Physics-Informed Machine Learning for Extreme Rainfall Nowcasting: A Rapid Review Toward Sustainable Early Warning Systems
Resumo
Extreme rainfall nowcasting is essential for urban early warning systems, yet remains a challenging task. Data-driven machine learning models show promising performance but often suffer from limited generalization, physical inconsistency, and high computational cost. Physics-Informed Machine Learning (PIML) addresses these issues by integrating physical knowledge into learning algorithms. This paper presents a Rapid Review of PIML for extreme rainfall nowcasting, organizing integration strategies and addressing terminological fragmentation. The results reveal different levels of physical integration and limited focus on nowcasting. Most works neglect computational efficiency and resource-aware evaluation, highlighting challenges for developing reliable and sustainable nowcasting systems.Referências
Breder, G. B., Vasconcelos, G., and Ferro, M. (2025). Beyond accuracy: A comparative study of machine learning models for extreme weather forecasting in rio de janeiro with a green ai perspective. In Anais do Encontro Nacional de Inteligência Artificial e Computacional (ENIAC), pages 1996–2007, Fortaleza, CE, Brasil. Sociedade Brasileira de Computação.
Brum, D., Teylo, L., da Silva, F. P., and et al. (2025). Towards sustainable nowcasting: Assessing the environmental costs of ai-driven extreme rainfall prediction. Preprint. Version 1.
Camuffo, D., Becherini, F., and della Valle, A. (2020). Relationship between selected percentiles and return periods of extreme events. Acta Geophysica, 68:1201–1211.
Cartaxo, B., Pinto, G., and Soares, S. (2020). Rapid reviews in software engineering. arXiv preprint.
Clark, S. R. and Jaffrés, J. B. (2025). Associations between deep learning runoff predictions and hydrogeological conditions in australia. Journal of Hydrology, 651:132569.
da Silva, F. P., da Silva, A. S., and da Silva, M. G. A. J. (2022). Extreme rainfall events in the rio de janeiro city (brazil): description and a numerical sensitivity case study. Meteorology and Atmospheric Physics, 134:77.
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):073116.
de Vasconcelos, G. (2026). Replication package for physics-informed machine learning in extreme rainfall nowcasting.
de Vasconcelos, G. S. (2017). Evaluation of precipitation forecasts from the ncar/cam 3.1 model over brazil for the period 2012–2017. Master’s thesis, Universidade Federal Fluminense, Niterói, Brazil.
Dereczynski, C., Calado, R., and Barros, A. B. (2017). Chuvas extremas no município do rio de janeiro: Histórico a partir do século xix. Anuário do Instituto de Geociências UFRJ, 40(2):17–30.
Fonseca, A., Porto, F., Ferro, M., Ogasawara, E., and Bezerra, E. (2022). Analysis of precipitation data in rio de janeiro city using extreme value theory. In Anais Estendidos do XXXVII Simpósio Brasileiro de Bancos de Dados, pages 193–198, Porto Alegre, RS, Brasil. SBC.
Frnda, J., Durica, M., Rozhon, J., Vojtekova, M., Nedoma, J., and Martinek, R. (2022). Ecmwf short-term prediction accuracy improvement by deep learning. Scientific Reports, 12:7898.
Hao, Z., Liu, S., Zhang, Y., Ying, C., Feng, Y., Su, H., and Zhu, J. (2023). Physics-informed machine learning: A survey on problems, methods and applications. arXiv preprint.
Hsu, L.-H., Chiang, C.-C., Lin, K.-L., Lin, H.-H., Chu, J.-L., Yu, Y.-C., and Fahn, C.-S. (2024). Downscaling taiwan precipitation with a residual deep learning approach. Geoscience Letters, 11:23.
Karpatne, A., Atluri, G., Faghmous, J. H., Steinbach, M., Banerjee, A., Ganguly, A., Shekhar, S., Samatova, N., and Kumar, V. (2017). Theory-guided data science: A new paradigm for scientific discovery from data. IEEE Transactions on Knowledge and Data Engineering, 29:2318–2331.
Kashinath, K., Mustafa, M., Albert, A., Wu, J.-L., Jiang, C., Esmaeilzadeh, S., Azizzadenesheli, K., Wang, R., Chattopadhyay, A., Singh, A., Manepalli, A., Chirila, D., Yu, R., Walters, R., White, B., Xiao, H., Tchelepi, H. A., Marcus, P., Anandkumar, A., Hassanzadeh, P., and Prabhat (2021). Physics-informed machine learning: case studies for weather and climate modelling. Philosophical Transactions of the Royal Society A, 379(2194):20200093.
