Influence of underlying surface change caused by urban renewal on land surface temperatures in Central Guangzhou

Land use change during urban renewal can have important impacts on ecological and environment indices. Exploring the impact of different urban renewal modes on land surface temperature (LST) is significant to city planning and the realization of urban sustainability. In this study, we selected five urban renewal regions of Central Guangzhou that are undergoing comprehensive or micro transformation. Deeplab V3+ was used to extract land use data from high-resolution remote sensing images; combined with LST data retrieved from Landsat series images, Pearson correlation analysis and geographically and temporally weighted regression (GTWR) analysis were used to explore the spatiotemporal heterogeneity of LST affected by underlying surface changes that were caused by urban renewal. The results indicate that urban renewal in all regions generated positive effects on LST reduction. However, comprehensive transformations in the diversity of land use types and more landscape pattern changes had higher potential for LST cooling effect than micro transformations. Owing to the heating effect of concrete, the LST peak point usually occurred in the block demolition phase of urban renewal. The transformation from dense low-rise to scattered high-rise buildings achieved LST cooling effect through shadows and ventilation to heat dissipation. In addition, parks with large areas of trees had stronger cooling capacity than other vegetation or farmland patches, and scattered vegetation patches played the role of connecting the influence of vegetation. Overall, appropriate transformation of architecture and green space landscape patterns are crucial to achieve LST cooling effect during urban renewal.