Comparative evaluation of daily rainfall forecasting models using INMET data

• Leonardo Costa de Souza USP
• Iberê Luiz Caldas USP
• Diego Minatel USP

DOI: https://doi.org/10.5753/eniac.2026.31109

Resumo

Accurate rainfall forecasting remains a complex challenge, yet it is vital for agriculture, water resource management, and urban planning. Given Brazil’s climatic diversity, understanding how different forecasting models perform across regions is particularly important. Using historical data from 46 selected INMET automatic stations distributed across all five Brazilian geographic regions and spanning 2015 to 2026, we evaluated six machine learning models–Random Forest, XGBoost, LightGBM, MLP, LSTM, and Reservoir Computing–for daily precipitation forecasting, using ARIMAX as a baseline. Our results suggest that the machine learning approaches consistently outperform the statistical baseline across all Brazilian regions, with neural-network-based models achieving the highest overall performance. These findings highlight the potential of nonlinear learning methods for improving daily precipitation forecasting across diverse climatic conditions in Brazil.

Referências

Alejo-Sanchez, L. E., Márquez-Grajales, A., Salas-Martínez, F., Franco-Arcega, A., López-Morales, V., Acevedo-Sandoval, O. A., González-Ramírez, C. A., and Villegas-Vega, R. (2025). Missing data imputation of climate time series: A review. MethodsX, 15:103455.

Atkinson, B. (2019). Precipitation. In Man and Environmental Processes, pages 23–37. Routledge.

Barrera-Animas, A. Y., Oyedele, L. O., Bilal, M., Akinosho, T. D., Delgado, J. M. D., and Akanbi, L. A. (2022). Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting. Machine Learning with Applications, 7:100204.

Burlando, P., Montanari, A., and Ranzi, R. (1996). Forecasting of storm rainfall by combined use of radar, rain gages and linear models. Atmospheric research, 42(1-4):199–216.

Chow, V. T. (1971). Handbook of applied hydrology.

Cutler, A., Cutler, D. R., and Stevens, J. R. (2012). Random forests. In Ensemble machine learning, pages 157–175. Springer.

da Silva, S. T., Milani, L. C., Gabrick, E. C., Iarosz, K. C., Viana, R. L., Caldas, I. L., and Batista, A. M. (2025). Rainfall forecast in brazil using machine learning. Chaos: An Interdisciplinary Journal of Nonlinear Science, 35(7).

Das, P. K., Sahu, R. L., and Swain, P. C. (2024). Comparative analysis of machine learning models for rainfall prediction. Journal of Atmospheric and Solar-Terrestrial Physics, 264:106340.

Domingos, M. C., Santiago Júnior, V. A., Anochi, J. A., Shiguemori, E. H., Costa dos Santos, L. M., Pereira, H. C. S., and Oliveira, A. E. C. (2025). Exploring machine learning, deep learning, and explainable ai methods for seasonal precipitation prediction in south america. arXiv preprint arXiv:2512.13910.

Gardner, M. W. and Dorling, S. R. (1998). Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences. Atmospheric Environment, 32(14-15):2627–2636.

Griffith, A., Pomerance, A., and Gauthier, D. J. (2019). Forecasting chaotic systems with very low connectivity reservoir computers. Chaos: An Interdisciplinary Journal of Nonlinear Science, 29(12).

Jaeger, H. (2001). The “echo state” approach to analysing and training recurrent neural networks-with an erratum note. Bonn, Germany: German national research center for information technology gmd technical report, 148(34):13.

Kumar, D. and Richariya, P. (2024). Leveraging sarimax for accurate rainfall forecasting in india. International Journal of Innovations in Science, Engineering And Management, pages 37–47.

Liu, Z. (2015). Comparison of versions 6 and 7 3-hourly trmm multi-satellite precipitation analysis (tmpa) research products. Atmospheric Research, 163:91–101.

Lukoševičius, M., Jaeger, H., and Schrauwen, B. (2012). Reservoir computing trends. KI-Künstliche Intelligenz, 26(4):365–371.

Monteiro, L. A. F., do Nascimento, F. I. C., de Oliveira-Júnior, J. F., Nunes, D. D., Mendes, D., de Gois, G., Sanches, F. d. O., Wollmann, C. A., Watanabe, M., and Gobo, J. P. A. (2024). Rainfall projections for the brazilian legal amazon: An artificial neural networks first approach. Climate, 12(11).

New, M., Todd, M., Hulme, M., and Jones, P. (2001). Precipitation measurements and trends in the twentieth century. International Journal of Climatology: A Journal of the Royal Meteorological Society, 21(15):1889–1922.

Pathak, J., Lu, Z., Hunt, B. R., Girvan, M., and Ott, E. (2017). Using machine learning to replicate chaotic attractors and calculate lyapunov exponents from data. Chaos: An Interdisciplinary Journal of Nonlinear Science, 27(12).

Pei, L. and Wang, K. (2023). Precipitation forecast in china based on reservoir computing. The European Physical Journal Special Topics, 232(5):695–702.

Rodrigues, G. C., Martins, J. D., Da Silva, F. G., Carlesso, R., and Pereira, L. S. (2013). Modelling economic impacts of deficit irrigated maize in brazil with consideration of different rainfall regimes. biosystems engineering, 116(1):97–110.

Rozante, J. R., Gutierrez, E. R., Fernandes, A. d. A., and Vila, D. A. (2020). Performance of precipitation products obtained from combinations of satellite and surface observations. International Journal of Remote Sensing, 41(19):7585–7604.

Rozante, J. R., Moreira, D. S., de Goncalves, L. G. G., and Vila, D. A. (2010). Combining trmm and surface observations of precipitation: technique and validation over south america. Weather and forecasting, 25(3):885–894.

Solomatine, D. P. and Ostfeld, A. (2008). Data-driven modelling: some past experiences and new approaches. Journal of Hydroinformatics, 10(1):3–22.

Tinôco, I. C. M., Bezerra, B. G., Lucio, P. S., and Barbosa, L. M. (2018). Characterization of rainfall patterns in the semiarid brazil. Anuário do Instituto de Geociências, 41(2):397–409.

Wallace, J. M. and Hobbs, P. V. (2006). Atmospheric Science: An Introductory Survey, volume 92 of International Geophysics Series. Elsevier Academic Press, Amsterdam, 2nd edition.

Wangwongchai, A., Waqas, M., Dechpichai, P., Hlaing, P. T., Ahmad, S., and Humphries, U. W. (2023). Imputation of missing daily rainfall data; a comparison between artificial intelligence and statistical techniques. MethodsX, 11:102459.

Wani, O. A., Mahdi, S. S., Yeasin, M., Kumar, S. S., Gagnon, A. S., Danish, F., Al-Ansari, N., El-Hendawy, S., and Mattar, M. A. (2024). Predicting rainfall using machine learning, deep learning, and time series models across an altitudinal gradient in the north-western himalayas. Scientific Reports, 14(1):27876.