How to attain the high-resolution spatial climate grid data by high accuracy spatial interpolation method according to the climate observed data scattered over China, which can satisfied with the demands of ecological process, structure and landscape simulation model, and ecosystem integrated assessment model, is the focus issue in the field of ecological modeling. To address the above issue, the property feature data from 752 weather stations during the period from 1964 to 2007 are transferred into spatial feature data in terms of coordinate information supplied by every meteorological station meteorological data. SRTM DEM of China on a spatial resolution of 90m×90m is transferred into DEM on a spatial resolution of 1km×1km using the resampling method. The DEM is used as auxiliary data to develop the statistical transfer functions of mean annual temperature and precipitation. The spatial distribution trend model of mean annual temperature and mean precipitation are respectively developed by spatial statistical analysis method combined with the factors of elevation, longitude, latitude, aspect, and slope, which are integrated into the High Accuracy Surface Modeling (HASM) method. Spatial change trends of mean annual temperature and mean annual precipitation in China in the four periods of time of C1 (1964-1974), C2 (1975-1985), C3 (1986-1996) and C4 (1997-2007) are analyzed in terms of results from running these models. The simulation results indicate: (1) Mean annual temperature has shown an accelerating increase trend with the rate of speed of 0.28℃ per ten years since 1964, respectively rose 0.1631℃ during the period from C1 to C2, 0.3835℃ from C2 to C3 and 0.7007℃ from C3 to C4. (2) Mean annual precipitation increased by 4.7069mm during the period from C1 to C2, decreased by 2.1335mm from C2 to C3 and continuously decreased by 2.3735mm from C3 to C4. Furthermore, the conclusion can be got that the continuous and high-resolution climate data can be quickly attained from the observed climate data on scattered distribution by the HASM method integrated with the factors of elevation, longitude, latitude, aspect, and slope, and are available used to the parameters of spatially explicit multi-scale ecosystem model.