The Relationship Between Landscape Pattern And Land Surface Temperature At Multi-scale

As global urbanization continues to accelerate,understanding relationships between urban landscape pattern and land surface temperature(LST)is of great value to mitigate the urban heat island(UHI)effect and create a better living environment.Existing studies have investigated the relationship between LST and landscape patterns based on land use land cover(LULC).However,multi-scale studies of the relationships between LST and landscape patterns are still relatively lacking.In this study,Changsha was selected as the study area.Landsat 8 data was used to obtain the surface distribution map of LST and green space /impervious surface,and OSM data,Google Map and other data were used to obtain the distribution map of urban functional areas,and multi-scale analysis,correlation analysis,regression analysis,spatial statistical analysis,and urban-rural gradient analysis were used to quantify the multiscale correlation between urban landscape and LST from macroscopic to microscopic,using green space / impervious surface and urban functional areas as the division of urban landscape,and the conclusions of the study are as follows.(1)The warming effect of impervious surface on LST and the cooling effect of green space on LST.The warming effect of commercial and industrial areas in inner city functional areas and the cooling effect of water bodies in ecological functional areas also deserve to be noticed because they have the best warming and cooling effect among all urban functional areas.(2)The landscape configuration of urban landscapes also plays an important role in influencing LST.Aggregated,large impervious surfaces significantly increase LST,while aggregated,large green spaces significantly mitigate the heat island effect.The landscape configuration of commercial land,residential land and Public management-services land is not significantly related to LST.While clustered,large-area industrial land have a greater impact on LST.Aggregated,large-area water bodies have a better cooling effect on LST than aggregated urban villages and green spaces.The relationship between urban landscape pattern and LST has a multi-scale effect.In the conversion process of urbanized land cover,the warming of urban villages into industrial land is greater than that of commercial land.When the scale is larger than 21×21 pixel,the warming rate of urban villages conversion to commercial area is larger than that of industrial land.The influence of impervious surface and green space on LST is small at small scales,and when the scale is larger than 9×9 pixel,the influence of impervious surface and green space on LST tends to be smooth,and the best scale for the correlation between landscape components and LST is a square area of 210m-270 m.The multi-scale effect of the relationship between landscape configuration and LST varies significantly among different urban landscapes.increasing the area of green space and decreasing the area of impervious surface at the 11×11pixel scale,increasing the aggregation of green space and decreasing the aggregation of impervious surface at the 30×30 pixel scale have significant effects on reducing LST.the aggregation of industrial areas above the 0.33km×0.33 km can significantly increase LST,0.33 km×0.33 km or more,large water bodies,0.63 km×0.63 km or more,large green space parks,1.47 km× 1.47 km and larger,and large urban villages can significantly reduce LST.(4)On the urban-rural gradient,the variation of green space/impervious surface landscape components is an important reason for the variation of LST along the urban-rural gradient.Residential land and urban villages components affect the spatial distribution characteristics of LST as a whole,while commercial land affect LST mainly in the 1-3km range.Aggregated,large,and complex-shaped green spaces significantly reduce LST,while aggregated,large,and complex-shaped impervious surfaces increase LST.complex-shaped,large-area commercial land and large residential land increase LST,and aggregated,large-area urban villages and green land and parks reduce LST.