Speaker
Description
Open spaces in cities, serving as vital carriers of blue-green infrastructure and urban wind corridors, are crucial for mitigating urban heat island effect (UHI). This study aims to further investigate the complex relationship between open space morphology and UHI, propose corresponding optimization strategies, and validate them through simulation.
Taking Xishan District in Wuxi City, Jiangsu Province as the study area, the study classifies open spaces into seven categories by using Morphological Spatial Pattern Analysis (MSPA) method and Geographic Information System (GIS) technology. The study uses Python and machine learning algorithms to explore the linear and non-linear correlations between various open space patterns and UHI, elucidating their roles in urban microclimate regulation. The study selects 9 typical area samples in the district and optimizes their open space patterns. This includes increasing green space areas, adjusting the morphology of open spaces and increase the connectivity of green space and water bodies. The urban climate environment before and after optimization is simulated by CFD tools. This simulation verifies the effectiveness of the optimization measures in UHI mitigation.
The results indicate that increasing Cores and Bridges in open spaces positively impacts the mitigation of UHI, while Islets in open spaces exacerbate it. The findings are confirmed through simulations in the four sample areas. Increasing the area of open space, reducing point-like open space and enhancing continuity of green belts can effectively reduce urban surface temperature and improve urban microclimate conditions.
In conclusion, this study explored the relationship between urban open space morphology and UHI through spatial data analysis and simulations. This can provide empirical evidence and practical solutions for urban planning and design, contributing significantly to enhancing urban resilience and addressing climate change challenges.
| Keywords | Urban open space, heat island effect (UHI), machine learning |
|---|---|
| Best Congress Paper Award | Yes |
