Improved monitoring of urbanization processes in China for regional climate impact assessment

Regional climate is influenced by land surface processes through energy exchange between land and atmosphere at various scales. The performance of climate model simulation is largely influenced by land cover parameterization, especially over areas that experience rapid change of land surface characterization. Accurate land cover datasets suited for climate modeling are urgently needed to improve model parameterization for better simulation. In this study, fused urban land cover datasets have been developed by combining multi-source urban land cover products based on cross-comparison and detailed validation from medium resolution images during recent decades. Fractional cover values at different spatial scales were aggregated from integrated land cover datasets to derive improved estimates of the urban extent of big cities in China. Urbanization in China as captured by the new dataset can be divided into steadily and rapidly increasing periods with the urban area increasing by 2.5 and 6.7 % per year, respectively. Rapid urbanization mostly happens in the urban fringe, especially in eastern China, with mass conversion of cropland and woodland to urban land. An improvement in climate model simulation could be achieved using the fused data applied in climate modeling for examining the impact of rapid urbanization on the regional climate.