WATER STORAGE KEY FACTOR IN COALBED METHANE PRODUCTION

Brian J. Luckianow
Taurus Exploration Inc.
Birmingham, Ala.

William L. Hall
Dames & Moore
Atlanta

Storage ponds provide a cost-effective means to temporarily retain water produced with coalbed methane and permit gas production during times when stream flow rates drop,

Normally, water produced with the gas is run into nearby streams, with the dilution rate closely monitored and controlled by environmental agencies. During low stream flow in the Black Warrior basin, Ala., large volumes of produced water must be stored to prevent shut-in of coalbed methane fields.

Taurus Exploration Inc. constructed such production water facilities for the Cedar Cove field to eliminate periodic field shut-ins as a result of excess water production.

The effectiveness of such a storage approach is governed by receiving stream flow variability, production water flow characteristics, and the economics of storage pond construction.

Typically, the major tributaries of the Black Warrior basin, as well as the Black Warrior River itself, are capable of assimilating during the winter and spring far more production water than current plans can generate.

However, during the summer and fall, stream discharges, especially within the tributaries, can approach near-zero flow.

During periods of insufficient flow for dilution within a receiving stream, field shut-ins can be avoided only through transfer of production water to other tributaries with excess capacity or through storage.

Due to the relatively large volumes of production water to be discharged into the Black Warrior over the next several years, periodic exhaustion of assimilative capacity even within the Black Warrior is almost a certainty. Storage will be necessary, however, providing it is dependent upon the cost of storage relative to the potential revenue loss and/or well damage associated with shut-in. Typically, two alternative designs are available-lined storage and slurry wall sealing.

The slurry wall and hydraulic barrier concepts are similar in that both involve a nonsynthetic, nondegradable, virtually indestructible barrier to production water migration. In addition, special protective measures are unnecessary.

Taurus' detailed assessment of alternatives shows the slurry wall/hydraulic barrier to be a technically attractive option in some instances. Regulatory resistance to unusual techniques as well as contractor inexperience, however, create limitations to economical installation.

Design requirements, hydrogeologic constraints, and regulatory conditions affect the installation and economics of the hydraulic barrier method. Barring regulatory constraints associated with the uncertainties born of inexperience, generally this storage sealant approach is highly economical if the proper geologic conditions can be found.

Lined storage designed in accordance with industry standards for industrial lagoons is the conventional approach most recognized by the regulatory agencies. As of the summer of 1990, no true storage facilities had been constructed within the Black Warrior basin. Limited storage within treatment ponds in most cases provides no more than a surge capacity for 1-2 days.

Synthetic-lined, large-scale storage is more troublesome than treatment ponds due to the necessity of maintaining the pond storage capacity. An empty pond exposes the liner material to various forms of degradation, including ultraviolet deterioration and liner destruction from surface disturbance. Protection of the liner generally involves an additional protective layer, such as washed sand. Protection on top of the layer increases the attention that must be paid to the liner subbase. Cover activities also raise the potential for liner penetration.

STORAGE REQUIREMENT

Economic development of coalbed methane calls for a production water management strategy that allows year-round operation of the field.

CRITERIA

Use of in-stream dilution as a disposal technique is critical to the economics of methane gas recovery. Alternatives such as deep well disposal or treatment by removal of the dissolved solids in most instances are technically and economically impractical. Dilution as a management technique, however, requires close coordination of the production water flow with the variable assimilation capacity of the field's receiving streams.

The capability of a receiving stream to assimilate production water discharge is a function of stream flow seasonal variation, production water chloride concentrations, sub-basin specific production water discharge variability, and permit criteria. The actual assimilative capacity available within a particular field varies significantly over both the seasons and the life of the project. Typically, critical periods will be encountered during project start-up and the low flow months of August, September, and October.

Storage requirements are partially controllable with respect to field production schedules. Parameters such as competitive use of a particular basin, water quality, and well-specific flow rates are uncontrollable, however.

BLACK WARRIOR BASIN

Table 1 indicates the flow characteristics of the Black Warrior basin as well as the variation between sub-basins.1 Flows within the Black Warrior River itself are highly regulated in the immediate vicinity of the methane recovery area by the Army Corps of Engineers (ACOE) locks and dams at Holt and Bankhead. Since 1968, the low flow regimes (i.e., 7Q10 or 10-year, 7-day average low flow) have more than doubled from the 150 cfs range to over 300 cfs.

