Identifying the Driving Factors of Urban Land Surface Temperature

Abstract

Land surface temperature (LST) has a profound impact on urban climate and ecology, and is widely used to quantify surface urban heat islands. The spatial heterogeneity of LST is affected by natural and human factors, with seasonal differences. This study selected Dongguan, a rapidly urbanizing city in China, as an example to analyze the relationship between the spatial heterogeneity of LST in different seasons and influencing factors in six dimensions. Multi-source spatial data were combined, including Landsat images, meteorological data, digital elevation models, National Polar-Orbiting Partnership Visible Infrared Imaging Radiometer Suite nighttime light, and points of interest. The results show that spatial patterns of LST across different seasons were consistent, although there were local differences. Based on the GeoDetector model, the result indicated differences between separate effects and interactive effects, and identified the high temperature risk areas.