Kitchenham, B. and Charters, S. (2007). Guidelines for performing systematic literature reviews in software engineering. Technical Report EBSE-2007-01, Keele University and University of Durham, UK.
Kurth, T., Subramanian, S., Harrington, P., Pathak, J., Mardani, M., Hall, D., Miele, A., Kashinath, K., and Anandkumar, A. (2023). Fourcastnet: Accelerating global high-resolution weather forecasting using adaptive fourier neural operators. In Proceedings of the Platform for Advanced Scientific Computing Conference (PASC ’23), pages 1–11, New York, NY, USA. ACM.
Li, B., Zhu, Z., Zhong, X., Tan, R., Wang, Y., Lan, W., and Li, H. (2025). One-kilometer resolution forecasts of hourly precipitation over china using machine learning models. Atmospheric Science Letters, 26(3):1–21.
Li, W., Liu, Z., Chen, K., Chen, H., Liang, S., Zou, Z., and Shi, Z. (2024). Deepphysinet: Bridging deep learning and atmospheric physics for accurate and continuous weather modeling. arXiv preprint.
Liu, Q., Xiao, Y., Gui, Y., Dai, G., Li, H., Zhou, X., Ren, A., Zhou, G., and Shen, J. (2025). MMF-RNN: A multimodal fusion model for precipitation nowcasting using radar and ground station data. IEEE Transactions on Geoscience and Remote Sensing, 63:1–16.
Ma, Z., Zhang, H., and Liu, J. (2023). Mm-rnn: A multimodal rnn for precipitation nowcasting. IEEE Transactions on Geoscience and Remote Sensing, 61:1–14.
Migliani, F. (2025). Rio 60°c aquecimento extremo, tempestades violentas. a cidade do rio de janeiro encara sua nova realidade climática. estamos preparados? Pulitzer Center.
Nerini, D., Foresti, L., Leuenberger, D., Robert, S., and Germann, U. (2019). A reduced-space ensemble kalman filter approach for flow-dependent integration of radar extrapolation nowcasts and nwp precipitation ensembles. Monthly Weather Review, 147(3):987–1006.
Page, M. J., Moher, D., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., et al. (2021). Prisma 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ, 372:n160.
Petersen, K., Vakkalanka, S., and Kuzniarz, L. (2015). Guidelines for conducting systematic mapping studies in software engineering: An update. Information and Software Technology, 64:1–18.
Porto, F., Ferro, M., Ogasawara, E., Moeda, T., Barros, C. D. T., Silva, A. C., Zorrilla, R., Pereira, R. S., Castro, R., Silva, J. V., Salles, R., Fonseca, A., Hermsdorff, J., Magalhães, M., Sá, V., Simões, A. A., Cardoso, C., and Bezerra, E. (2022). Machine learning approaches to extreme weather events forecast in urban areas: Challenges and initial results. Supercomputing Frontiers and Innovations, 9(1):49–73.
Pristo, M. V. d. J., Dereczynski, C. P., Souza, P. R. d., and Menezes, W. F. (2018). Climatologia de chuvas intensas no município do rio de janeiro. Revista Brasileira de Meteorologia, 33(4):615–630.
Ravuri, S., Lenc, K., Willson, M., Kangin, D., Lam, R., Mirowski, P., Fitzsimons, M., Athanassiadou, M., Kashem, S., Madge, S., Prudden, R., Mandhane, A., Clark, A., Brock, A., Simonyan, K., Hadsell, R., Robinson, N., Clancy, E., Arribas, A., and Mohamed, S. (2021). Skillful precipitation nowcasting using deep generative models of radar. Nature, 597(7878):672–677.
Sistema Alerta Rio (2025). Termos meteorológicos. Sistema Alerta Rio. Accessed: 07 Feb 2026.
Ta Gia, K., Tran Van, H., Phan Thanh, A., and Phan Minh, N. (2025). Thor: A motion-dependent physics-informed deep learning framework with constraint-centric theory of functional connections for rainfall nowcasting. Research Square. Preprint.
World Economic Forum (2025). The global risks report 2025.
World Meteorological Organization (2023). Guidelines on the definition and characterization of extreme weather and climate events.
