Home

Current Activity Sales All about us

Areas of Interest Carbon Dioxide Coal Coalbed Natural Gas Gemstones Geologic Mapping Geophysical Modeling GIS Groundwater Industrial Minerals Metals Modeling Oil and Gas Surface Water / Climate Surficial Processes Uranium

Copyright Privacy Policy

WSGS Oil and Gas Group

Subsidence potential related to water withdrawal in the Powder River Basin

Recent concerns have been raised about the potential of coalbed-methane-related water withdrawal in the Powder River Basin (PRB) to induce ground surface subsidence. While it is true that withdrawal of water or oil from the subsurface can cause the surface to subside, the amount of subsidence and the time over which it occurs are the most important factors in determining if measurable damage will occur.

Documented Cases

A number of documented cases exist where fluid withdrawal has caused ground subsidence. One of the most famous cases is Mexico City. The city is directly underlain by unconsolidated and saturated lacustrine (fresh water lake) clays.Beneath the clays are thick saturated deposits of unconsolidated sand and gravel beds interbedded with clayey silts. The primary water for the city is derived from the sands and gravels, and when water is pumped from them, the overlying clays are slowly dewatered (Figueroa Vega, 1984). The porosity (volume of voids divided by the total unit volume of soil or rock) in the unconsolidated clays is high, and when water is removed they can compress significantly. There had been so much water pumped from the sand and gravel beds that between 1940 and 1960, the resulting dewatering and compression of the overlying clays led to as much as 27 feet of ground surface subsidence (Poland, 1984).

In the United States, ground surface subsidence related to fluid withdrawal has been documented at a number of localities. The best known localities include the San Joaquin Valley in California; Las Vegas, Nevada; New Orleans, Louisiana; and Houston, Texas.The common geological tie between the sites is that all are underlain by saturated, unconsolidated sands and gravels with interbeds and/or overlying beds of saturated clays. Water or oil is being removed (pumped) from the sands and gravels, and effects similar to those in Mexico City have been observed.

Powder River Basin Conditions

The geologic conditions in the PRB are not the same as those observed in the cases cited above. The bedrock underlying the surface is compacted and consolidated.Instead of loose sand, sandstone is present; instead of unconsolidated clay, shale is present. The porosity of saturated, unconsolidated clay can be as high as 88% (Poland, 1984), whereas the porosity of shale is usually in the 0 to 10% range.The porosity of unconsolidated sand is usually in the 25 to 50% range, whereas the porosity of sandstone is usually between 5 and 30% (Freeze and Cherry, 1979). As sedimentary rocks are formed and compressed, the void space, which can hold water, is decreased.

Even saturated bedrock, such as sandstone, can compress if water is removed under certain conditions. If an aquifer is buried and overlain by relatively tight shale, the aquifer is said to be confined. The weight of the overlying rock and the fluids it contains is supported by both the water in the aquifer and the bedrock itself. In the case of a buried and confined sandstone aquifer, if water under pressure is removed from the sandstone, then the sandstone could be further compressed by the weight of the overlying rock and water. The degree of compression depends upon the amount of water removed (and the resultant decrease in water pressure) and upon the strength and elasticity of the bedrock (Poland, 1984). Removing the water will decrease its hydrostatic head, which is the level to which the water will rise if a well were drilled into the confined aquifer. In reality, if an aquifer is confined, the water released from storage when the hydrostatic head declines comes primarily from compression of the aquifer and to a minor extent, from expansion of the water.

Coal beds can also serve as aquifers because water can flow through fractures or cleats as well as along bedding or depositional planes within the coal. If the coal is buried, confined, and saturated, it will also compress when water is removed. Some (but not all) water is being pumped from specific coal beds in the PRB to allow for the release of coalbed methane. The water pressure in the coals has to drop below a certain level to bring about the release of the gas. If all the water were pumped from the coal, the micropores in the coal would probably close, which would stop most of the flow of methane through the coal. Coalbed methane producers try to keep enough water in the coal beds to prevent this from happening.

