Horizontal and vertical hydraulic conductivity, transmissivity, and storativity of the aquifer system at Centereach, New York, were estimated using analytical multiple-well aquifer test models and compared with results of numerical regional flow modeling and hydrogeologic framework studies. During the initial operation of production well S125632 in May 2008, continuous water-level and temperature data were collected at a cluster of five partially penetrating observation wells, located about 100 feet (ft) from S125632, and at observation well S33380, located about 10,000 ft from S125632. Data collection intervals ranged from 30 seconds to 30 minutes and analytical model calibration was conducted using visual trial-and-error techniques with time series parsed to 30-minute intervals. The following assumptions were applied to analytical models: (1) infinite aerial extent, (2) homogeneity, (3) uniform 600-ft aquifer thickness, (4) unsteady flow, (5) instantaneous release from storage with the decline in head, (6) no storage within pumped wells, (7) a constant-head plane adjacent to bounding confining units, and (8) no horizontal component of flow through confining units.
Preliminary estimates of horizontal and vertical hydraulic conductivity of 50 ft per day horizontal and 0.5 ft per day vertical were extrapolated from previous flow modeling and hydrogeologic framework studies of the Magothy aquifer. Two applications were then developed from the Hantush analytical model. Model A included only the pumping stress of S125632, whereas model B included the concurrent pumping stresses from two other production well fields (wells S66496 and S32551). Model A provided a sufficient match to the observed water-level responses from pumping, whereas model B more accurately reproduced water levels similar to those observed during non-pumping of S125632, as well as some effects of interference from the concurrent pumping nearby. In both models, storativity was estimated to be 0.003 (dimensionless) and the Hantush leakage parameter '1/B' was estimated to be 0.00083 ft-1. Representation of leakage across the overlying confining layer was likely complicated by: (1) irregularities in surface altitude and (2) groundwater recharge due to rainfall during the aquifer test.