The UNSAT-H model was developed at Pacific Northwest National Laboratory (PNNL) to assess the water dynamics of arid sites and, in particular, estimate recharge fluxes for scenarios pertinent to waste disposal facilities. During the last 4 years, the UNSAT-H model received support from the Immobilized Waste Program (IWP) of the Hanford Site's River Protection Project. This program is designing and assessing the performance of on-site disposal facilities to receive radioactive wastes that are currently stored in single-and double-shell tanks at the Hanford Site (LMHC 1999). The IWP is interested in estimates of recharge rates for current conditions and long-term scenarios involving the vadose zone disposal of tank wastes. Simulation modeling with UNSAT-H is one of the methods being used to provide those estimates (e.g., Rockhold et al. 1995;Fayer et al. 1999).To achieve the above goals for assessing water dynamics and estimating recharge rates, the UNSAT-H model addresses soil water infiltration, redistribution, evaporation, plant transpiration, deep drainage, and soil heat flow as one-dimensional processes. The UNSAT-H model simulates liquid water flow using Richards' equation (Richards 1931), water vapor diffusion using Fick's law, and sensible heat flow using the Fourier equation.This report documents UNSAT-H Version 3.0. The report includes the bases for the conceptual model and its numerical implementation, benchmark test cases, example simulations involving layered soils and plants, and the code manual. Version 3.0 is an enhanced-capability update of UNSAT-H Version 2.0 (Fayer and Jones 1990). New features include hysteresis, an iterative solution of head and temperature, an energy balance check, the modified Picard solution technique, additional hydraulic functions, multiple-year simulation capability, and general enhancements. This report includes eight example problems. The first four are verification tests of UNSAT-H capabilities, three of which are repeats of the tests used for previous versions of UNSAT-H. The first test examines the ability of UNSAT-H to simulate infiltration compared to separate analytical and numerical solutions. This test was repeated using the modified Picard solution technique. The second test examines the ability of UNSAT-H to simulate drainage compared to measurements and a numerical solution. The third test examines the ability of UNSAT-H to simulate heat conduction compared to an analytical solution. The fourth test is new for UNSAT-H and examines the ability of UNSAT-H to simulate hysteresis compared to measurements and a numerical solution. The results of all four tests showed that the tested processes were correctly implemented in UNSAT-H. The repeat of the first test with the modified Picard solution technique successfully demonstrated a 10 4 to 10 5 reduction in mass balance errors.The second four example problems are demonstrations of real-world situations. The first three are repeat problems from previous versions of UNSAT-H. The first demonstration involves a 1-year s...