Optimizing　urban　patterns　is　increasingly　recognized　as　an　effective　technological　strategy　to　mitigate　the　urban　heat　island　(UHI)　effect.　Taking　Xiamen　City　as　a　case　study,　this　research　extracts　and　quantifies　city　spatial　characteristics　from　multiple　data　sources.　Key　factors　influencing　the　urban　thermal　environment　were　integrated　into　three　primary　urban　pattern　elements:　Mountain,　City,　and　Sea.　The　spatial　autocorrelation　and　heterogeneous　impacts　of　these　urban　pattern　elements　on　the　thermal　environment　were　analyzed　using　Moran’s　I　and　Geographically　Weighted　Regression　(GWR)　modeling,　followed　by　impact-based　zoning　using　K-means　clustering　algorithms.　The　results　revealed　a　significant　positive　correlation　between　Mountain　and　City　elements　and　the　thermal　environment,　whereas　Sea　elements　exhibited　a　notable　cooling　effect.　Furthermore,　each　factor　demonstrated　significant　spatial　heterogeneity.　Based　on　local　GWR　regression　coefficients　and　spatial　variations　in　factor　intensity　and　directionality,　Xiamen　was　partitioned　into　four　distinct　regulatory　zones:　City-dominated　zones,　Sea-dominated　zones,　Mountain–Sea　co-dominated　zones,　and　Comprehensive　transitional　zones　influenced　by　Mountain–City–Sea　interactions.　Customized,　spatially　targeted　regulatory　strategies　were　subsequently　proposed　for　each　zone.　This　study　provides　an　innovative　methodological　framework　for　targeted,　region-specific　policy　interventions　to　alleviate　urban　thermal　stress　under　climate　change,　thereby　contributing　to　the　optimization　of　future　urban　planning　and　promoting　sustainable　and　adaptive　development.　©　2025　by　the　authors.