, how the well is designed, built, and operated, the quality of the material being injected, and volumes of material being injected. Obviously, wells which inject beneath the USDW are least likely to contaminate groundwater. Wells injecting directly into the USDW are potentially the most harmful. Wells discharging above the USDW allow for the removal of some contaminants by attenuation, adsorption, and degradation in the injection zone above the USDW (EPA, 2001).

     There are approximately 170,000 Class V wells in the United States. About 83% of these wells fall into two main categories, drainage wells (57%) and sewage related wells (26%) (EPA, 2001). As almost half the U.S. population receives its drinking water from aquifers or sources into which Class V wells discharge, it is important that these sources are not contaminated through reinjection of contaminated water. Awareness and means to minimize these impacts are needed (Lee, 1996).

Injection of CBM Product Water

     The recent boom in CBM mining in the Powder River Basin in eastern Wyoming and Montana is another setting which may offer opportunity for the use of reinjection technology. About 4,000 gas production wells have already been drilled in the area, with projections anywhere from 40,000 (EPA, 2001) to over 100,000 (Morrison, 2001) wells in the area within the next few years. Each well produces an average of 12,000 gallons/day (8-10 gallons/minute) as a by-product (McCarthy, 2001). Methane extraction occurs when water is kept at the top of, or just below the top of, the coal seam aquifer by continuous pumping, thereby reducing the pressure which keeps methane adsorbed to the coal's surface. Much of the water is now being impounded, largely due to concerns of environmental groups and soil and water scientists. Reinjecting the water back into the ground is an option being investigated to deal with the potential impact of surface discharge of CBM product (Schneider, 2001).

     Cost is a major obstacle to the widespread adoption of reinjection of CBM product water in Wyoming (Dobkins, Terry3, personal correspondence, 2001). The fact that CBM mining was well established in the Powder and Tongue River basins before efforts were made to determine the potential impacts on soil and water quality is another reason reinjection has not been widely adapted (Morrison, 2001). Therefore, there is currently little incentive for industry to reinject CBM product water (Steve Custer1). Currently, most CBM operations outside of the Powder River Basin do reinject discharge waters to deeper formations, largely because the quality of the water is so much worse, in terms of TDS and SAR, than the water in eastern Wyoming and Montana. Waters with TDS values as high as 75,000 mg/l are present in areas such as Colorado, Arizona, and New Mexico (EPA, 2001). Also, coal reserves are more concentrated in these areas, making it more economically feasible for industry to adopt injection (Wheaton, John4, personal correspondence, 2001). This water is not being reinjected back into the producing coal seam, however. It is being injected into deeper aquifers, therefore avoiding the complications of putting water back into a saturated system (EPA, 2001).

     Whether or not CBM water can be reinjected into the producing seam, or if it must be put into a shallower or deeper aquifer, is a difficult question. However, in other situations of injecting CBM product water, there is little evidence of reinjection into the original coal seam aquifer. During CBM extraction, hydrostatic pressure is kept at the top of, or just below the top of, the coal seam, which is kept saturated. A large reason for this is that it is possible that if the coal seam were to go unsaturated, formation permeability, affecting water and methane gas migration, would be damaged (John Wheaton4). Whether reinjection is feasible would have to be looked at on a case by case basis. Not only is every seam different, but the overlying soils and storativity vary as well. Finding the critical distance between the producing well and the reinjection well so that the respective drawdown and recharge fields do not significantly overlap is key (John Wheaton4). If they exhibit too much influence on each other, reinjected water would likely reappear as surface water being pumped out. Either that or hydrostatic pressure in the coal seam would be increased to the point where reinjection would become difficult if not impossible (Jim Bauder2).

     Storativity is another critical factor. If the water can be placed into an already depleted aquifer, this obviously makes injection much more feasible (Bouwer, 1998). For instance, Pennaco is reinjecting water back into the already depleted aquifer which supplies water for the city of Gillette. Because the aquifer has been drawn down for sixty years, there is ample room to store incoming water (Terry Dobkins3). There is certainly some space in the coal aquifer during CBM production, which also encourages the migration of methane gas, but whether or not this is sufficient for reinjection is unclear, as the coal seam itself is not actually dewatered (John Wheaton4). Continual pumping of water to keep the head at or just beneath the top of the coal seam would lead to void space somewhere above the coal seam, although the continual flow of groundwater and localized structural differences and storativity of subsurface formations would have to be taken into account to find exactly where these voids occur.

