A digital simulation of the glacial-aquifer system in Sanborn and parts of Beadle, Miner, Hanson, Davison, and Jerauld counties, South Dakota

Abstract

The drought in South Dakota from 1974-76 and the near-drought conditions in 1980-81 have resulted in increased demands on the groundwater resources within many of the irrigated areas of the James River basin in eastern South Dakota. These increases in demand for irrigation water from the glacialaquifer system and continued requests to the State of South Dakota for additional irrigation well permits have created a need for a systematic watermanagement program to avoid overdevelopment of this system in the James River basin. The aquifer system which can be unconfined at shallow depths and confined at greater depths in the same section has a thickness ranging from less than 10 feet to greater than 200 feet and an average hydraulic conductivity of 316 feet per day. Calculated storage coefficients of the aquifer system range from 0.00039 to 0.000017 and specific yield values are as great as 0.28. Calculated recharge rates to the unconfined aquifer range from 0.9 to 3.4 inches per year and to the confined aquifer, 0.24 to 0.72 inch per year. Evapotranspiration, which accounts for most of the natural discharge from the aquifer, was estimated to be as great as 36.2 inches per year in some locations. An equally spaced grid containing 56 rows and 52 columns was used to simulate the glacial-aquifer system. The steady-state simulation was calibrated using water-level data collected before significant groundwater development (before 1973). The aquifer was also simulated in 11 annual transient stress periods from 1973 through 1983 and in 12 monthly transient stress periods for 1976. The simulated predevelopment potentiometric heads were compared to average water levels from 32 observation wells to check the accuracy of the simulated potentiometric surface. The average arithmetic difference between the simulated and observed water levels was 1.68 feet and the average absolute difference was 4.38 feet. The nonpumping steady-state simulated water budget indicates that recharge from precipitation accounts for 97.1 percent of the water entering the aquifer and evapotranspiration accounts for 98.2 percent of the water leaving the aquifer. The sensitivity analysis of the steady-state model indicates that the model is most sensitive to reductions in recharge and least sensitive to changes in hydraulic conductivity. In the annual transient simulation, recharge, evapotranspiration, and pumpage were adjusted annually. The maximum annual recharge varied from 1 0.10 inch in 1976 to 8.1M inches in 1977. The potential annual evapotranspiration varied from 29.9 inches in 1982 to M8.9 inches in 1976. Withdrawals from the glacial-aquifer system increased 2.6 times between 1975 and 1976. Since 1976, the pumpage has fluctuated annually in both distribution and quantity, however, the maximum annual withdrawals have not increased significantly since 1976. The average annual arithmetic difference between the simulated and observed water levels ranged from-3.88 feet in 197 1* to 2.23 feet in 1982; the average absolute difference ranged from M.70 feet...