Three-dimensional (3-D) simulations of the spatial and temporal variations in chloride and strontium concentrations in porewaters were performed to constrain infiltration rates, flow paths, and mixing processes in the unsaturated zone (UZ) at Yucca Mountain, NV. Chloride concentrations in infiltrating water were calculated from aerial distributions of precipitation and infiltration rates for the current climatic conditions and for the last glacial maximum, combined with effective chloride concentrations in precipitation. Modeled concentrations are roughly similar to measured porewater chloride concentrations from the Paintbrush nonwelded tuffs in the Exploratory Studies Facility (ESF) tunnel and in boreholes suggesting that the mean infiltration rate over the site is unlikely to be higher than the calculated mean infiltration rate for the modern climate (∼5 mm/year; [Flint, A.L., Hevesi, J.A., Flint, L.E., 1996. Conceptual and Numerical Model of Infiltration for the Yucca Mountain Area, Nevada. Milestone 3GU1623M. U.S. Geol. Surv. Water Res. Invest. Rep. U.S. Geological Survey, Denver, CO]). Porewaters from the late Pleistocene (>10 ka) could be present in the Paintbrush bedded tuffs and in the underlying Topopah Spring welded tuffs (TSw), predominately under regions of thick alluvium having little infiltration. However, porewaters at the potential repository level may have a higher proportion of Holocene recharge due to the higher calculated infiltration rate in this region. Dual-permeability simulations show that in low infiltration regions chemical disequilibrium can exist between fracture and matrix porewaters, as a result of the climate change 10,000 years ago. Below the potential repository level, simulations show significant mixing due to lateral flow on top of the low permeability basal vitrophyre in the Topopah Spring unit and on zeolitized tuffs in the Calico Hills unit. Perched water chloride concentrations are closely matched using the calculated conditions for the last glacial maximum climate, with some component of Holocene recharge. Measured strontium concentrations in the UZ and in the perched water bodies can be roughly matched by assuming conservative behavior in nonzeolitic units and strong ion exchange in zeolitic units, and indicate that the perched water bodies are poorly mixed. Differences in the Cl contents of samples having a bomb-pulse Cl-36 signature and those with a modern ratio indicate that waters of intermediate 36 Cl/ Cl ratios may be mixtures, that without other isotopic data could be inferred as either Pleistocene or Holocene age waters.