Application of an Analytical Ground Water Flow Model to a Pseudokarst Setting in a Surface Coal Mine Spoil

Abstract

Attempts were made to apply an analytical groundwater flow model (MINEFLO) to a reclaimed surface coal mine spoil in West Virginia, U.S.A., that exhibited characteristics of both a conventional porous medium and pseudokarst flow. The model utilizes the method of analytic elements to derive a steady-state solution to the ground water flow equation for a single aquifer. Two simulations were performed, using hydrologic features and parameter values determined from field observations, slug withdrawal tests (1989) and slug injection tests (1990). Simulated heads at base-flow conditions for 1989 were about 12% different ihan measured values, with a root-mean-square (RMS) en-or of 0.82 m. A similar simulation for I 990 produced a somewhat better match, with an average head difference of less than 10% (RMS e1Tor of0.66 m). However, in both cases, the simulated flow rate at the toe-of-spoil seep was over 75% less than the measured rate. This suggested that the hydraulic conductivity values obtained from the slug tests may have underestimated the overall hydraulic conductivity of the aquifer materials, perhaps because of the influence ofpseudokarst and/or fracture flow mechanisms. It was concluded that the model could predict (within 10%) the base-flow water levels at a spoil site where recharge is known to occur primarily from adjacent unmined strata. The use of the model's only transient feature (pumping well) at the study site is doubtful due to the large RMS en-or in the steady-state simulation.

Hydrogeologic Characterization of Local Scale Hydraulic Properties in a Weathered Acidic Minespoil

“…This large difference may help explain the discrepancy between test results and model requirements observed by Hawkins (1994) and Aljoe (1994). The Type land 2 response characteristics would not be observed in slug tests unless bails are removed abruptly and the response in the first 20 seconds of the test is measured accurately.…”

“…Several investigations have encountered difficulty in applying results of local well tests to parameterize groundwater flow models calibrated using observed hydraulic heads. Aljoe (1994) dismissed the applicability of local-well t~st data, indicating they underestimated measured flows by as much as 80% when used to parameterize an analytic-clement flow model. Hawkins (1994) had-to increase field-test values of hydraulic conductivity as much as 200 Limes to derive an acceptable model fit to match heads and spring flows.…”