Xucai Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xucai Zhang has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Haosheng Huang
Ghent University
4 shared publications Xiaoping Liu
Guangdong Key Laboratory for Urbanization and Geo-simulation, School of Geography and Planning, Sun Yat-sen University
4 shared publications- Hongkai Gu
Department of Geography, Ghent University
3 shared publications Xiaobing Wei
Department of Geography, Ghent University
3 shared publicationsYue Zheng
Guangdong Key Laboratory for Urbanization and Geo-simulation, School of Geography and Planning, Sun Yat-sen University
3 shared publicationsYangzi Che
Guangdong Key Laboratory for Urbanization and Geo-simulation, School of Geography and Planning, Sun Yat-sen University
2 shared publications
Latest Publications
- Open Access
- 19 October 2026
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 123 (43)
Toward cooler cities by larger homogeneous functional clusters
Urban heat island (UHI) intensifies heat-related risks and environmental degradation in cities worldwide, yet existing mitigation strategies have largely overlooked how the spatial scale of homogeneous functional building clusters influences local climate. This scale may matter because buildings with similar functions often share similar morphology, activity schedules, and anthropogenic heat-release patterns, allowing their spatial contiguity to shape ventilation and heat accumulation. Here, we introduce the Urban Homogeneous Scale (UHS)—an indicator that measures the spatial scale of contiguous, functionally similar building clusters—and analyze 233,844 observations across 101 global cities to assess its potential for mitigating surface UHI. We find that larger UHS is associated with lower surface UHI: every 100-meter increase in UHS corresponds to 4.5% decrease in daytime UHI. This daytime cooling effect is particularly more pronounced in inland, humid, and high-income regions. In low-income regions, larger building area and higher building volume more strongly weaken UHS cooling capacity than in wealthier regions, revealing a socioeconomic inequality in UHS’s mitigation benefits. Our findings identify the spatial scale of building clusters as a essential dimension of urban form associated with urban heat burden, and suggest that considering UHS alongside targeted controls on built environmental characteristics may inform future heat-mitigation planning.
- 19 October 2026
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 123 (43)
Open AccessToward cooler cities by larger homogeneous functional clusters
Urban heat island (UHI) intensifies heat-related risks and environmental degradation in cities worldwide, yet existing mitigation strategies have largely overlooked how the spatial scale of homogeneous functional building clusters influences local climate. This scale may matter because buildings with similar functions often share similar morphology, activity schedules, and anthropogenic heat-release patterns, allowing their spatial contiguity to shape ventilation and heat accumulation. Here, we introduce the Urban Homogeneous Scale (UHS)—an indicator that measures the spatial scale of contiguous, functionally similar building clusters—and analyze 233,844 observations across 101 global cities to assess its potential for mitigating surface UHI. We find that larger UHS is associated with lower surface UHI: every 100-meter increase in UHS corresponds to 4.5% decrease in daytime UHI. This daytime cooling effect is particularly more pronounced in inland, humid, and high-income regions. In low-income regions, larger building area and higher building volume more strongly weaken UHS cooling capacity than in wealthier regions, revealing a socioeconomic inequality in UHS’s mitigation benefits. Our findings identify the spatial scale of building clusters as a essential dimension of urban form associated with urban heat burden, and suggest that considering UHS alongside targeted controls on built environmental characteristics may inform future heat-mitigation planning.
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 64 (01962892) , 1-10
Enhancing Geo-Transferability in Building Function Classification Through Similar Place Knowledge Retrieval
Understanding the use of buildings (i.e., building functions) is crucial for effective urban management and policy making. Recently, many deep learning-based models have been developed to identify building functions. However, their high performance depends on a large labeled dataset, which is a significant challenge for cities with limited labeled data. To close this gap, we propose a similar-place knowledge-retrieval-based framework for building function identification that enables a trained model to be directly applied/transferred to other cities without additional training needed. Our approach combines remote sensing imagery, POI data, and building footprints to capture the macro and micro built environment. Specifically, this framework includes two different modules: 1) a geo-aware transformer-based model to capture fine-grained spatial interactions between buildings; 2) a hierarchical node clustering algorithm with a similar-place knowledge-retrieval layer for portraying the coarse-grained features. This combination clusters all neighbors of a target building into a ’place’ and then conducts a coarse-grained knowledge transfer based on the similarity match. The evaluation conducted across nine cities demonstrates that the proposed model outperforms all state-of-the-art baselines, achieving an average improvement of 21.98% in cross-city predictions, as measured by Cohen’s kappa coefficient. More importantly, the ablation study shows that integrating our knowledge-retrieval layer into other deep-learning baselines significantly improves their cross-city prediction performance. Additionally, the transferability performance analysis indicates that greater similarity between the sample distributions in the source and target cities contributes to the proposed model’s higher transferability. In summary, the proposed knowledge-retrieval layer is an effective and reliable component for enhancing cross-city prediction and has strong potentials for application in cities without labeled data to support city management and sustainable urban planning.
- 28 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 64 (01962892) , 1-10
Enhancing Geo-Transferability in Building Function Classification Through Similar Place Knowledge Retrieval
Understanding the use of buildings (i.e., building functions) is crucial for effective urban management and policy making. Recently, many deep learning-based models have been developed to identify building functions. However, their high performance depends on a large labeled dataset, which is a significant challenge for cities with limited labeled data. To close this gap, we propose a similar-place knowledge-retrieval-based framework for building function identification that enables a trained model to be directly applied/transferred to other cities without additional training needed. Our approach combines remote sensing imagery, POI data, and building footprints to capture the macro and micro built environment. Specifically, this framework includes two different modules: 1) a geo-aware transformer-based model to capture fine-grained spatial interactions between buildings; 2) a hierarchical node clustering algorithm with a similar-place knowledge-retrieval layer for portraying the coarse-grained features. This combination clusters all neighbors of a target building into a ’place’ and then conducts a coarse-grained knowledge transfer based on the similarity match. The evaluation conducted across nine cities demonstrates that the proposed model outperforms all state-of-the-art baselines, achieving an average improvement of 21.98% in cross-city predictions, as measured by Cohen’s kappa coefficient. More importantly, the ablation study shows that integrating our knowledge-retrieval layer into other deep-learning baselines significantly improves their cross-city prediction performance. Additionally, the transferability performance analysis indicates that greater similarity between the sample distributions in the source and target cities contributes to the proposed model’s higher transferability. In summary, the proposed knowledge-retrieval layer is an effective and reliable component for enhancing cross-city prediction and has strong potentials for application in cities without labeled data to support city management and sustainable urban planning.
- Top Cited Papers
- 6 May 2024
- Published by Elsevier in Sustainable Cities and Society
- Vol. 108
Identifying building function using multisource data: A case study of China's three major urban agglomerations
- 6 May 2024
- Published by Elsevier in Sustainable Cities and Society
- Vol. 108
Top Cited PapersIdentifying building function using multisource data: A case study of China's three major urban agglomerations
- 27 February 2024
- Published by Elsevier in Sustainable Cities and Society
- Vol. 105
Characterizing the 3-D structure of each building in the conterminous United States
- 27 February 2024
- Published by Elsevier in Sustainable Cities and Society
- Vol. 105
Characterizing the 3-D structure of each building in the conterminous United States
- 10 December 2022
- Published by Elsevier in Sustainable Cities and Society
- Vol. 89
Inferring building function: A novel geo-aware neural network supporting building-level function classification
- 10 December 2022
- Published by Elsevier in Sustainable Cities and Society
- Vol. 89
Inferring building function: A novel geo-aware neural network supporting building-level function classification