Scholarlink

<p>Urban heat in high-density cities is strongly shaped by complex urban morphology, yet existing studies based on Local Climate Zones (LCZs) often overlook microthermal variability within and between zones. This study proposes a spatial modeling framework integrating LCZ and Multiple Local Climate Zones (MLCZ) to quantify the microthermal effects of urban morphology at a fine scale (50 × 50&#xa0;m) in Hong Kong. A multi-source dataset combining remote sensing (LST from Landsat 8), built environment data (OpenStreetMap), and street view imagery was employed. Six morphological indicators representing both two-dimensional (building density, compactness, neighbor number, building spacing) and three-dimensional structures (urban canyon, street enclosure) were derived. A modeling approach integrating Pearson correlation and Ordinary Least Squares (OLS) regression was developed to assess their spatiotemporal impacts on land surface temperature (LST), while SARIMA was applied to reconstruct missing thermal data. Results reveal that (1) the conventional LCZ framework fails to fully capture intra-zone morphological heterogeneity, especially for 3D structures; (2) MLCZ significantly improves the explanation of microthermal variability in transition zones; and (3) building density, compactness, and urban canyon exhibit the strongest positive contributions to LST, with standardized coefficients up to 0.075. In contrast, street enclosure shows a cooling effect. This study provides an explicit modeling framework linking urban morphology to thermal environments, offering quantitative evidence for urban design strategies to mitigate heat risks in high-density cities.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A spatial modeling framework of microthermal effects driven by urban morphology across Local Climate Zones

  • 期刊论文

  • 发表日期：

    2026年5月19日

查看全文
  • Xinyue Gu

摘要

Urban heat in high-density cities is strongly shaped by complex urban morphology, yet existing studies based on Local Climate Zones (LCZs) often overlook microthermal variability within and between zones. This study proposes a spatial modeling framework integrating LCZ and Multiple Local Climate Zones (MLCZ) to quantify the microthermal effects of urban morphology at a fine scale (50 × 50 m) in Hong Kong. A multi-source dataset combining remote sensing (LST from Landsat 8), built environment data (OpenStreetMap), and street view imagery was employed. Six morphological indicators representing both two-dimensional (building density, compactness, neighbor number, building spacing) and three-dimensional structures (urban canyon, street enclosure) were derived. A modeling approach integrating Pearson correlation and Ordinary Least Squares (OLS) regression was developed to assess their spatiotemporal impacts on land surface temperature (LST), while SARIMA was applied to reconstruct missing thermal data. Results reveal that (1) the conventional LCZ framework fails to fully capture intra-zone morphological heterogeneity, especially for 3D structures; (2) MLCZ significantly improves the explanation of microthermal variability in transition zones; and (3) building density, compactness, and urban canyon exhibit the strongest positive contributions to LST, with standardized coefficients up to 0.075. In contrast, street enclosure shows a cooling effect. This study provides an explicit modeling framework linking urban morphology to thermal environments, offering quantitative evidence for urban design strategies to mitigate heat risks in high-density cities.