AQUIFER PROPERTIES DETERMINED FROM 2 ANALYTICAL SOLUTIONS

In the analysis of pumping test data, the quality of the determined aq uifer parameters can be greatly improved by using a proper model of th e aquifer system, Moench (1995) provided an analytical solution for fl ow to a well partially penetrating an unconfined aquifer, His solution , in contrast to the Neuman solution (1974), accounts for the noninsta ntaneous decline of the water table (delayed yield). Consequently, the calculated drawdown in these two solutions is different under certain circumstances, and this difference may therefore affect the computati on of aquifer properties from pumping test data. This paper uses an in verse computational method to calculate four aquifer parameters as wel l as a delayed yield parameter, alpha(1), from pumping test data using both the Neuman (1974) and Moench (1995) solutions. Time-drawdown dat a sets from a pumping test in an unconfined alluvial aquifer near Gran d Island, Nebraska, were analyzed. In single-well analyses, horizontal hydraulic conductivity values derived from the Moench solution are lo wer, but vertical hydraulic conductivity values are higher than those calculated from the Neuman solution. However, the hydraulic conductivi ty values in composite-well analyses from both solutions become very c lose. Furthermore, the Neuman solution produces similar hydraulic cond uctivity values in the single-well and composite-well analyses, but th e Moench solution does not. While variable alpha(1) seems to play a ro le in affecting the computation of aquifer parameters in the single-we ll analysis, a much smaller effect was observed in the composite-well analysis, In general, specific yield determined using the Moench solut ion could be slightly higher than the values from the Neuman solution; however, they are still lower than the realistic values for sand and gravel aquifers.