Bedrock aquifers in the Denver basin, Colorado; a quantitative water-resources appraisal

Abstract

The Denver metropolitan area is experiencing a rapid population growth that is requiring increasing supplies of potable water to be pumped from bedrock aquifers in order to meet demand. In an effort to determine the ability of the aquifers to continue to meet this demand, the Colorado Department of Natural Resources, the Denver Board of Water Commissioners, and Adams, Arapahoe, Douglas, Elbert and El Paso Counties joined with the U.S. Geological Survey in undertaking a hydrologic evaluation of the ground-water resources of the basin. This involved mapping of aquifer extent, thickness, structure, hydraulic characteristics, and water-level and water-quality conditions. This enabled ground-water modeling techniques to be used to simulate aquifer response to various pumpage estimates and ground-water development plans. The Laramie-Fox Hills aquifer (the deepest aquifer) underlies the 6,700-square-mile study area and is overlain by the more permeable Arapahoe aquifer, the Denver aquifer, and the Dawson aquifer, which crops out in the southern part of the study area. It is estimated that 260x10 6 acre-feet of recoverable ground water are in storage in these four bedrock aquifers. However, less than 0.1 percent of this volume of water is stored under confined conditions. The larger volume of water stored under unconfined conditions will be available for use only when the water levels in the confined aquifers decline below the top of the individual aquifer, allowing water-table conditions to develop. Annual precipitation on the Denver basin supplies an average of 6,900 cubic feet per second of water to the area; about 55 cubic feet per second of this recharges the bedrock aquifers, principally through the Dawson Arkose. The direction of ground-water movement is generally from ground-water divides in the southern part of the area northward toward the margins of the aquifers. Pumpage has ranged from about 5 cubic feet per second in 1884 to about 41 cubic feet per second in 1978. Pumpage exceeds recharge in the metropolitan area and has caused water-level declines (1958-78) to exceed 200 feet in a 135-square-mile area of the Arapahoe aquifer southeast of Denver. A quasi-three-dimensional finite-difference model of the aquifer system was constructed and calibrated under steady-state and transient-state conditions. Steady-state calibration indicated that lateral hydraulic conductivity within the aquifers is about 100,000 times larger than the vertical hydraulic conductivity between the aquifers. Transient-state calibration indicated that between 1958 and 1978, 374,000 acre-feet of water was pumped from the aquifers, producing a 90,000-acre-foot net decrease in the volume of water in storage in the aquifers. During this time, pumpage also changed the rates of interaquifer flow, induced additional recharge, and caused capture of natural discharge. Three 1979-2050 pumpage estimates were made for use in simulating the effects of various ground-water development plans. Simulations using each of these pumpage estimates indicate that by the year 2050 large water-level declines could occur, particularly in the deeper aquifers. Maximum water-level declines of 410, 1,700, and 1,830 feet were produced using the small, medium, and large pumping rates. Four plans for supplementing the Denver water supply include pumping a satellite well field, pumping a municipal well field, pumping to irrigate parks, and injecting water during periods of low demand for later use during periods of peak demand. Model simulation of these plans indicates that the satellite well field will yield twice as much water as the municipal well field, but will produce larger and more widespread water-level declines in the four aquifers. The municipal well field would not significantly affect water levels in the Dawson aquifer. Pumping the Arapahoe aquifer to supply irrigation water to selected parks was shown to produce only small water-level declines in the aquifer. Results of simulating injection-pumpage well fields at two locations indicate that simulated injection rates could range from 1.7 to 10 cubic feet per second, depending on the choice of site. The volume of water that could be stored in the bedrock aquifer is, thus, sensitive to the hydrologic characteristics of the chosen site. More study is needed to evaluate water-chemistry compatibility of native and injected water.