The influence of the landscape pattern on the urban land surface temperature varies with the ratio of land components: Insights from 2D/3D building/vegetation metrics

Recently, the impact of multidimensional building/vegetation landscape patterns on urban land surface temperature (LST) has received extensive attention. However, investigations on how two-dimensional (2D)/three-dimensional (3D) building and vegetation landscape patterns affect the urban thermal environment under different cover types are still limited. Thus, we use the boosted regression tree model to explore the separate and combined relative contributions and marginal effects of 2D/3D building and vegetation landscape patterns on the LST and to examine the impacts of differences in building and vegetation coverage. The results indicate that the mean architecture height, building coverage ratio, high building ratio, vegetation coverage ratio, and architecture height standard deviation are the main metrics affecting the LST, with relative contributions of 24.8%, 14.9%, 14.7%, 8.2%, and 7.6%, respectively. The building landscape roughness and fragmentation have the greatest influence on the LST, with contributions of 1.2 °C and 0.8 °C, respectively. Furthermore, changes in the building and vegetation coverage can significantly affect the LST. When building coverage dominates, building-based metrics largely affect the LST. When vegetation coverage dominates, the impact of vegetation landscape diversity, roughness, and fragmentation on the LST gradually increases. This study provides insights for improving urban thermal environment.