     An expressed concern of many environmentalists and local residents is that drawdown of local and regional aquifers could reduce or eliminate crucial ground and surface water supplies, such as springs, streams, domestic and livestock wells, and subirrigated farmland (Regele and Stark, 2000). How much CBM drawdown will affect streams and springs is largely dependent on where the coal outcrops (John Wheaton4). Because the coal seams are the primary aquifers in the area and also have a higher hydraulic conductivity than the surrounding shales and sands, CBM drawdown will definitely have an effect on wells and springs. Industry personnel, such as Pennaco's Vice President-Production Terry Dobkins3 feels that aquifer drawdown is an issue, which needs to be addressed especially for ranchers who have water wells in the same zone from which water is being pumped. Dobkins states that currently available data show that the water drawdown of the entire aquifer may be on the order of 2 to 10%. How fast the aquifer recharges is a function of proximity to the outcrop and the aquifer's characteristics. He believes that the aquifer is sufficiently large so that the drawdown by this CBM extraction, the 2 to 10%, will not negatively impact the lifestyles of the residents in areas of CBM development. He also indicated that there are EIS studies underway that will better define this issue (Terry Dobkins3).

     Others, such as MSU groundwater specialist Steve Custer, predict that aquifer recharge would be much slower, up to a thousand years or more, and therefore more of a concern. He feels the very high clay soils in the area and generally arid conditions would cause this slow recharge time. (Steve Custer1). John Wheaton puts the recharge time at somewhere between a few years and twenty years, although he indicates that there is a good deal of uncertainty about this (Wheaton, 2001). A case by case study would be the only appropriate way of answering this question, because of the large degree of variability of soil type in the area (Bauder, 2001) which in turn effects drainage efficiency and how deep the water will penetrate and where it will wind up - on the surface, in streams and lakes, or in aquifers (Wheaton, 2001).

     Aquifer recharge is a key question; 41.5% of the groundwater used in Montana is for irrigation (USGS, 2001). And if aquifers are not recharged quickly, particularly if they are the same aquifers being used for irrigation, this could be disastrous to the farming community. While most groundwater irrigation in the area comes from shallower sources than the producing seams (Bauder, 2001), it is certainly possible that coal seam aquifer drawdown can, in turn, draw down water higher up in the profile. Tests to determine connectivity of aquifers and efforts to map out aquifers very precisely are expensive and often not done by industry because of the cost involved (Custer, 2001).

     Fault blocks and other trapping mechanisms would also aid in reinjection. Injection manifolds and pumps could be put into place at discharge points. Analysis of individual wells, patterns of reinjection, position relative to aquifers and other subsurface structures, soil structure and permeability, and other factors must also be considered. How deep reinjection wells need to be is a major consideration, particularly to industry, as deeper wells are obviously more expensive. In areas with sandy soils, shallow reinjection wells or trenches are used, in which gravity does the work in carrying the water to the aquifer. The extensive clay soils in the Powder River Basin and limited vertical transmissivity between geologic strata generally prohibit this. (Schneider, 2001)

     The Powder River Basin Resource Council (PRBRC, 2001), a conservationist group based out of Sheridan, Wyoming, worries that if industry adopts reinjection on a widespread basis, they will be inclined to inject water into shallow aquifers to save money, which would make it likely that much of this water would reappear as springs, basically negating the effects of reinjection (Morrison, 2001). Injecting into shallower aquifers would require a careful examination of where the coal seam outcrops and soil permeability, as well as predicted water flow (horizontal and vertical), to ensure that springs do not pop up everywhere (Wheaton, 2001). Writing to Wyoming's Environmental Quality Council, Jill Morrison of PRBRC also argues that permits and monitoring requirements be put into place to ensure that injection is done in an environmentally responsible manner. Wheaton also worries about the potential for the formation of new springs, particularly in areas where the coal seam outcrops at or near the surface (Wheaton, 2001).

     Thomas Schneider (2001) argues that surface discharge poses severe and irreversible threats to the surface environment and that reinjecting CBM discharge water is the best approach towards preserving surface waters and aquifers. Schneider argues that the best method is to reinject water back into the coal seams which are the producing zones. Oil companies have been reinjecting water back into aquifers since the 1970s, to dispose of water and to guard against surface flooding (Schneider, 2001) and to enhance recovery of oil (Rice and Nuccio, 2000). The BLM's (Bureau of Land Management) Onshore Oil and Gas Order Number 7 - Disposal of Produced Water - states that:

     "All produced water from Federal/Indian leases must be disposed of by (1) injection into the subsurface; (2) into pits; or (3) other acceptable methods approved by the authorized officer, including surface discharge under NPDES permit. Injection is generally the preferred method."

