Groundwater withdrawals in Las Vegas Valley, Nevada, primarily for municipal supplies, totaled more than 2.5 million acre-ft between 1912 and 1981, with a peak annual withdrawal rate of 88,000 acre-ft in 1968. Effects of heavy pumping are evident over large areas of the valley but are more pronounced near the major well fields. Secondary recharge from lawn irrigation and other sources is estimated to have totaled more than 340,000 acre-ft during 1972-81. Resulting rises in water-level in shallow, unconfined aquifers in the central and southeastern parts of the valley have caused: widespread water-logging of soils; increased groundwater discharge to Las Vegas Wash and its tributaries; and potential for degradation of water quality in deeper aquifers by accentuating downward vertical hydraulic potential in areas where shallow groundwater has high concentrations of dissolved solids and nitrate. A 3-dimensional groundwater flow model of the valley-fill aquifer system was constructed for use in evaluating possible groundwater management alternatives aimed at alleviating problems related to overdraft and water-logging while maximizing use of the groundwater resources. Natural recharge to the valley-fill aquifers is about 33,000 acre-ft/yr; in 1979, an estimated 44,000 acre-ft of secondary recharge infiltrated to the near-surface and developed-zone aquifers. Peak water use for lawn irrigation during summer results in rates of secondary recharge that may increase threefold from winter rates. Simulated rates of seepage to washes in the valley increased correspondingly from an average of 850 acre-ft/mo in winter to about 1,300 acre-ft/mo in the summer. Groundwater withdrawals by pumping totaled 620,000 acre-ft during 1972-81, and model results indicate that about 190,000 acre-ft of that total was derived from storage. Use of the model as a predictive tool was demonstrated by simulating the effects of using most municipal wells only during the peak-demand season of June 1 through September 20. Results of the 9-year simulation indicated that: (1) long-term rates of water-level decline near the municipal well field would be less than rates for 1972-81, but the magnitude of seasonal fluctuations would increase, and (2) total volume of water released from storage as a result of subsidence would be only 42,000 acre-feet per year, or about half the volume during 1972-81.