Assessing the Impact of Natural Conditions/Socioeconomic Indicators on the Urban Thermal Environment Based on Geographic Big Data

Abstract

Understanding correctly the factors influencing the urban thermal environment is a prerequisite and basis for formulating heat-island-effect mitigation policies and studying urban ecological issues. The rapid urbanization process has led to the gradual replacement of natural landscapes by products of socioeconomic activities, and although previous studies have shown that natural conditions and socioeconomic intensity can significantly influence land surface temperature (LST), few studies have explored the combined effects of both on LST, especially at a fine scale. Therefore, this study investigated the relationship between natural conditions/socioeconomic and summer daytime LST based on big data and a random forest (RF) algorithm using the city of Jinan as the study area. The results showed that the spatial pattern of LST, natural condition characteristics of the city, and socioeconomic characteristics are consistent in spatial pattern and have significant correlation. In the RF model, the fitted R2 of the regression model considering two influencing factors reaches 0.86, which is significantly higher than that of the regression model considering only one influencing factor. In the optimal regression model, topographic factors in natural conditions and socioeconomic factors in buildings and roads are very important factors influencing the urban thermal environment. Based on the results, strategies and measures for developing and managing measures related to the thermal environment are discussed in depth. The results can be used as a reference for mitigating urban heat islands in the study area or other cities with similar characteristics.