Abstract: Nowadays, world urbanization accelerates greatly and caused some environment problems, including the urban heat island (UHI) effect, especially in developing countries. Taking Harbin city as an example, using four Landsat remote sensing images (1989, 2007 and 2013 by TM sensors, 2001 by ETM+sensor), based on RS and GIS, the land surface temperature (LST) was inverted, which was normalized, classified and figured out, while the spatial distribution characteristics and spatio-temporal change law of LST were analyzed. Using geography graphic information theory, the change characteristics of space, procedure and attribute of UHI were quantitatively analyzed, to explore the response relationship between urban expanding and UHI. The results showed that:with the expanding urban area of Harbin city, the fourth grade UHI effect was growing, which were taken accounted for 4.36%, 5.69%, 6.29% and 7.12% in area proportion, respectively. Meanwhile, they were mainly located in Daowai district and railway regions. The region of vegetation and water body had the lower LST, even much lower on the edge of that region. The continual change map has the largest area, while the least proportion area was the anaphasic change map, which accounted for the proportion of 33.30% and 7.30%, respectively.Meanwhile, during the process of the urbanization and urban expansion, using geography graphic information theory can provide accurate and enough information for monitoring and estimating urban heat island effects.