A 24-hour aquifer test was conducted in Well Field 2 near Augusta, Georgia, October 21–22, 2009, to characterize the hydraulic properties of the Midville aquifer system. The selected well was pumped at a rate of 684 gallons per minute. At the initiation of aquifer-test pumping, water levels in each of eight wells monitored for the test were still recovering from the well-field production. Because water levels had not stabilized, data analyses were needed to account for the ongoing recovery. Hydraulic properties of the Midville aquifer system were estimated by an approach based on the Theis model and superposition. The Midville aquifer system was modeled as a Theis aquifer. The principle of superposition was used to sum the effects of multiple pumping and recovery events from a single pumped well and to sum the effects of all pumped wells as the estimated total drawdown at a monitored well. Simulated drawdown at each monitored well was determined by using a spreadsheet (SUMTheis) function of aquifer transmissivity and storativity. Simulated drawdown values were transformed into simulated water levels, accounting for longterm water-level trends. The transmissivity and storativity values that were used to calibrate the simulated water levels to measured water levels (roughly 4,000 square feet per day and 2E-04, respectively) provide estimates of the transmissivity and storativity of the Midville aquifer system in the vicinity of Well Field 2. The approach used in this study can be applied to similar well-field tests in which incomplete drawdown recovery or other known pumping is evident.