Compression Calculations for Aquifers

Hydrologists can use a relatively simple formula to estimate how much a confined aquifer may compress when water is removed. The analysis starts with an understanding of certain hydrologic parameters. The storage coefficient, S, is defined as the volume of water that an aquifer releases from storage per unit surface area of the aquifer per unit change in head (Heath, 1987). The volume of water released is for the entire thickness of the aquifer under the unit surface area. In other terms, S= (volume of water) / (unit area x unit head change).

The actual formula for calculating the amount of aquifer compression that occurs when water is withdrawn from an aquifer has been derived by Edgar and Case (2000).The change in aquifer thickness or compression (DB), is equal to the storage coefficient (S) times the change in head (Df). In other words, DB=S x Df. The storage coefficient is determined through a well monitored pump test, and the change in head is obtained by monitoring declines in water level over time. The compressibility of an aquifer decreases with increasing depth (Edgar and Case, 2000).

To summarize, in order to calculate the amount of aquifer compression that may occur, it is necessary to have data on both the storage coefficient and the head decline for the aquifer or system under investigation.

Subsidence Potential for the Powder River Basin

In the coalbed methane producing areas of the PRB, a widely used storage coefficient figure for both the coals producing coalbed methane and for underlying sands providing water for the town of Gillette is 1.0 x 10-4 (Larry Wester, Wester-Wetstein & Associates Consulting Engineers, personal communication). The proposed head declines for coal beds that may be or already are being produced for coalbed methane will vary from area to area. The Bureau of Land Management (1999) presented a series of maps that showed modeled drawdown (head decline) for a number of scenarios. In the Gillette area, the maximum modeled coalbed-methane-related drawdown for the upper Wyodak coal is approximately 150 feet. The modeled drawdown for the lower Wyodak coal is approximately 250 feet. Applying the formula above, the aquifer compression for the upper Wyodak coal would then be 0.015 feet, and the aquifer compression for the lower Wyodak coal would then be 0.0250 feet. The sum of the two compressions would equal 0.04 feet, or slightly less than 1/2 inch.

Because of the strength of materials above the coals, it is thought that only a part, if any, of the compression would be observed at the surface. It is expected that such subsidence would be uniform over the area, and would not result in significant damage. Significant quantities of water have already been pumped from sandstones underlying the Wyodak coal zone for the Gillette water supply.To date, no surface subsidence has been observed or associated with this water withdrawal. The storage coefficient of those sandstones is 1.0 x 10-4, and the measured drawdown is approximately 400 feet (Larry Wester, Wester-Wetstein & Associates Consulting Engineers, personal communication). These data have the same values as those projected for the Wyodak coal in the Gillette area.

Summary

It appears that minor aquifer compression up to 1/2 inch may occur in the coal beds that are being developed for coalbed methane in the Gillette area. That entire compression, however, may not be transmitted to the surface. To date, no surface subsidence has been associated with other equally significant water withdrawals in the Gillette area.

References cited

Bureau of Land Management, 1999, Wyodak coal bed methane project Draft Environmental Impact Statement: Bureau of Land Management, Buffalo Field Office, May, 1999, 289 p.

Edgar, T.V., and Case, J.C., 2000, Pumping induced settlement of aquifers: Wyoming State Geological Survey Preliminary Hazards Report PHR 00-1, 9 p.

Figueroa Vega, G.E., 1984, Case history No. 9.8 Mexico, D.F., Mexico, in Poland, J.F., editor, Guidebook to studies of land subsidence due to ground-water withdrawal: Studies and Reports in Hydrology 40, prepared for the International Hydrological Programme, Working Group 8.4, United Nations Educational, Scientific, and Cultural Organization (UNESCO), Paris, France, p. 217-232.

Freeze, R.A., and Cherry, J.A., 1979, Groundwater: Prentice-Hall, Inc., New Jersey, 604 p.

Heath, R.C., 1987, Basic ground-water hydrology: U.S. Geological Survey Water Supply Paper 2220, 84 p.

Poland, J.F., 1984, Mechanics of land subsidence due to fluid withdrawal, in Poland, J.F., editor, Guidebook to studies of land subsidence due to ground-water withdrawal: Studies and Reports in Hydrology 40, prepared for the International Hydrological Programme, Working Group 8.4, United Nations Educational, Scientific, and Cultural Organization (UNESCO), Paris, France, p. 37-54.