Parameter Estimation of Constant Head Test in the Unconfined Aquifer with Partially Penetrating Effect

• Main Authors: Hsiu-Chuan Kao, 郭綉娟
• Other Authors: Chia-Shyun Chen
• Format: Others
• Language: zh-TW
• Published: 2004
• Online Access: http://ndltd.ncl.edu.tw/handle/93582169287423114015

碩士 === 國立中央大學 === 地球物理研究所 === 92 === Transmissivity (T) values estimated using pumping test data normally vary with distance, suggesting the aquifer be heterogenous. However, we found that this result may be not so much due to the spatial variability of T as due to the data analysis method failing to account for field test conditions. Here, we use three different methods to analyze drawdown data produced from a constant-head pumping test in an unconfined till aquifer in Iowa. The saturated till thickness is about 2.5 meter. The water depth (the constant head) maintained in the pumping well was only 1.5 meter. There were four observation wells, and their maximum drawdown changes from 5.62% to 15.43% of the saturated thickness. Method I ignores the possible partial penetrating effect in the wells by assuming a fully penetrating condition for all the wells. The estimates of T vary relatively randomly, lacking a sound explanation. Method II assumes that the pumping well is partially penetrating while the observation wells are fully penetrating, and the spatial variability of the T estimates decreases. Method III takes into account the partial penetrating effect of the wells, and a constant T value is obtained using the drawdown data of all the observation wells. This constant T is also used to analyze drawdown data from a constant-rate test conducted in the same aquifer and the same wells. Considering the five observation wells are located within a short extent of 5 meters surrounding the pumping well, the T value being constant is plausible. For this particular case, it is thus concluded that the spatial variability of the estimates of T is an artifact due to the negligence of the partial penetrating effects of the pumping well and the observation wells. For the constant head data analysis, we develop a mathematical model which is suitable for the unconfined aquifer and involves four pertinent parameters; namely, the horizontal and the vertical hydraulic conductivity, the storage coefficient, and the specific yield.