Quantitatively comparing the morphological influences on the cool island effect in urban waterfront blue-green spaces across six cities near 30°N

Abstract

Urban blue-green spaces (BGS) are a proven strategy to mitigate urban heat islands, reduce energy consumption, and improve public health. This study examines the cool island effect of waterfront BGS in six Chinese cities near 30°N. A three-tiered road network partitions the urban waterfront into basic analysis units, while a multi-dimensional indicator system is constructed to analyse key morphological indicators and spatial differentiation characteristics of the cool island effect associated with BGS using multiple machine learning models. The results show that (1) the BGS in central urban areas exhibit superior cooling effects, with an average Land Surface Temperature (LST) reduction exceeding 2 °C; (2) the ranking of waterfront area LST aligns with thermal patterns observed in central urban areas, with a mean temperature increase of 1.26 °C compared to central urban areas; (3) two machine learning models, Random Forest (RF) model and Extremely Randomized Trees (ERT) model, demonstrate high robustness and applicability in exploring the quantitative relationship between cool island effect and waterfront BGS; and (4) the key morphological factors of the cool island effect of the BGS among the six cities near 30°N are highly consistent. The findings provide valuable insights for planning waterfront BGS and mitigating urban heat islands.