Abstract: Based on six land use and land cover change (LUCC) datasets, urban and suburban meteorological observation data and a Landsat TM image, this study analyzed the changes of land surface temperature and near surface air temperature with Beijing urban sprawl. Firstly, the urban land surface temperature and the urban underlying surface types were extracted by singlewindow inversion algorithm and linear spectral mixture analysis model. Then, six LUCC datasets and two meteorological observation datasets were used to analyze the impact of urban sprawl on the regional thermal environment. The results showed that the value of land surface temperature in a pixel mainly depended on the fraction of impervious surface of the pixel. The warming effect of impervious surface for land surface temperature was greater than the cooling effect of the vegetation fraction in the pixel. In the process of urban sprawl, urban heat island intensity increased along with the urban sprawl continuously in early stage. However, urban heat island intensity might be stable and even descend under a certain condition.

Key words: natural spruce-fir forest, dominant tree species, competition