Estimation of hydraulic parameters from pumping tests in a multiaquifer system

Abstract

The hydrogeologic conditions in Tianjin, China, are characterized by a system of multiple alternating aquifers and aquitards and a high groundwater level. Generally, the groundwater system to a depth of 70 m includes one phreatic aquifer and three semiconfined aquifers (AqI–AqIII). The distribution of aquifers varies greatly. Accurate estimates of aquifer parameters from pumping tests are essential in geotechnical engineering practice. In this study, the applicability of different models for estimating hydraulic parameters in a multiaquifer system is analyzed. To avoid errors in the graphic-analytical process and to enhance the efficiency and accuracy, a hybrid algorithm, called GALMA (a genetic algorithm (GA) combined with the Levenberg–Marquardt (LM) algorithm), is used with the Neuman and Witherspoon model and ratio method. The estimated values in the second and third semiconfined aquifers (AqII and AqIII) from various solutions are compared with the measured data. Further validations based on the prediction of the drawdown values are performed via a three dimensional numerical simulation.

Keywords

• Hydraulic parameter
• Leaky aquifer
• Pumping test
• Optimal approach