Quantifying the aggravation and mitigation of urban heat island through differential dynamic changes in impervious surface
Xu T. Gong J. Yang Z. Wang Y. Jin T. Duman I. Kerimbaevich E.F.
1 September 2025Elsevier Ltd
Sustainable Cities and Society
2025#131
Impervious surfaces changes significantly impact urban heat island patterns during rapid urbanization. However, existing extraction models face challenges in capturing the temporal impervious surfaces dynamics and identifying reverse changes, mainly due to regional heterogeneity and temporal instability. Moreover, the impacts of these changes on surface urban heat island intensity (SUHII) trends and its patterns remain unclear. Therefore, this study developed a framework for estimating dynamic impervious surface percentage (DISP) by integrating machine learning and time optimization. The framework was used to quantify the SUHII trend under four types of impervious surface changes (increased, decreased, unchanged, and other). This classification can investigate the distinct thermal environmental responses associated with different DISP. Firstly, spatiotemporal consistency screening and extraction rules were designed to generate training samples. Subsequently, multi-temporal image fusion strategies were implemented, and regionally adaptive impervious surface percentage estimation models were constructed based on multiple feature combinations. Additionally, a time-series correction method was introduced to optimize DISP across different change trajectories. Finally, the dynamic trends of SUHII were captured under time series changes and regional differences, based on various impervious surface change types. The results show that DISP can achieve continuous and accurate estimations (R² = 0.848, RMSE = 0.106), outperforming existing products. Different types of impervious surface changes exhibit distinct thermal responses, with increased DISP intensifying the heat island effects and decreased DISP mitigating them. These effects are more pronounced in tropical and temperate zones than in arid regions. And high-density impervious surfaces also correspond to higher temperatures. Consequently, DISP provides robust and continuous reference for monitoring urban impervious surface dynamics and associated thermal environmental changes within cities.
Impervious surface , Machine learning , Multi-source remote sensing , Surface urban heat island , Time series
Text of the article Перейти на текст статьи
Key Laboratory of Western Chinas Environmental Systems (Ministry of Education), College of Earth and Environmental Sciences, Lanzhou University, Lanzhou, 730000, China
Center for Remote Sensing of Ecological Environments in Cold and Arid Regions, Lanzhou University, Lanzhou, 730000, China
Scientific and Educational Technology Platform, Kazakh National Agrarian Research University, Almaty, 050010, Kazakhstan
Key Laboratory of Western Chinas Environmental Systems (Ministry of Education)
Center for Remote Sensing of Ecological Environments in Cold and Arid Regions
Scientific and Educational Technology Platform
10 лет помогаем публиковать статьи Международный издатель
Книга Публикация научной статьи Волощук 2026 Book Publication of a scientific article 2026