Urban agglomerations environmental heterogeneity and heatwave risks: spatiotemporal insights from remote sensing and public sentiment analysis

Urban agglomerations are global hotspots for escalating heatwave risks, where the complex interplay between environmental drivers and public emotional responses remains underexplored due to limitations in traditional research methods and data on heatwave impacts to public health. Pioneering a novel framework, this study integrates multi-source remote sensing data (2014–2023), 3.2 million social media posts, and advanced causal models across five major Chinese urban agglomerations. Our analysis reveals that environmental factors are the dominant drivers of heatwaves, with surface properties—notably albedo, vegetation cover (FVC), and thermal indices (NDTI)—exerting a far stronger influence than hydrological or topographical features. Concurrently, we uncover distinct, nonlinear emotional responses to escalating heat. While heat perception rises directly with intensity, health concerns exhibit a critical threshold effect, surging only during moderate-to-high events and preceded by behavioral shifts. Crucially, each urban agglomeration displays a unique socio-ecological response signature, highlighting significant regional heterogeneity. These findings provide critical evidence for developing tailored, place-based policies to mitigate thermal risks and enhance urban resilience. Unveils asymmetric causalities between urban environment and heatwaves using causal inference.Integrates 3.2 million social media posts to map public sentiment response to heatwaves.Quantifies multilevel effects with Bayesian GLMMs across five Chinese urban agglomerations.Reveals nonlinear, region-specific emotional responses to heatwave indicators.Provides actionable insights for climate adaptation in rapidly urbanizing regions. Unveils asymmetric causalities between urban environment and heatwaves using causal inference. Integrates 3.2 million social media posts to map public sentiment response to heatwaves. Quantifies multilevel effects with Bayesian GLMMs across five Chinese urban agglomerations. Reveals nonlinear, region-specific emotional responses to heatwave indicators. Provides actionable insights for climate adaptation in rapidly urbanizing regions.