Machine Learning-Driven Hourly Heat Index Mapping via Multi-Source Data Fusion: Optimizing Blue-Green Infrastructure and Ventilation Corridors for Thermal Resilience in Wuhan

Wuhan, a metropolitan area in central China, faces escalating thermal health risks due to its unique climatic and geographical conditions. The city's extreme heatwaves, high humidity (dew point >25°C), and extensive water networks (26% urban coverage) amplify heat stress through dual mechanisms: (1) Urban heat islands elevate temperatures 2-5°C via anthropogenic heat and restricted ventilation, while (2) water bodies intensify humidity through daytime evaporation and nocturnal heat retention. This diurnal hydro-thermal coupling sustains critical wet-bulb temperatures, endangering 13.77 million vulnerable residents. To address this challenge, our study pioneers a high-resolution, hourly simulation of the Heat Index (HI) across Wuhan by integrating multi-source geospatial datasets. Leveraging dense meteorological observations (183 stations), daily 30m-resolution land surface temperature (LST) from satellite remote sensing, and 3D urban canopy parameters (building morphology and tree canopy coverage), we developed a robust random forest model to predict spatiotemporal thermal stress patterns. Our analysis reveals intricate interactions between heterogeneous urban landscapes and microclimatic variability, emphasizing the dual role of water bodies as both heat sinks and humidity sources. Furthermore, we quantify the cooling efficiency of blue-green infrastructure configurations and urban ventilation corridors through scenario-based simulations. Results demonstrate that strategically optimized green spaces can reduce peak-hour heat stress by up to 2–4°C, while river-aligned wind pathways enhance nighttime cooling rates by 15–20%. These findings provide actionable insights for climate-resilient urban planning, enabling targeted mitigation strategies such as heat-vulnerability zoning, green network optimization, and ventilation corridor preservation. This research framework advances fine-scale thermal environment modeling and supports evidence-based policymaking for sustainable megacity development in humid subtropical regions.