Exploring Urban Heat Distribution via Intra- and Extra-Block Morphologies with Integrated Stacked Models

The spatial variability of land surface temperature (LST) is considerably affected by urban morphology. Previous research has focused separately on the thermal effects of urban morphology and the cooling effects of water bodies and urban parks. However, the combined influence of intra- and extra-block factors on LST has not been thoroughly examined. To bridge this research gap, we conducted an extensive analysis of 17 urban morphology factors in Hangzhou by employing a novel stacked ensemble approach. Results showed that the stacked ensemble models outperformed commonly used techniques, such as random forest and boosted regression trees. Extra-block factors, alongside building density, average building height, and vegetation coverage within blocks, predominantly influenced the LST distribution across all seasons. Building density was positively correlated with LST, with a maximum influence of 1.5 °C in spring, whereas building height was negatively correlated with it, with a maximum influence of 1.8 °C in winter. The cooling distance of the Qiantang River extends up to 2500 m into the urban blocks and has a maximum effect of 2 °C in summer. These insights deepen our comprehension of the interplay between LST and intra- and extra-block urban morphologies, thus offering valuable guidance for urban planners and policymakers.

This creative work has been published under the Creative Commons Attribution 4.0 International (CC-BY 4.0) license which allows copying, and redistribution as well as adaptation of the original work provided appropriate credit is given to the original author and the conditions of the license are met.