Mitigating Urban Heat Islands Through Critical Edge Deletion in Thermal Environment Networks: A Case study of the Central Guizhou Urban Agglomeration in Southwest China

The urban heat island (UHI) effect increasingly threatens ecosystem function and human well-being in rapidly urbanizing regions. However, existing studies predominantly describe the static morphology and land-use correlates of heat island patches, lacking a network-based understanding of how critical heat flow pathways can be identified and strategically disrupted for thermal remediation. To address this shortcoming, by taking the central Guizhou urban agglomeration in Southwest China as the study area. We integrated morphological spatial pattern analysis (MSPA), circuit theory, and complex network theory to construct a thermal environment network, quantitatively evaluated source importance via node weight, betweenness, and PageRank, and applied three edge-deleting strategies—random deletion (RD), high-degree-first (HDF), and high-betweenness-first (HBF)—to sever critical thermal corridors. The remediation effect was verified through connectivity robustness and recovery robustness analysis. The results indicated the total area of heat island patches expanded from 7484.63 km² in 2010 to 11 096.94 km² in 2022, with core-type patches becoming dominant. Both the number of sources and corridors in the thermal environment network and the complexity of the structure increased. Compared with the corridor remediation effect, the HDF strategy performed better in the robustness analysis, with the lowest original degree of connection robustness (decreasing from 0.82 to 0.73) and the fastest rate of decrease. Moreover, the node restore robustness reduction effect was the best, and the edge restore robustness reduction effect was comparable to that of the HBF strategy. The proposed framework provides concrete spatial guidance for mitigating heat island risks and promoting sustainable urban planning in karst urban agglomerations.