     While Schneider and environmental organizations such as PRBRC and the Northern Plains Resource Council (NPRC) in Montana feel that this should apply to the disposal of CBM product water, it was written with gas and oil production in mind (Schneider, 2001). Injection to a deeper aquifer could also be problematic, as water would be lost to the original coal aquifer (Schneider, 2001). Again, this is something that would have to be dealt with on a site by site basis. If CBM drawdown causes streams used for irrigation and drinking or livestock wells to dry up, injecting into an aquifer beneath the coal seam will most likely do nothing to restore these surface water sources (Schneider, 2001). To make this deeper water accessible, industry may be required to dig deep wells after gas production ceases, in order to ensure that local residents have a usable and immediate water source.

     Schneider also argues reinjection is the only way to preserve the hydrostatic balance in the coal bed aquifers, and argues for reinjecting water in a closed loop system into the producing zone where the water came from. Many engineers involved in the CBM process argue that this would totally defeat the entire idea behind CBM extraction, that is drawing down the pressure to release the CBM gas adsorbed to the coal's surface by water pressure. If the pressure head drops below the top of the coal seam, there might be space to reinject water. Otherwise, reinjection into the producing CBM seam becomes very expensive and technically difficult, if not impossible (Bauder,2001). A possible solution to this would be to reinject water into a lower formation at the onset of drilling, in order to get the necessary head loss, and then reinject water into the coal seam later on. This would only work if the deeper aquifer is not already saturated (Bauder, 2001). Schneider argues that reinjecting water to areas either above or below the aquifer avoids surface impacts but does not deal with the depletion of the coal bed aquifer itself. Most of the drinking wells in the area come from either coal seams or aquifers above the coal seams (Bauder, 2001).

     Another potential problem of injecting into alternate sub-surfaces is that the water chemistry of the receiving formation may be substantially different from that of the coal seam, causing chemical reactions or plugging. While Schneider states that injection should not degrade the quality of water in the receiving formation, this is generally not a concern in eastern Montana and Wyoming, as the water quality is generally better (lower TDS/ SAR) in the coal seams than it is in either shallower or deeper formations (Bauder,2001).

     Perhaps mechanical ways of periodically sectioning off the coal aquifer can be found, allowing for simultaneous drawdown and reinjection, although such methods are hypothetical and most likely expensive. Further testing is necessary to determine if there is a possibility that producing and reinjection wells can be placed far enough apart so that they do not effectively cancel each other out. Many wells produce high water/gas ratios, and it is these wells that Schneider argues should be converted to reinjection wells. These wells are obviously highly permeable to water and would have high reinjection rates. Wells which are structurally lower than other wells, are on the edges of reservoirs, or other depleted areas are also likely reinjection candidates.

     For the Gillette injection well system , which receives water from seven production wells, construction cost was about $1 million, and maintenance costs were about 1 penny/ barrel, or $625,000/year. Because the water is going into a public water supply, federal regulations require that it must be treated. Both chlorine and ultraviolet treatments are being used (chlorine treatments were less expensive to install but more expensive to maintain) (Dobkins, 2001). According to MSU Soil/ Water Quality Specialist Jim Bauder, these treatments are generally unnecessary, and are done because EPA regulations state that any water entering a public water supply must be treated with chlorine or ultraviolet radiation, whether or not harmful contaminants actually exist.

Netherlands case study on depleted aquifer recharge and application to CBM

     Subsidence is a major concern in areas where groundwater levels have dropped dramatically (Bouwer, 1988). This is particularly important in urban areas, in which an inch or two or subsidence may cause massive structural damage. This was initially a concern in relation to CBM mining. However current projections are that CBM product water withdrawals are not likely to cause subsidence in the Powder River Basin (WY DEQ, 2001). A study in The Netherlands addresses this problem using innovative and cost effective reinjection technology. Instead of using expensive mechanical wells, which are not only much more expensive to install, but are also costly to maintain, this study used a totally different technique. Infiltration wells were installed directly in the lakes around the Markerwaard district to recharge the depleted aquifer. This approach utilized the natural head loss between the surface water in the lakes and the piezometric head in the deep ground water to drive the infiltration of water. Also, the aquifer had already been depleted through drainage, leaving ample space to inject the water back into the aquifer (Hebbink, 1988).

     Large-scale drainage in the Markerwaard district of the Netherlands has resulted in drawdown of the piezometric head of groundwater. This has caused concern about subsidence and structural damage to surface buildings and infrastructures. Similar to the situation with CBM in the Powder River Basin, a major concern was the amount of money it would take to reinject the large amounts of water involved. Industry people involved in coal bed methane mining in eastern Wyoming argue that reinjection may be too expensive to make it a feasible option (Dobkins, 2001). In the Markerwaard situation, the head difference between the surface impoundment and the receiving aquifer is what drives the reinjection and saves a tremendous amount of money.

