Machine Learning-Based Spatial Modeling of Land Surface Temperature in Butantã: Insights from SHAP Interpretability

• Williane G. S. Pereira UFPA https://orcid.org/0009-0005-0611-2462
• Gabriel V. de Melo UFPA https://orcid.org/0009-0008-8429-7281
• Waldemiro J. A. G. Negreiros UFPA https://orcid.org/0009-0007-2694-6274
• Fernando A. R. Costa UFPA https://orcid.org/0000-0002-0226-7505
• Leonardo de O. Tamasauskas UFPA https://orcid.org/0009-0006-8049-8997
• Claudomiro de S. de S. Junior UFPA https://orcid.org/0000-0002-2735-1383
• Marcos C. R. Seruffo UFPA https://orcid.org/0000-0002-8106-0560

DOI: https://doi.org/10.5753/wcama.2026.21147

Resumo

Accelerated urbanization has expanded urban heat islands, necessitating the precise monitoring of Land Surface Temperature (LST). This study spatially predicts LST in Butantã (SP) for 2018 and 2021 using a high-resolution multiscale approach. To capture the relationships between urban morphology and LST, XGBoost and LightGBM models were compared, with XGBoost demonstrating superior predictive performance (R² of 0.890 and 0.848). Applying the SHapley Additive exPlanations (SHAP) method to the winning model ensured explainability, revealing the dominance of impervious surfaces in warming, the strong mitigating potential of water bodies and vegetation, and the heat retention induced by topography in valleys.

Referências

Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019). Optuna: A next-generation hyperparameter optimization framework. In The 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pages 2623–2631.

Arunab, K. and Mathew, A. (2024). Exploring spatial machine learning techniques for improving land surface temperature prediction. Kuwait Journal of Science, 51(3):100242.

Bai, X., Yu, Z., Wang, B., Zhang, Y., Zhou, S., Sha, X., Li, S., Yao, X., and Geng, X. (2024). Quantifying threshold and scale response of urban air and surface temperature to surrounding landscapes under extreme heat. Building and Environment, 247:111029.

Bueno, E. S. and Ximenes, D. S. S. (2011). A importância da infraestrutura verde no desenho ambiental: Estudo da área da cidade universitária e instituto butantã. Revista LABVERDE, (3):128–154.

Hoang, N.-D., Pham, P. A. H., Huynh, T. C., Cao, M.-T., and Bui, D.-T. (2025a). Geospatial urban heat mapping with interpretable machine learning and deep learning: A case study in hue city, vietnam. Earth Science Informatics, 18(1):64.

Hoang, N.-D., Tran, V.-D., and Huynh, T.-C. (2025b). From data to insights: Modeling urban land surface temperature using geospatial analysis and interpretable machine learning. Sensors, 25(4).

Kim, M., Kim, D., and Kim, G. (2022). Examining the relationship between land use/land cover (lulc) and land surface temperature (lst) using explainable artificial intelligence (xai) models: A case study of seoul, south korea. International Journal of Environmental Research and Public Health, 19(23).

Landis, J. R. and Koch, G. G. (1977). The measurement of observer agreement for categorical data. biometrics, pages 159–174.

Li, S., Wong, M. S., Zhu, R., Shi, G., and Yang, J. (2025). Impacts of land surface temperature and ambient factors on near-surface air temperature estimation: A multisource evaluation using shap analysis. Sustainable Cities and Society, 122:106257.

Lundberg, S. M., Erion, G. G., and Lee, S.-I. (2018). Consistent individualized feature attribution for tree ensembles. arXiv preprint arXiv:1802.03888.

Ma, X. and Peng, S. (2022). Research on the spatiotemporal coupling relationships between land use/land cover compositions or patterns and the surface urban heat island effect. Environmental Science and Pollution Research, 29(26):39723–39742.

Raufu, I. O. and Adediran, A. (2025). Influence of topographic elements on land surface temperature: A case study using remote sensing and gis. Journal of Geography, Environment and Earth Science International, 29(12):249–264.

Shui, C., Shan, B., Li, W., Wang, L., and Liu, Y. (2025). Investigating the influence of land cover on land surface temperature. Advances in Space Research, 75(3):2614–2631.

Snaiki, R. and Merabtine, A. (2025). Recent advances on machine learning techniques for urban heat island applications: a review and new horizons. Sustainable Cities and Society, page 106943.

Sobrino, J. A., Jiménez-Muñoz, J. C., and Paolini, L. (2004). Land surface temperature retrieval from landsat tm 5. Remote Sensing of environment, 90(4):434–440.

Tamasauskas, L. d. O., Pereira, W. G., Negreiros, W. J., Guimaraes, P. H. d. V., Dias, J. A., Corrêa, A. B., Costa, G. B., and Seruffo, M. C. d. R. (2025). Comparison of lstm and sarima models for air temperature forecasting in belém, amazônia, pará. In Workshop de Computação Aplicada à Gestão do Meio Ambiente e Recursos Naturais (WCAMA), pages 256–265. SBC.

Tanoori, G., Soltani, A., and Modiri, A. (2024). Machine learning for urban heat island (uhi) analysis: Predicting land surface temperature (lst) in urban environments. Urban Climate, 55:101962.

Yin, H. and Zhao, X. (2024). Urban heat island analysis based on high resolution measurement data: A case study in beijing. Sustainable Cities and Society, 106:105389.