Subseasonal Precipitation Forecasting in Southeastern Brazil Using Machine Learning with Explainable Artificial Intelligence
Resumo
Subseasonal precipitation forecasting is essential for mitigating climate-related risks, but prediction remains challenging for traditional numerical weather prediction models. This study evaluates machine learning (ML) models—MLP, LSTM, GRU, and CNN-1D—for subseasonal precipitation forecasting in Southeastern Brazil. ERA5 reanalysis data (1991–2018) were used as the observational reference, with 1991–2016 for training and 2017–2018 for testing. The ML models were compared to the Brazilian Global Atmospheric Model (BAM), and the results show that all ML models outperform BAM, with MLP achieving the best overall performance. SHAP values were used to explain the MLP predictions, highlighting the importance of the input variables.Referências
Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019). Optuna: A nextgeneration hyperparameter optimization framework. CoRR, abs/1907.10902.
Anochi, J. A., de Almeida, V. A., and de Campos Velho, H. F. (2021). Machine learning for climate precipitation prediction modeling over south america. Remote Sensing.
Anochi, J. A. and Shimizu, M. H. (2025). Precipitation forecasting and drought monitoring in south america using a machine learning approach. Meteorology.
Chung, J., Gulcehre, C., Cho, K., and Bengio, Y. (2014). Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555.
Copernicus Climate Change Service (C3S) (2017). Era5: Fifth generation of ecmwf atmospheric reanalyses of the global climate.
Domingos, M., Júnior, V. S., Anochi, J., and Shiguemori, E. (2025). Exploring deep learning techniques for seasonal prediction of autumn precipitation in south america. In Anais do XVI Workshop de Computação Aplicada à Gestão do Meio Ambiente e Recursos Naturais, pages 157–166. SBC.
Figueroa, S. N., Bonatti, J. P., Kubota, P. Y., et al. (2016). The brazilian global atmospheric model (bam): Performance for tropical rainfall forecasting and sensitivity to convective scheme and horizontal resolution. Weather and Forecasting, 31(5):1547–1565.
Hersbach, H. et al. (2020). The era5 global reanalysis. Quarterly Journal of the Royal Meteorological Society.
Hochreiter, S. and Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8):1735–1780.
Lam, R., Sanchez-Gonzalez, A., Willson, M., Wirnsberger, P., Fortunato, M., Alet, F., Ravuri, S., Ewalds, T., Eaton-Rosen, Z., Hu, W., Merose, A., Hoyer, S., Holland, G., Vinyals, O., Stott, J., Pritzel, A., Mohamed, S., and Battaglia, P. (2023). Learning skillful medium-range global weather forecasting.
LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324.
Lundberg, S. M. and Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems, volume 30.
Monego, V. S., Anochi, J. A., and de Campos Velho, H. F. (2022). South america seasonal precipitation prediction by gradient-boosting machine-learning approach. Atmosphere, 13(2).
Morrison, H., van Lier-Walqui, M., Fridlind, A. M., et al. (2020). Confronting the challenge of modeling cloud and precipitation microphysics. Journal of Advances in Modeling Earth Systems, 12.
NOOA (2017). Subseasonal and seasonal forecasting innovation: Plans for the twenty-first century - annotated outline. Technical report, NOAA.
Rumelhart, D. E., Hinton, G. E., and Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323:533–536.
Sales, R. A., de Oliveira, E. C., Delgado, R. C., Leite, M. C. T., Ribeiro, W. R., and Berilli, S. S. (2018). Sazonal and interanual rainfall variability for colatina, espírito santo, brazil.
Yan, Z., Lu, X., Wu, L., Liu, F., Qiu, R., Cui, Y., and Ma, X. (2025). Evaluation of precipitation forecasting based on graphcast over mainland china.
Anochi, J. A., de Almeida, V. A., and de Campos Velho, H. F. (2021). Machine learning for climate precipitation prediction modeling over south america. Remote Sensing.
Anochi, J. A. and Shimizu, M. H. (2025). Precipitation forecasting and drought monitoring in south america using a machine learning approach. Meteorology.
Chung, J., Gulcehre, C., Cho, K., and Bengio, Y. (2014). Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555.
Copernicus Climate Change Service (C3S) (2017). Era5: Fifth generation of ecmwf atmospheric reanalyses of the global climate.
Domingos, M., Júnior, V. S., Anochi, J., and Shiguemori, E. (2025). Exploring deep learning techniques for seasonal prediction of autumn precipitation in south america. In Anais do XVI Workshop de Computação Aplicada à Gestão do Meio Ambiente e Recursos Naturais, pages 157–166. SBC.
Figueroa, S. N., Bonatti, J. P., Kubota, P. Y., et al. (2016). The brazilian global atmospheric model (bam): Performance for tropical rainfall forecasting and sensitivity to convective scheme and horizontal resolution. Weather and Forecasting, 31(5):1547–1565.
Hersbach, H. et al. (2020). The era5 global reanalysis. Quarterly Journal of the Royal Meteorological Society.
Hochreiter, S. and Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8):1735–1780.
Lam, R., Sanchez-Gonzalez, A., Willson, M., Wirnsberger, P., Fortunato, M., Alet, F., Ravuri, S., Ewalds, T., Eaton-Rosen, Z., Hu, W., Merose, A., Hoyer, S., Holland, G., Vinyals, O., Stott, J., Pritzel, A., Mohamed, S., and Battaglia, P. (2023). Learning skillful medium-range global weather forecasting.
LeCun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324.
Lundberg, S. M. and Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems, volume 30.
Monego, V. S., Anochi, J. A., and de Campos Velho, H. F. (2022). South america seasonal precipitation prediction by gradient-boosting machine-learning approach. Atmosphere, 13(2).
Morrison, H., van Lier-Walqui, M., Fridlind, A. M., et al. (2020). Confronting the challenge of modeling cloud and precipitation microphysics. Journal of Advances in Modeling Earth Systems, 12.
NOOA (2017). Subseasonal and seasonal forecasting innovation: Plans for the twenty-first century - annotated outline. Technical report, NOAA.
Rumelhart, D. E., Hinton, G. E., and Williams, R. J. (1986). Learning representations by back-propagating errors. Nature, 323:533–536.
Sales, R. A., de Oliveira, E. C., Delgado, R. C., Leite, M. C. T., Ribeiro, W. R., and Berilli, S. S. (2018). Sazonal and interanual rainfall variability for colatina, espírito santo, brazil.
Yan, Z., Lu, X., Wu, L., Liu, F., Qiu, R., Cui, Y., and Ma, X. (2025). Evaluation of precipitation forecasting based on graphcast over mainland china.
Publicado
19/07/2026
Como Citar
COSTA, João Vyctor Ferreira da; DOMINGOS, Matheus Corrêa; SANTIAGO JÚNIOR, Valdivino Alexandre de; GOMES, Jorge Luís.
Subseasonal Precipitation Forecasting in Southeastern Brazil Using Machine Learning with Explainable Artificial Intelligence. In: WORKSHOP DE COMPUTAÇÃO APLICADA À GESTÃO DO MEIO AMBIENTE E RECURSOS NATURAIS (WCAMA), 17. , 2026, Gramado/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 304-313.
ISSN 2595-6124.
DOI: https://doi.org/10.5753/wcama.2026.22630.
