Study On The Spatial Characteristic Of Land Surface Temperature And The Relationship Between The Spatial Characteristic Of Land Surface Temperature And Land Surface Characteristic In Beijing Area

Landsat 7 Enhanced Thematic Mapper Plus (ETM+)dated on July 1, 1999 was used in this research. Firstly, the land surface temperature (LST) and the types of land use/land cover of urban and surrounding area of Beijing is retrieved by using RS and GIS technology. Then, the spatial characteristic of land surface temperature and the relationship between land surface temperature and characteristic of land cover was analyzed. This paper exposes that characteristic of land cover impacts on urban thermal environment, and provides scientific bases for reducing extend of the thermal environment deteriorating.The urban heat island ratio index (URI) was selected as the indicator to estimate Spatial Characteristic of Land Surface Temperature. The paper quantitatively analyzed the spatial characteristic of LST by calculating URI of Beijing area, including all administrated district, each zone between ring road as well as each transect buffer. The results show that difference of LST is significant in Bejing area. LST of most urban area (including East district, West district, Chongwen district and Xuanwu district) is higher than other regions, and LST of part urban area is the higest among all regions. UHI show multi-center distribution pattern, some UHI center is distributed the area beyond urban area. Among all administrative districts, the URI of Shunyi district and its mean LST is the highest, and the value is 0.6210, the mean LST is 38.08℃. The URI of urban area is 0.6118, its mean LST is 37.18℃. The mean LST is 36.97℃in the fifth ring road zone, the mean LST is 35.18℃beyond of the fifth ring road zone. Among all zones between ring roads, the highest URI is in the zone between the third ring road and the fourth ring road, the value is 0.6327, and the mean LST is 37.48℃. Among eight transect buffers, the URI of east transect and south transect buffer are higher than others, and the URI of west transect and north transect buffer are lower.In this research, a supervised classification with the maximum likelihood algorithm was conducted to classify the image, NDVI and NDBI was retrieved by band math. We analyze the relationship between LST among classes, NDVI, NDBI. And then, the effect of land cover characteristic on LST was exposed. The result show that LST is different if land cover classes is different, the LST of water and forest is lower, the LST of urban and boil soil is higher; The greater NDBI value is, the higher LST is, but the higher LST is, the smaller NDVI is, LST possessed negative correlation (r=0.7044); Different land cover have different effect on LST, water and green land will reduce extend of UHI , By covering with buildings, roads and other impervious surfaces, urban areas generally increase extend of UHI.