The low flows that could create the need for extended curtailment of production water discharges are limited to the August-October time frame. Because of regulation by the ACOE, limited periods of extreme low flow can be experienced any time of the year. With shut-in of the locks and power houses, river flow may be reduced to seepage which averages in the range of less than 100 cfs. However, extended periods (more than 7 days) of insufficient flow for assimilation of all the production water currently discharging into the main item is unlikely.

Use of tributaries is much more problematic than use of the Black Warrior because the variation is extreme both between basins and between seasons within a single basin. Table 1 shows the characteristics of three basins which cover the north to south geographic range of the methane recovery area. Except for Valley and Big Sandy creeks, most of the tributaries approach a near-zero flow for periods of up to several months.

The potential storage for a particular field is a function of both the average low stream flow and the variation within a drought produced by short-term storms. Even during a low-flow period, storm cells over the basin periodically raise the water flow for short durations (Fig. 1). Consideration of these bursts is critical in the planning for storage needs because they allow the recovery of storage capacity periodically during the storage period.

PRODUCTION WATER

Four parameters must be considered with the production water flow:

• Per well flow rates

• Well flow rate variation over the life of a project

• Water quality

• Total discharge within the basin.

Based on data developed by Taurus for fields that cover the extent of the basin, flow rates may range from less than 10 b/d to over 1,500 b/d The likely average start-up flow across the basin is in the range of 150 b/d.

Variation is increased further by the impact of long-term pumping of a field. A typical well reduces flow rate over time from approximately 375 b/d to approximately 80 b/d (Fig. 2). Data covering more than 2 years of pumping in the northern portion of the basin indicate flows will decrease to the range of 1020 b/d. Initial data from the southern portion of the basin indicate long-term flows are likely to remain much higher, in the range of 200 b/d.

Production-water quality is also a significant issue, specifically with respect to chloride concentrations. Chloride concentrations range from less than 1,000 mg/l. to over 30,000 mg/l.

The impact of average per well flow rates and chloride concentrations on the Black Warrior basin assimilative capacity is shown in Table 2. Estimates are the minimum total flow required within the basin to allow full production water discharge under different combinations of average water quality and well flow. The estimates assume 3,500 operating wells.

This estimate was developed by Taurus in the summer of 1990 as the number of wells likely to be drilled by the end of 1990. The actual number of wells that may ultimately be brought into operation cannot easily be estimated from either drilling reports or permits. Production economics are undefined for many existing wells, and the impact of shifts in the economy cannot be effectively predicted.

With the available estimates, however, significant storage is likely to be required within the basin even without consideration of the impact of production-water distribution and regulatory constraints.

Capacity may be available, but due to the mixing zone constraints and the lack of even distribution throughout the basin of points of discharge (POD), basin-wide shutdown of Tier 11 discharges is possible.

REGULATORY CONSTRAINTS

The National Pollution Discharge Elimination System (Npdes) regulatory program defines the allowable mixing zones, the edge of mixing-zone allowable chloride concentrations, and the criteria for discharge shut-in. The mixing zone and shut-in criteria create the potential for discharge termination despite available capacity within the river.

Specifically, permits issued since the spring of 1990 contain the provision that the mixing zone may only encompass 50% of the stream width. As a result, without multiple PODs by an individual discharger, the full capacity of the stream cannot be utilized. Up to four PODs with extremely effective vertical and horizontal mixing are required to obtain up to 90% of the available capacity.

STORAGE ALTERNATIVES

Four alternatives for lined storage are plastic liners, bentonite-clay liners, bentonite slurry walls, and hydraulic barriers.

A plastic liner has the following drawbacks:

• High maintenance/repair problems

• Labor-intensive installation requiring rigorous quality control

• Perforation by objects or animals

• Degradation when subjected to alternating wet/dry exposure

• Photodeterioration

• Strict slope requirements to avoid liner gaping

• High capital cost.