     As the CBM extraction process generally results in large volumes of water being produced initially, one feasible way to potentially deal with CBM product water is to dig a large pit, essentially a small reservoir at a site where water is going to be discharged. From an engineering perspective, the greater the elevation difference between the water source and the receiving aquifer, the greater the infiltration rate. The large initial amounts of water could be used to fill up the reservoir. Infiltration wells could be constructed with a mechanism to keep the wells sealed off until the reservoir has been filled, at which point they could be opened and water allowed to infiltrate. To avoid excessive sedimentation, the entrance of the well should somewhat higher than the bottom of the reservoir. As salts will tend to settle in the bottom of the reservoir, placing the well's entrance closer to the surface may further reduce the salinity of the water going back into the ground. Depending on where the water is targeted for reinjection, the injection wells could be used in conjunction with some sort of treatment to clean up the water prior to it being injected or reinjected. For instance, if it is going into a public water supply, ultraviolet or chlorine treatments would be applied. If it is going to shallow aquifers used for irrigation, some sort of treatments to remove salts, and particularly sodium, could be applied ( EPA, 2001).

     The Netherlands approach offers potential for providing cost-effective ways of getting water back into the ground, assuming the right or acceptable physical criteria can be defined. It is potentially useful for any injection operation, but seems well suited for dealing with CBM product water. Infiltration wells along these lines may well provide a viable alternative for dealing with CBM product water, as it appears to offer a reasonable compromise between the economic concerns of industry and the environmental concerns of landowners and scientists.

Modeling reinjection patterns

     Many of the same problems encountered with traditional pumping wells apply to reinjection wells. These include: water quality decline, pumping water level decline, lowering of specific capacity or yield, complete loss of production, sand/silt clogging, chemical incrustration, biofouling plugging (bacteria and other sources), pump or well corrosion and power-related pump aging and failure. In terms of CBM, the potentially highly saline water may lead to salt encrustration and clogging of transmission pipes. Wells must also be built to withstand the high initial levels of product water discharge, and keep functioning well when water levels drop off. Some ways to prevent or minimize deterioration are using designs which maximize efficiency and seek to minimize incrustation, using materials which resist corrosion and biofouling, using better designed and more efficient pumps, careful construction, sealing, and disinfection, and thorough testing to see how a well is likely to hold up over time (Smith, 2000).

     One of the most difficult problems with reinjection is that it is very difficult to predict where water will go once it enters the ground. The capacity of individual aquifers and how they are connected is also not very well understood. Recent efforts in computer modeling are beginning to make headway in understanding reinjection patterns. Generally, confined aquifers with a uniform thickness are easier to model than unconfined aquifers because their transmissivity does not vary. Aquifers with varying thicknesses are inherently more difficult to model and the consequences of reinjection are difficult to predict (EPA, 1991). Both the difficult science and expense in mapping aquifers and interlying areas account for this. For instance, Pennaco and other companies reinjecting CBM discharge water are basically pumping water into the ground and when they can no longer get water into the ground, they assume they have reached infiltration capacity or have maximized storativity (Dobkins,2001).

Summary and Conclusions

     While reinjection technology is currently being used in a multitude of ways, there is still much to learn, particularly about how to simultaneously reinject and withdraw water from saturated confined aquifers or those with positive pieziometeric head, as is the case with CBM reinjection. With very real concerns about reduction and contamination of groundwater supplies, it would seem prudent to invest in gaining a better understanding of subsurface water. The role that reinjection plays in mitigating the effects of CBM development to groundwater resources is certainly not insignificant at present, but the role it plays could certainly be expanded upon.

     Injection technology may play a role in the Powder River Basin, as a means of disposing of CBM product water, although whether or not this will include reinjecting into the coal seam is uncertain. Our current understanding of CBM aquifers and the need to control the piezometric head within these confined aquifers poses obstacles to reinjection and must be understood better if this is a route industry looks at seriously. The two main problems associated with CBM mining, saline/sodic waters questionable for surface discharge and aquifer drawdown, can potentially both be solved through reinjection. The loss of soil structure caused by sodium and the damage done to plants by excessive salts would obviously not be an issue if product water was not applied to or disposed of on the surface. Aquifer drawdown may still be an issue, particularly in areas in which high pumping well concentrations make reinjection into the coal seam very difficult. Selective reinjection following careful mapping out of the aquifers in the area, in combination with maps of wells and springs in the area, could minimize the drying up of critical surface water supplies. While the role reinjection will play in the Powder River Basin is uncertain, innovative techniques such as those employed in the Netherlands study, in conjunction with good mapping and engineering, may contribute to environmentally responsible disposal of CBM product water.

