A simulation of climatic change induced by land use modification from 1976 to 2006 on summer over Tokyo metropolitan area, Japan

Numerical simulations on the climatic change associated with the land use modification of 30 years are executed by using the Digital National Land Information dataset of 1976 and 2006, targeted on the summer of Tokyo metropolitan area, Japan. The model used for the simulation is the Japan Meteorological Agency Non Hydrostatic Model (NHM), with which the single-layered square prism urban canopy scheme is coupled. The simulations are executed with the same initial and boundary conditions of 31 days of August 2006. The land surface parameters from each land use information of 1976 and 2006 are only the difference. We assume that urban buildings had lower aspect ratios (height / width = 0.25) in 1976 and they grew up higher twice (four times in the center of Tokyo's 23 wards) in 2006. We add the spatially distributed anthropogenic heat to the simulation of 2006 case. On the other hand, we add no anthropogenic heat in 1976 case because of the lower penetration rate of cooling systems on those years.

From the differential of two cases, the monthly mean temperature rise at around the center of Tokyo (Tokyo's 23 wards) and along with some trunk railways, where the land use changed notably from vegetation (such as paddy rice fields and forests) to urban surfaces, is calculated over 1.3 °C to the maximum. On the other hand, the temperature rise of less than 1 °C could be seen at the center of Tokyo though the land use fractions of urban area didn't change in 30 years. As for the temperature rise in the center of Tokyo, it can be understood the effect of adding anthropogenic heat and also some effects of aspect ratio change of the buildings.

The slight negative trend of vapor pressure for 30 years can be seen from the simulated result. This negative trend is conflicting with the positive trend derived from the observational dataset though the significances of the observational trend are low.