A standard bentonite-clay liner has the following drawbacks:

• Permeability

• Permeability of existing soils can influence installation

• Amount of clay is dictated by fill material needed to achieve impermeability

• Installation depends on slope

• Limited chemical resistivity to chloride attack.

The advantages of a clay liner are:

• Maintenance

• Less rigorous quality control requirements

• Unaffected by sharp objects because of thickness

• No deterioration when exposed to wet/dry cycles because of good shrink/swell qualities

• No photodeterioration

• Modification possible to prevent deterioration from chloride or sulfate attack

• Somewhat less expensive to obtain and install.

Uncontaminated groundwater levels and gradients surrounding the impoundment must be higher than the expected maximum operating water level. The impoundment and groundwater beneath it must be underlain by a relatively impermeable and unfractured rock stratum.5

The concept assumes some moderate but uncompromising degree of exchange will take place between production water in the impoundment and the underlying groundwater, whether by mechanical dispersion or molecular diffusion.6 The concept also requires cutoff or containment of any down gradient (downstream) migration of groundwater underlying the impoundment.7

CONVENTIONAL LINER

Flexible membrane liners, or geomembranes, have been used for dependable seepage control in many surface impoundments including reservoirs, landfill liners and caps, tank farm liners, and canals. Over 20 different materials in over 50 thicknesses and reinforcement formats are offered as pond and pit liners. Sales claims and technical arguments are made for and against virtually all of the available liner selections.

The impermeability of synthetic membranes, in combination with a proper subbase, protective cover, and installation control, provides a highly secure sealant. The permeability of synthetic materials is less than 10-12 cm/sec. In contrast, good clays have a permeability of about 10-6 cm/sec, and bentonite-treated soils have a permeability of about 10-6-10-9 cm/sec.

Based upon these low permeabilities, plastic linings theoretically can be effective in thicknesses of 10-30 mil and still be less permeable than soils of great thickness. The concern, however, lies in long-term integrity and installation failures. The issues of concern are material selection, fabrication, and installation.

In contrast to the preceding methodologies, the hydraulic barrier approach depends on maintaining a positive hydraulic gradient into the storage facility.3 The hydraulic barrier requires the presence of a freshwater barrier with a potentiometric surface higher than the maximum level of proposed storage.

For the typical sites available within the Taurus field, storage is most economically obtained with embankments across natural or mine reclamation drainage swales. In such instances the hydraulic barrier must be implemented in combination with an effective seal of the dam. If a sufficiently impermeable clay or unfractured rock is present within 80 ft of the surface, a bentonite slurry wall can be effective and economical.4

MATERIAL SELECTION

Good engineering practice suggests that if more than one solution exists for the same problem, the best choice is generally the simplest, most economical, or most efficient. For example, an earth-covered, 20-mil PVC liner (properly installed) may cost $0.50/sq ft, including the placement of the earth cover. Under the right circumstances this may be a better choice than an exposable liner that might cost $1.00/sq ft. As an example, a 6-mil, earth-covered liner installed 27 years ago at the Bermuda Dunes Clubhouse lake in Bermuda Dunes, Calif., is still in service. The same liner system failed after only 6 months when tried with an industrial waste effluent containing an acid that attacked the mastic and tape seams.8

It is imperative that an operator investigate the particular characteristics of a liner system through examination of existing installations. Operational capabilities of any given system should not be assumed based on sales or product specification data that may not address specific installation conditions.

FABRICATION

Along with recommended installation procedures, each product has a minimum material properties specification which lists various test procedures and corresponding test values a given material must possess to be considered a good quality product.

Specifications should require that a sample be taken from every tenth roll or each 10,000 lb of roll stock, whichever is more frequent, and submitted for quality assurance testing. The results of this testing should be obtained by the buyer and kept on file for future reference.

In addition, the specification should require that every roll of material be unwound and both sides inspected for defects prior to fabrication. All welds should be continuously inspected and an inspector's stamp or other means of inspection verification periodically applied along each weld during the fabrication process. Test welds should be made and tested every 4 hr of operation during fabrication.

The preceding steps will ensure a properly manufactured and properly fabricated liner. Although extensive testing increases cost, it is preferable to correcting leak seams or a material failure once a pond has been put into service.

