Assessing the impact of adjacent urban morphology on street temperature: A multisource analysis using random forest and SHAP

Abstract

Urbanization has significantly transformed land use patterns, intensifying environmental challenges such as the urban heat island (UHI) effect and increasing health risks in urban public spaces. Urban streets, as vital public spaces, frequently experience heat accumulation during summer due to various environmental factors. Existing research has focused primarily on microscale case studies, leaving the broader impact on adjacent areas unclear. Therefore, this study examines the influence of morphology features within street-adjacent buffers on street temperatures in Nanjing, China, utilizing multisource data and machine learning. The random forest algorithm, combined with the Shapley additive explanation (SHAP) interpretation, was applied to analyze the impact of adjacent street morphological features on the land surface temperature (LST) of streets. The results suggest that greenery, buildings, and surface morphology features within street-adjacent buffers are crucial in regulating street temperatures. Furthermore, this study explains the variations in the factors influencing the thermal environments of different typical street types via K-means clustering analysis. The findings offer insights for sustainable urban planning strategies aimed at mitigating extreme heat and enhancing thermal comfort in urban pedestrian spaces.