Spatiotemporal impact mechanisms of urban multidimensional form on land surface temperature: A case study of representative cities in five climate zones of China

Rapid urbanization and global warming have intensified the frequent occurrence of abnormal LST phenomena. Research on the regulation of thermal environment by urban multidimensional form (UMF) across different climate zones remains relatively limited, and the interpretation of the UMF’s characteristics is still incomplete. This study extracted LST data from five typical cities spanning various climate zones, innovatively constructed a 1D-2D-3D urban morphological indicator system, and utilized the XGBoost-SHAP model to reveal the response mechanisms of LST to UMF. The results indicate that: (1) The three cities with the highest multi-year average area proportion of high-temperature zones are Lhasa, Beijing, and Urumqi, with proportions of 18.00 %, 16.46 %, and 12.97 %, respectively. (2) DEM is the most important factor influencing LST in multiple cities, and the interaction between high DEM and low NDVI values in Shanghai exhibits a negative impact. (3) In Urumqi and Lhasa, the thresholds for the warming effects of NDVI and PWG are 0.50 and 0.95, respectively. In dry-climate cities like Beijing, the warming threshold for PIS (0.07) is lower than that in humid-climate cities like Shanghai (0.09 and 0.22). This study aims to provide insights for regulating urban thermal environment and enhancing climate resilience.