Examining water bodies' cooling effect in urban parks with buffer analysis and random forest regression

Abstract

As urban thermal environments continue to deteriorate, the capacity of urban parks to mitigate the Urban Heat Island (UHI) effect has gained increasing attention, particularly the role of water bodies within these parks. However, the relationship between water body characteristics and cooling effect of urban park remains underexplored. This study utilizes Landsat 8 OLI_TIRS imagery from 28 urban parks in Jiangsu, China, applying buffer analysis and random forest regression to evaluate the cooling effects of water bodies based on their coverage and complexity. The key findings are as follows: (1) For every 0.1 increase in water body shape index in urban parks, the cooling area increases by 3.746 ha; (2) larger water body coverage enhances cooling, but the cumulative cooling efficiency decreases beyond a certain threshold; (3) from a cost-effectiveness perspective, there exists an optimal threshold value of capacity (TVoC) for water bodies in urban parks to alleviate the UHI, which is influenced by the coverage and complexity of urban parks; (4) for every 1 ha increase in urban park area raises the TVoC by 0.017 ha, and every 0.1 increase in urban park shape index raises the TVoC by 0.117 ha. Based on these findings, we recommend that urban planners, when renovating urban parks, adjust the size of water bodies to approach the optimal TVoC, enriching the shapes of water bodies, and integrating urban parks with blue-green infrastructure. These insights are crucial for enhancing the cooling effect of urban parks and promoting sustainable urban renewal.