Spatiotemporal Relationship Between Human Activities and Urban Heat in Chinese Megacities Based on Multisource Remote Sensing

Human activities have significantly reshaped urban thermal environments and intensified heat-related risks in rapidly urbanizing areas. Owing to its broad coverage and long-term monitoring capabilities, satellite remote sensing has become essential for tracking land surface temperature (LST) dynamics. In this study, MODIS LST data and WorldPop population density (PD) data retrieved from thermal infrared and multi-source remote sensing were used. Employing correlation analysis, standard deviational ellipse, and a coordination model, we comprehensively assessed the spatiotemporal relationship between human activities and urban heat in six Chinese megacities from 2005 to 2020. The main results revealed that: (1) although both PD and LST exhibited consistent upward trends in all cities during 2005–2020, the correlation between the two decreased while remaining significantly positive, suggesting that governance measures in megacities have helped improve urban thermal environments; (2) the correlation displayed periodic fluctuations. Dominant coordination types alternated between “heating advance” and “coordinated enhancement,” with “heating hysteresis” tending to occur alongside “coordinated enhancement,” reflecting dynamic policy adaptations to different stages of urban development; (3) significant intercity variation in the PD–LST relationship was attributed to differences in geographical conditions, development models, and industrial structures. For example, Guangzhou exhibited high PD and LST but the lowest PD–LST correlation (0.31) and the fastest decline (−1.12%/a), whereas Beijing exhibited lower PD and LST but the highest correlation (0.50) and lowest change rate (−0.60%/a). Based on these findings, we propose targeted heat environment regulation strategies tailored to each city's characteristics, providing practical guidance for enhancing heat environment governance.