INSTALLATION

To achieve the desired impermeability in synthetic membranes, the material must be free of holes, and all seams should be watertight. This means that the manufactured material must be of the highest quality and that the installation and field seaming must be accomplished with utmost quality control.

There is always a concern, however, regarding liner placement. It is extremely important that the liner be installed with supervision to ensure that the earth surface upon which the liner is placed has been adequately prepared, properly graded, and compacted and that all large stones and sharp objects that could puncture the liner have been removed. The large panels, which can be fabricated up to 100 ft in width and 200 ft or more in length, are then spread over the prepared surface and joined together with a solvent adhesive.

Earth-moving equipment should not be driven directly on the liner to avoid damaging it when covering with earth fill. When placing the earth cover on the liner, the material should be deposited on the liner and pushed forward over the liner to obtain at least 1-2 ft of earth cover while the equipment is operating over it.

The installation specification should require that the liner installer submit a list of at least 10 references with names and phone numbers of the proper contact at each job site. The 10 projects should comprise at least 1,000,000 sq ft of successfully installed liner.

Example seams should be made at the beginning and middle of each shift. These seams are then tested and the results made available to the owner. Seam strength should equal 80% of the specified parent material strength after allowing 7 days of cure time.

It is essential that an inspector be on the job site during the installation to inspect all seams and seals to penetrations. Once the installation is complete, a leakage test should be conducted either by the head loss method or by a monitoring system.

HYDRAULIC BARRIER

Selection of a potential site was conducted through a comprehensive review of the area within and surrounding the Taurus well field (Figs. 34). Review of hydrogeological conditions at the prospective site indicated that groundwater conditions lent themselves to containment through an hydraulic barrier. In addition, existing subsurface clay strata are sufficiently extensive to enable use of a dam core or slurry wall to prevent downstream or vertical migration.

Results of foundation investigations showed the prevailing hydraulic gradients upstream of the proposed dam site are all positive, i.e., all groundwater flows are toward the center or thalweg of the proposed impoundment. The foundation investigations also indicated the presence of a nearly horizontal, uniform, unfractured, and very impermeable shale/sandstone layer approximately 510 ft below the ground surface of the valley floor throughout the area underlying the proposed impoundment.

Given that the core of the proposed dam would be constructed of an impermeable bentonite slurry wall into the shale/sandstone layer or into an impervious hard clay just above it, any down gradient groundwater flow would be cutoff; thus, conditions for containment of any dispersing of stored, treated production water were satisfied. Dispersion of the stored water would be prevented by the positive, opposing hydraulic gradients.

A conservative estimate of the likely diffusion rates has shown diffusion of the water during the limited time of storage, a maximum of 90 days, would be negligible and would be contained within the impounded area. Given the limited areal extent of the impoundment and the limited duration of storage, any losses of stored water to the underlying groundwater within the impoundment area would be minimal.

Following draining of the impoundment, the positive, opposing hydraulic gradients will flush any diffused water back to the surface. Any emerging groundwaters high in chlorides, as monitored via the down gradient surface water monitor, will be piped to the Hurricane Creek POD if necessary.

Regulatory concern over the hydraulic barrier/slurry wall concept centers on the lack of absolute proof the hydraulic barrier will prevent migration of chlorides into the surrounding groundwater aquifers, and the relationship between potential impact, risk of impact, and technical and economic difficulties entailed in avoiding potential impact.

A considerable amount of information exists about the use of the hydraulic barrier concept. In most cases, the concept is used very effectively to mitigate or prevent the effects of salinity intrusion in coastal areas.

In the metropolitan areas of New York and New Jersey, injection of highly treated wastewater into groundwater along the coast is providing an effective barrier against intrusion of saline ocean waters into freshwater aquifers. In the Midwest, the hydraulic barrier concept was combined with installation of a bentonite clay slurry wall to form an effective impoundment and seal against the migration of hazardous/toxic waste into the groundwaters underlying the Department of Defense's Rocky Mountain arsenal. In nearly all cases, the hydraulic barrier concept is reported technically equal, or superior to, more conventional liners and has the additional advantage of being very cost effective.

COSTS

A comparison of the alternative costs can best be based on actual quotes for each of the facilities. A quoted price for the site civil work is available for a lined treatment pond of approximately 3.5 acre-ft and a slurry wall pond of 55 acre-ft. For comparison purposes, the costs are reduced to the common basis of cost/acre-ft.

