Exploring the spatial heterogeneity of urban heat island effect and its relationship to block morphology with the geographically weighted regression model

An increasing number of studies in recent years have investigated the relationship between urban morphology and the urban heat island (UHI) effect in the context of global climate change and urbanization. However, most research does not consider the spatial heterogeneity of UHI effect and its relationship to urban morphology at the block level. In this study, we used 410 management units (MUs) of Xi'an, China, as the spatial scale and qualified the relationships between UHI effect and several influencing factors of block morphology. Geographically weighted regression (GWR) models were adopted combining multi-source data such as remote sensing images and building footprints. Compared to the ordinary least squares (OLS) models, the GWR models considerably improve modeling fit by capturing the spatial heterogeneity. The results show significant spatial variations of different variables. The impervious surface ratio (ISR) and building density (BD) are the top 2 urban morphology features intensifying the UHI effect, while green ratio (GR) is a critical factor forming a cool urban island in the dense urban areas. More importantly, floor area ratio (FAR) and sky view factor (SVF) show the strong non-stationary effect on the local UHI effect. These findings suggest that morphological variables significantly impact the UHI effect, and it is necessary to consider the spatial context. This study provides useful insights to understand the UHI effect as a function of urban morphology and substantial implications for sustainable urban planning, especially in high-density urban areas.