Estimates of ground-water recharge rates developed from hydrologic modeling studies are presented for 3-Springs and East Stewart basins. two small basins (analog sites) located in central Nevada. The analog-site studies were conducted to aid in the estimation of recharge to the paleohydrologic regime associated with ground water in the vicinity of Yucca Mountain under wetter climatic conditions. The two analog sites are located to the north and at higher elevations than Yucca Mountain, and the prevailing (current) climatic conditions at these sites is thought to be representative of the possible range of paleoclimatic conditions in the general area of Yucca Mountain during the Quaternary. Two independent modeling approaches were conducted at each of the analog sites using observed hydrologic data on precipitation, temperature, solar radiation stream discharge, and chloride-ion water chemistry for a 6-year study period (October 1986 through September 1992). Both models quantify the hydrologic water-balance equation and yield estimates of ground-water recharge, given appropriate input data. The first model uses a traditional approach to quantify watershed hydrology through a precipitation-runoff modeling system that accounts for the spatial variability of hydrologic inputs, processes, and responses (outputs) using a dailycomputational time step. The second model is based on the conservative nature of the dissolved chloride ion in selected hydrologic environments, and its use as a natural tracer allows the computation of acoupled, water and chloride-ion, mass-balance system of equations to estimate available water (sum ofsurface runoff and groundwater recharge). Results of the modeling approaches support the conclusion that reasonable estimates of average-annual recharge to ground water range from about 1 to 3 centimeters per year for 3-Springs basin (the drier site), and from about 30 to 32 centimeters per year for East Stewart basin (the wetter site). The most reliable results are those derived from a reduced form of the chloride-ion model because they reflect integrated, basinwide processes in terms of only three measured variables: precipitation amount, precipitation chemistry, and streamflow chemistry.