Non-Darcian effect on slug test in a leaky confined aquifer

Abstract

Summary Slug test has been commonly applied to determine the near-well aquifer properties because of its logistical and economic advantages over other aquifer tests. Instantaneously increasing or decreasing the water level inside a slug test well will generate a significant hydraulic gradient near the well screen-aquifer interface (particularly at early stage of slug test), which may result in near-well aquifer flow deviated from Darcian behavior. However, previous studies on slug test rarely considered non-Darcian flow process associated with the slug test. In this study, a transient model of slug test in a leaky confined aquifer was established considering non-Darcian horizontal flow in the aquifer, and Darcian and vertical flow in the aquitard. The non-Darcian flow in the aquifer was described by the Izbash (power-law) equation. A finite-difference solution was presented for solving the non-linear model of slug test, the accuracy of which was tested through comparison with a previous analytical solution. By analyzing the sensitivities of dimensionless aquitard leakance (αD), non-Darcian index (n), and ratio of well screen radius versus well casing radius (χ) on the water level recovery (WLR) of slug test, one could conclude that the rate of WLR increased with the decrease of n for a given time. The WLR curve of Darcian flow (n = 1) was above the curves of non-Darcian flow, including fully developed turbulent flow. There was a deflection point for each WLR curve without the aquitard leakance in log–log scales, and the slope of the curve approached a constant after such a point. The effects of aquitard leakance on WLR for non-Darcian case appeared later than Darcian case. The normalized sensitivity coefficients of WLR with respect to n, αD and χ were negative, and they were most sensitive to n and least to αD.