Description
Transmissivity (T) is an important aquifer parameter which indicates how much water an aquifer can transmit. Numerical models may require large numbers of transmissivity values to accurately predict water level changes in an aquifer. Specific capacity (SC) data, defined as discharge drawdown (Q/s), are much more readily available than are transmissivity values since they can taken from drillers reports. Establishment of a useful relationship between T and SC would allow estimation of transmissivity using specific capacity data. Long-term well capacity (WC), or well yield, influences pump selection and placement, and determines adequacy of wells to meet water demands. A useful relationship between well capacity and specific capacity would provide an easy and inexpensive means of estimating well yield. The purpose of this investigation was to determine if useful relations exist between specific capacity and transmissivity and between specific capacity and well yield for wells completed in fractured rock aquifers. Linear regression analysis was used to determine the relationships between the variables. Frequency distributions and residuals were examined to validate use of the linear regression model. Frequency distributions of T, SC, and WC were all strongly skewed to the right indicating that the raw data were not normally distributed. Cumulative frequency distributions of logs of T, SC, and WC suggested that data is approximately log-normally distributed. The log of transmissivity was found to be linearly related to the log of specific capacity (r2 ≈ 0.80). Best fit lines through subsets of well test data (T vs. SC) including pump tests, step tests, specific capacity at 60 minutes (SC60), SC at 200 minutes (SC200), wells exhibiting delayed yield (DY), and wells without delayed yield (NDY) were plotted. No statistically significant differences existed between them, indicating that separating data into subsets was not necessary. Transmissivity estimates using specific capacity had a precision of about ± 0.6 log cycles, at a 95% confidence level. This relationship is useful since it significantly improves upon the range of transmissivities in a typical ground water basin. Linear regression analysis for log well capacity vs. the log of the product of specific capacity and maximum allowable drawdown were performed for all usable data, and various subsets, as with transmissivity. A strong relationship between log wC and log(SC.smax) (r2 ≈ 0.91) ·still had insufficient precision (± .44 log cycles) to make the relationship usable. Regressions through data subsets showed no statistically significant differences.
