More than 90 percent of the public water supply in the upper Rockaway River Valley in Morris County, New Jersey, is obtained from ground-water withdrawals from the valley-fill aquifers. During 1997, an average of 9.6 million gallons per day of ground water was withdrawn from these aquifers. The aquifer system consists of an unconfined aquifer (upper aquifer) and a locally confined aquifer (lower aquifer), which are composed of sands and gravels. These aquifers are separated by a discontinuous confining unit that consists mostly of silt and clay. Increases in ground-water withdrawals can induce movement of water from streams to wells, increase flow from the upper aquifer to the lower aquifer, and reduce base flow in the Rockaway River downstream. A ground-water-flow model was used to simulate and quantify the effects of current withdrawals on the valley-fill aquifer system under transient monthly conditions. Recharge over the model area varies both spatially and temporally. Part of model calibration consisted of adjusting percentages of monthly precipitation that recharges the valley-fill aquifer system. More recharge occurs during winter and spring than during summer and fall. This seasonal variation affects ground-water discharge to the Rockaway River. Ground-water withdrawals from the valleyfill aquifers also affect ground-water discharge to the Rockaway River. Three scenarios were simulated to observe the effects of ground-water withdrawals on ground-water discharge to the Rockaway River and to determine the extent to which variations in rates of withdrawals correspond to variations in rates of streamflow depletion. Streamflow depletion was estimated by comparing model-computed ground-water discharge for the three scenarios with the modelcomputed ground-water discharge under transient conditions. In scenario 1, all pumpage was removed from the model. In scenarios 2 and 3, 1 million gallons per day of ground-water withdrawals in excess of the current pumpage was withdrawn from the valley-fill aquifers. In scenario 2, the additional 1 million gallons per day of withdrawals were made from a hypothetical well located in the upper aquifer about 250 feet from the river. In scenario 3, the additional withdrawals were made from a hypothetical well located in the lower aquifer about 1,750 feet from the river. Results of scenario 1 indicated that the difference between the streamflow depletion and withdrawals is small; increases in ground-water withdrawals from the valley-fill aquifers correspond to decreases in ground-water discharge to the Rockaway River of approximately the same amount. Results of scenario 2 and 3 indicated that a lag time could occur between the introduction of withdrawals and the full magnitude of the effects of the withdrawals on streamflow depletion. A lag time of about seven months occurred for scenario 2 with the well placed in the upper aquifer. A longer lag time of more than 1.5 years occurred with the well placed in the lower aquifer and separated from the upper aquifer by a confining unit (scenario 3). Extreme low flow in the Rockaway River is mostly base flow. A flow-duration analysis of the Rockaway River at the surface-water gaging station upstream from the Boonton Reservoir during the drought of 1961-66 indicated that streamflow from the upper Rockaway River Basin alone might not be sufficient to meet the minimum passing flow of 7 million gallons per day during a drought. Under similar drought conditions today, during 3.2 percent of the drought time, streamflow at this station upstream from the reservoir would be less than the minimum passing flow requirement downstream from the reservoir.