How do 2D/3D urban landscapes impact diurnal land surface temperature: Insights from block scale and machine learning algorithms

文献类型： 外文期刊

第一作者： Han, Dongrui

作者： Han, Dongrui;Wang, Fei;An, Hongmin;Cai, Hongyan;Xu, Xinliang;Qiao, Zhi;Jia, Kun;Sun, Zongyao;An, Ying;Qiao, Zhi

作者机构：

关键词： Urban landscapes; Diurnal land surface temperature; Block scale; Machine learning algorithms; Urban heat island

期刊名称：SUSTAINABLE CITIES AND SOCIETY （ 影响因子：11.7； 五年影响因子：10.6 ）

ISSN： 2210-6707

年卷期： 2023 年 99 卷

页码：

收录情况： SCI

摘要： Urban landscapes significantly affect land surface temperature (LST) and are considered crucial factors affecting urban heat island (UHI). The impacts of urban landscapes on LST have been extensively explored, mainly focusing on grid scale and the daytime. However, how 2D/3D urban landscapes affect diurnal LST at the block scale is unclear. Therefore, taking 1, 536 blocks (including low-rise blocks (LRB), middle-rise blocks (MRB), and high-rise blocks (HRB)) in Beijing as samples, the performances of boosted regression tree (BRT) and random forest (RF) were first evaluated, and the impacts of 2D/3D urban landscapes on diurnal LST across different block types were explored. The results showed that the mean LST was the highest in MRB (daytime) and HRB (nighttime). BRT performed better than RF in investigating diurnal impacts at the block scale. Vegetation and buildings are the domain factors influencing daytime and nighttime LST in LRB and MRB, while buildings are the domain factor in HRB except at 03:09 (impervious surface). The relationships between the key 2D/3D urban landscape metrics and block diurnal LST are nonlinear. The findings can serve as f basis for UHI mitigation and urban renewal strategies by urban planners to develop thermal comfort.

分类号：

  • 相关文献
作者其他论文 更多>>

微信小程序