Zhang, Y., Long, M., Chen, K., Xing, L., Jin, R., Jordan, M. I., and Wang, J. (2023). Skilful nowcasting of extreme precipitation with nowcastnet. Nature 2023 619:7970, 619:526–532.
Brum, D., Teylo, L., da Silva, F. P., and et al. (2025). Towards sustainable nowcasting: Assessing the environmental costs of ai-driven extreme rainfall prediction. Preprint. Version 1.
Camuffo, D., Becherini, F., and della Valle, A. (2020). Relationship between selected percentiles and return periods of extreme events. Acta Geophysica, 68:1201–1211.
Cartaxo, B., Pinto, G., and Soares, S. (2020). Rapid reviews in software engineering. arXiv preprint.
Clark, S. R. and Jaffrés, J. B. (2025). Associations between deep learning runoff predictions and hydrogeological conditions in australia. Journal of Hydrology, 651:132569.
da Silva, F. P., da Silva, A. S., and da Silva, M. G. A. J. (2022). Extreme rainfall events in the rio de janeiro city (brazil): description and a numerical sensitivity case study. Meteorology and Atmospheric Physics, 134:77.
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):073116.
de Vasconcelos, G. (2026). Replication package for physics-informed machine learning in extreme rainfall nowcasting.
de Vasconcelos, G. S. (2017). Evaluation of precipitation forecasts from the ncar/cam 3.1 model over brazil for the period 2012–2017. Master’s thesis, Universidade Federal Fluminense, Niterói, Brazil.
Dereczynski, C., Calado, R., and Barros, A. B. (2017). Chuvas extremas no município do rio de janeiro: Histórico a partir do século xix. Anuário do Instituto de Geociências UFRJ, 40(2):17–30.
Fonseca, A., Porto, F., Ferro, M., Ogasawara, E., and Bezerra, E. (2022). Analysis of precipitation data in rio de janeiro city using extreme value theory. In Anais Estendidos do XXXVII Simpósio Brasileiro de Bancos de Dados, pages 193–198, Porto Alegre, RS, Brasil. SBC.
Frnda, J., Durica, M., Rozhon, J., Vojtekova, M., Nedoma, J., and Martinek, R. (2022). Ecmwf short-term prediction accuracy improvement by deep learning. Scientific Reports, 12:7898.
Hao, Z., Liu, S., Zhang, Y., Ying, C., Feng, Y., Su, H., and Zhu, J. (2023). Physics-informed machine learning: A survey on problems, methods and applications. arXiv preprint.
Hsu, L.-H., Chiang, C.-C., Lin, K.-L., Lin, H.-H., Chu, J.-L., Yu, Y.-C., and Fahn, C.-S. (2024). Downscaling taiwan precipitation with a residual deep learning approach. Geoscience Letters, 11:23.
Karpatne, A., Atluri, G., Faghmous, J. H., Steinbach, M., Banerjee, A., Ganguly, A., Shekhar, S., Samatova, N., and Kumar, V. (2017). Theory-guided data science: A new paradigm for scientific discovery from data. IEEE Transactions on Knowledge and Data Engineering, 29:2318–2331.
Kashinath, K., Mustafa, M., Albert, A., Wu, J.-L., Jiang, C., Esmaeilzadeh, S., Azizzadenesheli, K., Wang, R., Chattopadhyay, A., Singh, A., Manepalli, A., Chirila, D., Yu, R., Walters, R., White, B., Xiao, H., Tchelepi, H. A., Marcus, P., Anandkumar, A., Hassanzadeh, P., and Prabhat (2021). Physics-informed machine learning: case studies for weather and climate modelling. Philosophical Transactions of the Royal Society A, 379(2194):20200093.
Kitchenham, B. and Charters, S. (2007). Guidelines for performing systematic literature reviews in software engineering. Technical Report EBSE-2007-01, Keele University and University of Durham, UK.
Kurth, T., Subramanian, S., Harrington, P., Pathak, J., Mardani, M., Hall, D., Miele, A., Kashinath, K., and Anandkumar, A. (2023). Fourcastnet: Accelerating global high-resolution weather forecasting using adaptive fourier neural operators. In Proceedings of the Platform for Advanced Scientific Computing Conference (PASC ’23), pages 1–11, New York, NY, USA. ACM.
