Analysis on Relationship Between Urban Land Surface Temperature and Landcover from Landsat Etm+ Data

By selecting areas of inside outer-ring road in Shanghai, China as the study area, the paper selected Lands at ETM+ image and GIS land cover data as the major information. Urban land cover classification was firstly processed by using the maximum likelihood method from ETM+ visual and near infrared images combined with assistance of GIS land cover data. Then land surface temperature was retrieved from ETM+ thermal infrared data by using the mono-window algorithm. The statistical analysis and multiple parameters regression method were used to analysis the relationship between urban land surface temperature and land cover. Results revealed spatial variations of urban land surface temperature within different land cover. Built-up areas displayed larger temperature than water areas and green space. Furthermore influence degrees were different for various land cover types based on partial correlation and regression analysis, in which public building and service area increased land surface temperature stronger while green space decreased temperature clearer.