Accurate and reliable estimates of groundwater flow and contaminant transport models are dependent on an understanding of the aquifer properties used to create the models. The borehole flowmeter has been used with increasing frequency at a variety of sites to produce high resolution vertical hydraulic conductivity (K(z)) distributions [Boggs et al. 1990; Rehfeldt et al. 1989b; Molz et al. 1989, Boman et al. 1997; Dinwiddie et al. 1999]. In theory, the validity of measurements obtained using borehole flowmeters is contingent on the hydraulic head gradients near the well at each discrete depth resulting from the pumping-induced flow having reached quasi-steady-state. Previous studies to predict the hydraulic head gradients near a well under pumping conditions have been predicated on various assumptions and have resulted in conflicting estimates of the length of time required for these gradients to reach quasi-steady-state.

This study models hypothetical single-porosity, confined, multi-layer aquifers with a minimum of simplifying assumptions to gain further insight into near-well gradient behavior in aquifers. The challenge, presented through the comparison of models presented herein [Javandel and Witherspoon 1969; Ruud and Kabala 1996, 1997; Hemker 1999a, 1999b; Kabala and El-Sayegh 2002], is to create an independent model capable of accurately and reliably reproducing their bulk results while also addressing the smaller inconsistencies among them. The results of the modeling will be applied to flowmeter analysis so that semi-quantitative estimates of the time required for a system to reach the status where flowmeter readings are valid may be obtained for field testing.