A summary of the unit cost based on actual quoted figures is shown in the accompanying box. Only the site civil work is included since the electrical and mechanical monitoring and construction management are essentially the same. This is not strictly true because the cost of monitoring for the hydraulic barrier slurry wall will be higher. However, the amount is unknown until the Alabama Department of Environmental Management (ADEM) provides comments and review of any lined treatment pond design.

The unit cost for construction of a lined pond based on a total capital cost of $35,000-50,000 is from $10,000 to $14,300/acre-ft. This is derived using a storage pond volume of 3.5 acre-ft as estimated from constructed dimensions.

As designed, lined treatment ponds may or may not be acceptable to ADEM. It is likely, because of the size of the storage pond, that a permit modification will be required. Submittal of a design report would provide an opportunity for ADEM to tighten the standards for design.

It is typical for a storage pond of this nature to have a sub-base or a geofabric as well as a cover material of washed sand to prevent damage to the liner. This is specifically required when a pond is not to be kept filled. The total cost/acre-ft if the pond is designed to engineering standards may rise to $20,000-25,000 (Fig. 4).

The unit cost for site civil work for the hydraulic barrier slurry wall facility discussed here was approximately $12,600/acre-ft for the initial 55 acre-ft storage capacity. The unit cost for the specific facility will be reduced to less than $5,000/acre-ft as the storage volume is increased through excavation to its maximum capacity of 170 acre-ft.

It is likely that in most instances topography will dictate that several ponds must be constructed to reach the desired capacity for the single slurry wall lined facility. In such an instance, the mechanical and electrical costs will increase accordingly to discharge and transfer production water into and from the storage ponds.

Both approaches would be technically equal if the lined pond is provided with a protective cover to prevent disintegration or damage to the liner. Although lined treatment ponds currently are not equipped with monitors within the Black Warrior basin, it is likely that the increased sensitivity on the part of ADEM will result in the requirement of groundwater monitors in the vicinity of the storage facility. It is unlikely, however, that the monitoring requirements will be as extensive as those required for the slurry wall facility.

Cost for the hydraulic barrier/slurry wall will be very competitive in situations in which appropriate topographic and hydrogeologic conditions are accessible from a field.

REFERENCES

1. Earth Info Inc., USGS daily values-Eastern United States, 1989.

2. Chenisenoff, P.N., and Young, R.A., Pollution Engineering Practice Handbook, Ann Arbor, 1976, p. 542.

3. Hall, L.E., "Control of Ground Water Contamination in an Alluvial Fan Aquifer by a Dual Hydraulic Barrier," Paper proceedings of the First National Outdoor Action Conference on Aquifer Restoration, Ground Water Monitoring and Geophysical Methods, National Water Well Association, Dublin, Ohio, 1987, pp. 125-42.

4. Pendrell, D.J., and Zeltinger, J.M., "Contaminated Ground Water Containment/Treatment System at the Northwest Boundary, Rocky Mountain Arsenal, Colorado," Proceedings of the Third National Symposium on Aquifer Restoration and Ground Water Monitoring. National Water Well Association, Worthington, Ohio, 1983, pp. 453-61.

5. Skaggs, R.L., "Model Study of Selected Mitigative Strategies to Control Radionuclide Migration in Ground Water," The Fourth National Symposium and Exposition on Aquifer Restoration and Ground Water Monitoring, The Fawcett Central, Columbus, Ohio, May 23-25, 1984, pp. 151-61.

6. Shafer, J.M., "Determining Optimum Pumping Rates for Creation of Hydraulic Barriers to Ground Water Pollutant Migration," National Technical Information Service, DE84-013245, Springfield, Va., 1984.

7. Zheng, C., Bradbury, K.R., and Anderson, M.P., "Role of Interceptor Ditches in Limiting the Spread of Contaminants in Ground Water," Ground Water GRWAAR. Vol. 26, No. 6, November-December 1988, pp. 734-42.

8. Cain, R., "How to Select and Specify a Pond Liner," Pollution Equipment News, 1988.

Copyright 1991 Oil & Gas Journal. All Rights Reserved.