Multisource data-driven assessment of joint heatwave-flood risks: Spatial heterogeneity and key determinants

Urban heatwaves and floods increasingly threaten urban sustainability, yet most studies assess these hazards independently, limiting understanding of their combined impacts. This study quantifies the joint risk of UHR and UFR at the township level in Fuzhou, China, using a multisource data framework integrating remote sensing, meteorological, geographic, and socioeconomic datasets. The optimal parameter-based geographical detector identifies the contributions and interactions of multiple factors, while spatial autocorrelation and Getis-Ord Gi* hotspot analyses detect high-risk clusters. Results show the joint risk index ranges from 0.011 to 0.658, with high-risk (0.302–0.658) and medium-risk (0.099–0.302) areas, which account for 1.9 % and 12 % of the study area, respectively, concentrated in northern-central townships, and low-risk areas (0.011–0.099) in southeastern townships. Nightlight intensity (q = 0.8955), population density (q = 0.8928), impervious surface proportion (q = 0.8514), and building density (q = 0.8506) are the dominant factors, with interactions explaining up to q = 0.9641 (road density and proportion of the population aged 0–15 or > 65), exceeding individual factor contributions. Based on these results, response measures were proposed according to the characteristics of each township. Our research provides new insights for climate adaptation planning and the healthy development of cities. • Joint heatwave-flood risks were comprehensively assessed at the township scale. • An assessment framework integrated hazard, exposure, and vulnerability dimensions. • Game-theory weighting balanced expert judgment with data-driven objectivity. • OPGD revealed significant synergistic enhancement effects among all factors. • Findings highlight spatially uneven joint risks guiding climate-resilient planning.