The former attempts to divide the influence on the observed near surface air temperature into a part by global warming and a part by regional warming related with urbanization mainly depended on statistical analysis. But this method is invalid in the region where no observed data are given. In this study, numerical simulations of regional climate in a summer day about 150 years ago were performed with a meso-scale model. LUIS (Land Use Information System) is a digital land use data set covering all Japan with 2 km grid. The land use on each grid point circa 1850, circa 1900, circa 1955 and circa 1985 are compiled in LUIS and it is published by UNEP/GRID-Tsukuba. By the numerical simulations referring to LUIS, the authors attempted to pick up the influence on the near surface air temperature by regional warming related with land use change during recent 150 years. This method is applicable to any region where no observed data of temperature are given.
The numerical simulation model used in this study was based on the Colorado State University Mesoscale Model (CSU-MM) (Pielke, 1974). Each rectangular area around Tokyo, Osaka, Sapporo, Fukuoka and Sendai with the horizontal scale of several hundred kilometers was divided into grid cells with the horizontal size of around 10 km. The surface boundary conditions in each grid cell, i.e., albedo, evaporation efficiency, roughness length, density, specific heat capacity, heat diffusion coefficient, were calculated by the method of a weighted average for each share of land use referring to LUIS. The surface parameters for each land use were determined referring to Anthes et al. (1987). The time step for numerical integration was chosen as 60 seconds and results were calculated for 48 hours from 0 AM on July 26 in 4 periods. The results were analyzed for 24 hours from 0 AM on July 27. Anthropogenic heat with its diurnal variation was given only to urbanized area. Its dairy average for circa 1985 was assumed to be 25 W m-2. For former periods, 75%, 50% and 25% of the intensity for circa 1985 were assumed for circa 1955, circa 1900 and circa 1850, respectively. To evaluate the potential warming related with land use change, sea surface temperature around each studied area was set to be constant during 4 periods. The results of this study are summarized as follows;
1) During 4 periods, the area showing the regional warming related with land use change has expanded. Especially, this feature was significant around Tokyo, Osaka, Fukuoka, where urbanization was significant.
2) In 4 cities except Sapporo, the maximum differences between circa 1850 and circa 1985 emerged at 9 PM and the minimum emerged at 6 AM. The former was 1.8 Celsius degrees in Tokyo (Otemachi) and 2.2 Celsius degrees in Osaka.