The surface urban heat island (SUHI) exhibits pronounced spatiotemporal heterogeneity under rapid urbanization and climate change. However, the nonlinear response and interaction contribution of multiple factors to SUHI across different background climate zones remain underexplored. In this study, the summer SUHI trends and their responses to impact factors were investigated using an explainable machine learning approach across temperate, cold, tropical, and arid zones. The results revealed that (1) Multi-year averaged daytime SUHI was stronger in the southeast than in the northwest, and nighttime SUHI had a significant increasing trend in more than half megacities. (2) Daytime SUHI was mitigated by optimal water body fractions in temperate zones, lower GDP in cold zones, reduced specific humidity in arid zones, and higher electricity consumption in tropical zones. Night-time SUHI mitigation depended on higher humidity and lower impervious fraction in cold zones, higher NDVI in arid zones, and greater green space in tropical zones. (3) Interaction analysis revealed pronounced climate-specific nonlinear coupling effects on SUHI, where daytime SUHI was primarily regulated by vegetation, moisture, water fraction interactions, whereas nighttime SUHI was more strongly controlled by combined effects of humidity, solar radiation, impervious surfaces, water fraction, and anthropogenic activity. These results highlighted the necessity of region-specific SUHI mitigation strategies and provided scientific support for resilient urban development.
Cui et al. (Mon,) studied this question.