Context: Although many prior efforts have found that both spatial composition and configuration of greenspaces significantly affect the urban heat environment globally, the spatially heterogeneous effects of greenspace spatial patterns on the urban heat environment remain poorly understood for urban spaces. Objectives: We proposed a spatially explicit approach to investigate the spatially heterogeneous cooling effects of greenspaces and map the relative contributions of the greenspace spatial patterns to the characterization of the urban heat environment. Methods: The proposed approach integrated the best subsets regression method, geographically weighted regression (GWR), and hierarchical partitioning analysis. Two cities in southeastern China were selected to test our model. Landsat 5 image obtained in the summer was used to estimate the land surface temperature (LST) and greenspace spatial patterns were extracted from 0.5-m aerial images. Results: The results revealed that LST of Guangzhou can be well predicted by the percent cover (PER), the number of patches (NP), the area-weighted mean of the patch area (AREA_AM), and the area-weighted mean of the perimeter-area fractal dimension (FRAC_MN), while that of Shenzhen can be predicted by PER, NP, AREA_AM and the mean of the related circumscribing circle (CIRCLE_MN). The inclusion of additional landscape metrics did not yield significantly higher accuracies. The dominant landscape metrics of greenspace that determine the LST varied spatially across the two cities, with the PER accounting for the greatest variation. Conclusion: The results of our work demonstrate that the location of greenspace is a significant factor affecting the urban heat environment. The proposed approach provides a new understanding of the interaction between the greenspace spatial patterns and urban heat environments, providing useful information for tailoring greenspace planning policies for specific local sites.
Location of greenspace matters: a new approach to investigating the effect of the greenspace spatial pattern on urban heat environment
Zheng Z.
2021
Abstract
Context: Although many prior efforts have found that both spatial composition and configuration of greenspaces significantly affect the urban heat environment globally, the spatially heterogeneous effects of greenspace spatial patterns on the urban heat environment remain poorly understood for urban spaces. Objectives: We proposed a spatially explicit approach to investigate the spatially heterogeneous cooling effects of greenspaces and map the relative contributions of the greenspace spatial patterns to the characterization of the urban heat environment. Methods: The proposed approach integrated the best subsets regression method, geographically weighted regression (GWR), and hierarchical partitioning analysis. Two cities in southeastern China were selected to test our model. Landsat 5 image obtained in the summer was used to estimate the land surface temperature (LST) and greenspace spatial patterns were extracted from 0.5-m aerial images. Results: The results revealed that LST of Guangzhou can be well predicted by the percent cover (PER), the number of patches (NP), the area-weighted mean of the patch area (AREA_AM), and the area-weighted mean of the perimeter-area fractal dimension (FRAC_MN), while that of Shenzhen can be predicted by PER, NP, AREA_AM and the mean of the related circumscribing circle (CIRCLE_MN). The inclusion of additional landscape metrics did not yield significantly higher accuracies. The dominant landscape metrics of greenspace that determine the LST varied spatially across the two cities, with the PER accounting for the greatest variation. Conclusion: The results of our work demonstrate that the location of greenspace is a significant factor affecting the urban heat environment. The proposed approach provides a new understanding of the interaction between the greenspace spatial patterns and urban heat environments, providing useful information for tailoring greenspace planning policies for specific local sites.
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
