Abstract: Based on Landsat 8 remote sensing images with a 30m spatial resolution acquired in 2015 the Urban Development Area of  Wuhan, this paper retrieved Land Surface Temperature (LST）using atmospheric correction method. This study sought to analyze the spatial-temporal distribution of LST in urban development area, ecological green wedge and main urban area, so as to provide reference for the planning of urban spatial distribution and the mitigation effect of Urban Heat Island (UHI）more comprehensively and scientifically. The results show that: (1）Compared with ordinary least squares regression (OLS), linear stepwise regression (LSR）can find the multiple driving factors combination model. Compared with the results of LSR fitting, the R^2 value of GWR model is improved by 0.04~0.09. (2）The “high-high” cluster occurred mainly in the main urban area, the new town development area and the main traffic line covered with artificial surface, which was more prone to high temperature cluster. Furthermore, the “high-high” cluster had the largest number and proportion in summer, while the “low-low” cluster occurred mainly in the major lake water systems in four seasons. (3）The ecological land of the ecological green wedge had different cooling range. The cooling range was not obvious in spring, but was most obvious in summer. The temperature drop of each green wedge was more than 2℃. In a certain buffer range, LST increased with the distance from green wedge。When the buffer range reached a certain distance, LST tended to be gentle or downward with the increase of the distance from the green wedge. (4）Compared with previous studies, the explanation of LST change in landscape pattern is generally lower, which may be due to the increase of artificial surface area caused by rapid urbanization and the increase of artificial green space. So it makes the urban landscape pattern more fragmented and LST affected by a variety of interaction factors. (5）The driving factors of LST in four seasons were quite different in space. The regression coefficients of land cover and landscape pattern in summer were positive, and so were land cover, landscape pattern and human activities in winter. It indicated that these driving factors had obvious effect on integral warming under high or low temperature.