One of the primary uses for the Nebraska Cooperative Hydrology Study (COHYST) groundwater-flow models is to aid regulators, resource managers, and groundwater developers in assessing the interaction of surface water and groundwater in the Platte River Basin. Of particular importance is the influence of new and existing pumping wells on river flows. “Stream depletion” refers to either direct removal of water from a stream or indirect capture of aquifer water flowing to the stream by pumping wells. The stream depletion factor (sdf) is a concept developed by C.T. Jenkins (1968) that has been widely used in groundwater/surface-water management. This technique determines stream depletion using analytical techniques. The sdf has units of time and indicates when the volume of stream depletion equals 28% of the volume removed by a pumping well. To compute the location of the analytical 28% depletion sdf line, only aquifer transmissivity, specific yield, and distance to the stream are needed, but the method assumes a highly idealized system. However, this approach does not address variable geologic and hydrogeologic characteristics nor boundary conditions other than a straight stream. Alternatively, the line of 28% depletion by volume can be determined with a numerical groundwater-flow model, which accounts for real-world, complex hydrologic, geologic and boundary conditions. Preliminary, single-layer COHYST models with variable, distributed parameters and 1 square-mile cells were used to determine the 40-year depletion lines for the Platte River in the central COHYST area. The simulated well location was adjusted until the line of 28% depletion could be determined. The preliminary model indicated that the stream depletion line ranged from 0.8 miles to 16.5 miles from the Platte River in the area tested. The model lines were compared to sdf lines produced in a 1982 study (Missouri Basin States Association). On the south side of the river, the model-produced lines are similar to the earlier study, with the largest deviation less than 5 miles. North of the river, greater distances between the lines estimated by the two methods existed, with the largest deviations being nearly 7 miles near where the model accounted for a tributary that probably was not accounted for in the previous study.