Construction of a nested cold island network based on multiple urban scales: A study in Nanchang, China

Mitigating heat islands is a critical issue for many cities to improve urban environmental quality. Variations in topography, urban development patterns, and thermal environmental characteristics across different spatial scales lead to scale-dependent heat islands. The mitigation effects of various factors on urban thermal environments are also dependent on scales. However, existing studies on the construction of cold island networks have not adequately addressed the factor differences with scales. Therefore, this study used the random forest algorithm to explore the impact of various influencing factors on the urban thermal environment of Nanchang, China, across different spatial scales. Additionally, a multi-scale collaborative nested cold island network was constructed using the Morphological Spatial Pattern Analysis (MSPA). The results verified that the dependence of influencing factors with urban scale, especially regarding elevation and building index. In the nested analysis of the multi-scale cold island network, 13 cooling sources were identified, covering a total area of 40.707 km², of which 9 sources exceed 1 km². Moreover, 15 segments of nested cooling corridors were identified, totaling 21.275 km. The nested cooling sources and corridors are capable of exerting cold island effects across different scales and should be prioritized for protection. Overall, the construction of the nested cold island network promotes the understanding in developing cooling strategies from a multi-scale perspective and valuable insights into heat island mitigation.