388 / 2019-02-26 10:09:44

Investigating the mpacts of Three-Dimensional Urban Spatial Patterns on Urban Thermal Environment in Hangzhou, China

Land surface temperature; driving factors; spatial regression; random forest; cubist model; Hangzhou.

全体主题 > Urban Planning & Modeling

Understanding the relationship between different dimensional urban spatial pattern and land surface temperature (LST) is important for mitigating urban heat island. This study used Hangzhou as an example to explore the relationship between spatial variation of LST in summer and the influencing factors. Its objectives are as follows: (1) to build quantitative relationships between LST and three-dimensional (3-D) urban information at multiple resolutions; (2) to find suitable resolutions for explaining the relationships; (3) to find the most important factor affecting spatial heterogeneity of LST by using machine learning models. Results of this study indicate that the resolution of 600m are most suitable for measuring the relationships between 3-D urban spatial pattern and LST. At this resolution, the most important factor is the normalized difference vegetation index (NDVI) in summer. These two machine learning models performed well in LST fitting and could explain more than 85% variations of LST. These findings provided valuable insights into how thermal environment impacts of urbanization can be mitigated through local-level planning and zoning approaches.