Drone-based assessment of urban green space structure and cooling capacity

The urban green spaces (UGS) structure are important attributes affecting a wide range of ecosystem services in cities, especially the potential cooling effects of trees and vegetation. However, the coarse resolution of most satellite images limits the ability to detect and assess the structural characteristics of UGS and their effects on the thermal environment. In this direction, drones provide an ideal tool to overcome this issue. In this study, we employed Random Forest algorithms and Python tools modelling to produce detailed land use patterns for two typical UGS in China (i.e., urban parks) to explore the effect of UGS structural factors on the thermal environment and its cooling capacity. We used two UGS in Baoding as case studies: Jingxiu Park with diverse and complex spatial structures, and Military Academy Square characterized by single and simple structures. Our results suggest that Jingxiu Park (complex struct patterns) provides a higher cooling effect (0.551℃ lower) and temperature regulation than Military Academy Square (simple struct patterns). Additionally, land surface temperature comparisons of impervious surfaces with and without shaded areas demonstrate that shading has an enhanced cooling effect and varies across habitat types.