Vadose zone is an important sink and source of terrestrial carbon, and the foundation of agriculture and ecosystems. Therefore, it is important to predict the impacts of climate change on physical conditions of the vadose zone. In the meanwhile, various General Circulation Models (GCMs) have been developed. Though they are beneficial in predicting point-scale processes under climate change, still some downscaling schemes are required because of a big difference in spatial resolution (Wilby et al., 2004). Here, we applied the GCM projections to predict future soil moisture and thermal profiles. An arable bare land in the western suburb of Tokyo covered by Andisols was selected for the study site. HYDRUS-1D (Simunek et al., 2008) ver. 4.09 was used for the simulation of soil water and heat transport. Projections by MIROC ver.3.2 coded by Center for Climate System Research of the University of Tokyo, the National Institute for Environmental Study, and the Frontier Research Center for Global Change with SRES A1B were downscaled with cumulative distribution function method (Iizumi et al., 2010), and were used as boundary conditions for the HYDRUS. Soil hydraulic and thermal properties were determined experimentally (Kato et al., 2011). In the projected future, soil moisture as well as downward water flux increases in June due to an increase in rainfall. In September, drying of soil surface will be remarkable because of a decrease in rainfall and an increase in evaporation flux. Though global warming may raise soil surface temperature, rise in temperature would be alleviated within a shallower depth in September than that in June. It should be noted that these predicted results reflect both meteorological conditions and the soil physical properties such as high hydraulic conductivity, water retention, and low thermal conductivity.