Effects of changing spatial extent on the relationship between urban forest patterns and land surface temperature

Abstract

Abstract The effects of spatial pattern of urban greenspace on land surface temperature (LST) have been extensively documented. Previous studies have confirmed that the relationship between urban green spatial pattern and LST is sensitive to spatial resolution and remote sensing imagery with higher resolution could be more accurate on quantifying the urban green patterns, but little is known about another scaling issue—spatial extent. This paper examined whether the spatial extent applied to derive landscape metrics affect the relationship between LST and spatial pattern of urban forested areas of highly urbanized Shanghai and the seasonal variations using correlation analyses and regression analyses. Spatial pattern of forested areas was measured with eight class-level landscape metrics over four spatial extents/scales (90 m × 90 m, 180 m × 180 m, 360 m × 360 m and 720 m × 720 m) using moving-window approach based on a land-use and land cover (LULC) map derived from SPOT 6 datasets. Results demonstrated that changing spatial extent had significant impacts on the relationship between spatial pattern of urban forested areas and LST. The responses of correlations between spatial pattern metrics and LST to changing extent fell into three categories: correlation decreases with extent increases, correlation increases with extent increases and unpredictable pattern. In general, the amount of forested cover accounts for greater variability in LST than its spatial arrangement. This study extended our scientific understanding of the effects of spatial pattern of urban forested area on LST. In addition, it can provide insights for urban forest planning and management.