1Steve Custer, Earth Sciences Professor, 118 Traphagen Hall, MSU-Bozeman, Bozeman, MT 59715, uessc@montana.edu.
2Jim Bauder, Soil/Water Quality Specialist, 806 Leon Johnson Hall, MSU-Bozeman, Bozeman, MT 59715, jbauder@montana.edu.
3Terry Dobkins, Pennaco Vice-President. Pennaco Energy. Inc., 1050 17th St., Denver Colorado 80265, tdobkins@pennaco.com.
4John Wheaton, Groundwater Specialist, Montana Bureau of Mines/Geology 1300 N. 27th St., Billings, Montana, 59101, jwheaton@mtech.edu.

References

Bouwer, Herman. 1988. Systems for artificial recharge of groundwater. Artificial Recharge of Groundwater - Proceedings of the International Symposium. Annaheim, California. American Society of Civil Engineers. United States. 1-3.

Brooks, Patrick, Billy Kornegay, and Kevin Wood. 2001. Membrane systems can generate reuse-quality water cost-effectively. WET. 13: 36-40.

EPA. 1991 Handbook of ground water; Volume II, Methodology Environmental Protection Agency (Washington, D.C., U.S.) 1-25. < http://www.cepis.ops-oms.org/muwww/fulltext/repind46/models/models.html. 7/23/01

EPA. 2001 Subclass of Class V Underground Injection Control Wells, Title 40 of the Code of Federal Regulations (40CFR), sections 144.6 and 146.5. Region 5 Water Division section maintained by Steve Ray. <http://www.epa.gov/safewater/uic/cl5study.html. 7/22/01

Gill, Robert M. 2001. Massachusetts Military Reserve Groundwater treatment technologies: Fact Sheet 2001-01, March. <http:// www.mmr.org/cleanup/tech/gwtech01.htm#ctr 7/27/01

Hebbink, Arnold J. 1988. Infiltration wells in the lakes around the future Markerwaard polder. Artifical Recharge of Groundwater - Proceedings of the International Symposium. Annaheim, California. American Society of Civil Engineers. United States. 582-591.

Lee, Fred. 1996 Comments on Recharging to the Local Aquifer, LEHR Site Landfill Leachate - Poluted Groundwaters. <http://www. members.aol.com/dscso/1996roth8.htm 7/16/01

McCarthy, Terry. 2001 Plumbing the pasture. Time Magazine. July 16: 22-24.

Morrison, Jill. 2001 Powder River Basin Resource Council letter to Wyoming's Environmental Quality Council. Postel, Sandra. 2001 Growing more food with less water. Scientific American. 284:2. 46-51.

Powder River Basin Resource Council Website. 2001. <http://www.powderriverbasin.prbrc.org 8/13/01

Regele, Steve, and Judd Stark. 2000. Coal-Bed methane Gas Development in Montana, Some Biological Issues. Presented Sept. 1 at "Interactive Forum on Surface Mining Reclamation Approaches to Bond Release: Cumulative Impacts Assessment (CHIA) and Hydrology Topics for the Arid and Semi-arid West. Coal-bed Methane Workshop." Sponsored by USDI Office of Surface Mining, Denver CO; the Montana Department of Environmental Quality, Helena, MT; Montana Bureau of Mines and Geology, Butte, MT. <http:www.state.mt.us.coalbedmethane.pdf.fnl_cbm_txt3.PDF

Rice, Cynthia, and Vito Nuccio. 2000. Water produced with coal-bed methane. USGS Fact Sheet FS-156-00. <energy.cr.usgs.gov/oilgas/cbmethane/production.htm 8/15/01.

Schneider, Thomas J. 2001 Coal Bed Methane Produced Water Reinjection. <http:www.nprcmt.org/media/2001/reinjection%20main%20 page.asp>. 8/15/01.

Smith, Stuart A. 1998-2000. Prevention and maintenance for water supply wells: What goes wrong, how to prevent it, control it , and fix it. <http://www.groundwatersystems.com/m&rsemnr.html>. 7/14/01

TDEC (Tennessee Department of Environment and Conservation) Website. 2001. <http://www.state.tn.us/environment/dws/uicbroch.htm >. 7/09/01

U.S. Geological Survey (USGS). 2001. <http://www.usgs>. 8/13/01.

Wyoming Department of Environmental Quality (DEQ). 2001. <http://www.wsgsweb.uwyo.edu/

Home Page

Questions/Comments: waterquality@montana.edu