Li, B., Zhu, Z., Zhong, X., Tan, R., Wang, Y., Lan, W., and Li, H. (2025). One-kilometer resolution forecasts of hourly precipitation over china using machine learning models. Atmospheric Science Letters, 26(3):1–21.
Li, W., Liu, Z., Chen, K., Chen, H., Liang, S., Zou, Z., and Shi, Z. (2024). Deepphysinet: Bridging deep learning and atmospheric physics for accurate and continuous weather modeling. arXiv preprint.
Liu, Q., Xiao, Y., Gui, Y., Dai, G., Li, H., Zhou, X., Ren, A., Zhou, G., and Shen, J. (2025). MMF-RNN: A multimodal fusion model for precipitation nowcasting using radar and ground station data. IEEE Transactions on Geoscience and Remote Sensing, 63:1–16.
Ma, Z., Zhang, H., and Liu, J. (2023). Mm-rnn: A multimodal rnn for precipitation nowcasting. IEEE Transactions on Geoscience and Remote Sensing, 61:1–14.
Migliani, F. (2025). Rio 60°c aquecimento extremo, tempestades violentas. a cidade do rio de janeiro encara sua nova realidade climática. estamos preparados? Pulitzer Center.
Nerini, D., Foresti, L., Leuenberger, D., Robert, S., and Germann, U. (2019). A reduced-space ensemble kalman filter approach for flow-dependent integration of radar extrapolation nowcasts and nwp precipitation ensembles. Monthly Weather Review, 147(3):987–1006.
Page, M. J., Moher, D., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., et al. (2021). Prisma 2020 explanation and elaboration: updated guidance and exemplars for reporting systematic reviews. BMJ, 372:n160.
Petersen, K., Vakkalanka, S., and Kuzniarz, L. (2015). Guidelines for conducting systematic mapping studies in software engineering: An update. Information and Software Technology, 64:1–18.
Porto, F., Ferro, M., Ogasawara, E., Moeda, T., Barros, C. D. T., Silva, A. C., Zorrilla, R., Pereira, R. S., Castro, R., Silva, J. V., Salles, R., Fonseca, A., Hermsdorff, J., Magalhães, M., Sá, V., Simões, A. A., Cardoso, C., and Bezerra, E. (2022). Machine learning approaches to extreme weather events forecast in urban areas: Challenges and initial results. Supercomputing Frontiers and Innovations, 9(1):49–73.
Pristo, M. V. d. J., Dereczynski, C. P., Souza, P. R. d., and Menezes, W. F. (2018). Climatologia de chuvas intensas no município do rio de janeiro. Revista Brasileira de Meteorologia, 33(4):615–630.
Ravuri, S., Lenc, K., Willson, M., Kangin, D., Lam, R., Mirowski, P., Fitzsimons, M., Athanassiadou, M., Kashem, S., Madge, S., Prudden, R., Mandhane, A., Clark, A., Brock, A., Simonyan, K., Hadsell, R., Robinson, N., Clancy, E., Arribas, A., and Mohamed, S. (2021). Skillful precipitation nowcasting using deep generative models of radar. Nature, 597(7878):672–677.
Sistema Alerta Rio (2025). Termos meteorológicos. Sistema Alerta Rio. Accessed: 07 Feb 2026.
Ta Gia, K., Tran Van, H., Phan Thanh, A., and Phan Minh, N. (2025). Thor: A motion-dependent physics-informed deep learning framework with constraint-centric theory of functional connections for rainfall nowcasting. Research Square. Preprint.
World Economic Forum (2025). The global risks report 2025.
World Meteorological Organization (2023). Guidelines on the definition and characterization of extreme weather and climate events.
Zhang, Y., Long, M., Chen, K., Xing, L., Jin, R., Jordan, M. I., and Wang, J. (2023). Skilful nowcasting of extreme precipitation with nowcastnet. Nature 2023 619:7970, 619:526–532.
Publicado
19/07/2026
Como Citar
VASCONCELOS, Gyslla; DORNELLAS, Nayara; AGUIAR, Laisa; FERRO, Mariza.
Physics-Informed Machine Learning for Extreme Rainfall Nowcasting: A Rapid Review Toward Sustainable Early Warning Systems. In: SEMINÁRIO INTEGRADO DE SOFTWARE E HARDWARE (SEMISH), 53. , 2026, Gramado/RS.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 506-517.
ISSN 2595-6205.
DOI: https://doi.org/10.5753/semish.2026.23878.
