The impacts of landscape patterns spatio-temporal changes on land surface temperature from a multi-scale perspective: A case study of the Yangtze River Delta

Unordered and speedy urbanization is the foremost cause of land surface temperature (LST) rise in an urban area. Understanding the effects of landscape changes on LST is crucial for the urban sustainable development. In this study, we retrieved the LSTs of 26 cities in the Yangtze River Delta Urban Agglomeration with the Landsat images during the summer time (from June to August) of 2000 and 2019. From a multi-scale perspective, i.e. grids of 10 km and 20 km, county and city level, the partial correlation analysis, geographically weighted correlation analysis and local bivariate Moran's I were conducted to explore the influence of the landscape pattern changes of the built-ups on LST change. Our results have shown that, the scale change impacts the relationships between the landscape metric changes of built-ups and the LST change. As the scale upscales, the correlation between different landscape metric changes of built-ups and the LST change continues to increase. Among them, the area-related metrics (percentage and largest patch index) have the most significant impact on LST change, showing a positive correlation. Moreover, there are obvious spatial autocorrelation and spatial spillover effects between the landscape metric changes of built-ups and the LST change. These findings are helpful for understanding regional ecology as well as land use/land cover planning to minimize the negative environmental impacts of urbanization.