Quantifying the contributions of urban spatial morphology on the river cold island effect: Taking Changsha, China, as an example
Huiwen Wang, Cunyou Chen
Abstract
Urban rivers serve a vital role in mitigating urban heat islands, a phenomenon that is becoming increasingly apparent against the backdrop of rapid urban expansion and global warming. We combined with the cold island amplitude and the maximum cold island distance, eleven indicators of three types of indexes were selected, and the cold island effect of eight rivers in Changsha urban area was analyzed by Boosted Regression Tree model. Results show that (1) As the width of the river increases, the cold island effect increases. The cold island amplitude of the river with a width of 20.15–1717.35 m ranges from 2.23 to 8.93°C, and maximum cold island distance ranges from 187.5 to 16,666.67 m. (2) The order of contribution is land cover index > building space index > landscape pattern index, among which the contribution of normalized building index (NDBI), normalized vegetation index (NDVI) and Mean architectural height (MAH) is the largest. (3) According to the marginal effect curve, Mean of patch area (AREA_MN) and Shannon diversity (SHDI) at 0.36–0.4 and 0.6–1.5, were weakly positively and negatively correlated with surface temperature. NDBI has a significant positive correlation with the surface temperature, and NDVI has a negative correlation with the surface temperature. MAH and tolerance for staggering buildings (TSB) have a cooling effect on the river after >45 m and 34.2 m, respectively. The research results provide valuable insights into urban space management and planning regulation.