The town of Yucca Valley located in the southwest part of the Mojave Desert in southern California relies on groundwater pumping from the Warren groundwater basin as its sole source of water supply. This significant dependency has resulted in a large imbalance between groundwater pumpage and natural recharge, causing groundwater levels in the basin to decline more than 90 m from the late 1940s to 1994. Consequently, an artificial recharge program proposed by the Hi-Desert Water District, which provides water service to the town of Yucca Valley, was implemented for the purpose of recovering the groundwater levels; however, the rise in groundwater levels has caused nitrate (NO3)(NO3) concentration to increase simultaneously. The purpose of this study is to develop an optimal pump and recharge strategy for a planned conjunctive-use project to remove the high-nitrate concentration while maintaining groundwater levels at desired elevations at specified locations as well as meeting water demand. An optimization/management model is formulated with a linear objective function and nonlinear constraints. The response matrix approach is used to link the optimization model with the simulation model. Because of nonlinearity, the response matrix is updated and iteration is required for convergence. A systematic scheme is also developed for finding a feasible initial policy. Three different scenarios are considered in the management model. The results obtained from each scenario are analyzed and discussed.