A machine learning and statistical learning-based pipeline to perform multipoint rainfall forecasting

• Eduardo Carvalho Vale Institute of Technology
• Ewerton Oliveira Vale Institute of Technology / Federal University of Pará
• Rafael Rocha Vale Institute of Technology / Federal University of Pará
• Nikolas Carneiro Vale Institute of Technology
• Renata Tedeschi Vale Institute of Technology
• Ronnie Alves Vale Institute of Technology

DOI: https://doi.org/10.5753/eniac.2023.234376

Abstract

Analyzing and predicting precipitation is crucial for society, particularly when extreme rainfall and floods occur. Such events impact socio-economic structures and can lead to fatalities. Understanding rain formation and associated variables helps develop predictive models for precipitation levels, aiding decision-making in production chains and urban mobility. Recent advancements in this field result from increased computer processing power and the availability of meteorological data worldwide. This study focuses on evaluating SARIMA, RNN, and XGBoost models for predicting monthly rainfall along a northern Brazilian railway. Using lagged precipitation values, SARIMA performed better for 11 out of 13 points (with R² ranging from 0.704 to 0.817), while RNN outperformed in the remaining points (15.39% of the evaluated points).

Keywords: Machine Learning, Statistical Learning, Rainfall Forecast, Preciptation

References

Berrang-Ford, L., Sietsma, A. J., Callaghan, M., Minx, J. C., Scheelbeek, P. F., Haddaway, N. R., Haines, A., and Dangour, A. D. (2021). Systematic mapping of global research on climate and health: a machine learning review. The Lancet Planetary Health, 5(8):e514–e525.

Cerveny, R. S., Bessemoulin, P., Burt, C. C., Cooper, M. A., Cunjie, Z., Dewan, A., Finch, J., Holle, R. L., Kalkstein, L., Kruger, A., et al. (2017). Wmo assessment of weather and climate mortality extremes: lightning, tropical cyclones, tornadoes, and hail. Weather, climate, and society, 9(3):487–497.

Chhetri, M., Kumar, S., Pratim Roy, P., and Kim, B.-G. (2020). Deep blstm-gru model for monthly rainfall prediction: A case study of simtokha, bhutan. Remote sensing, 12(19):3174.

Fang, J., Li, M., and Shi, P. (2015). Mapping flood risk of the world. World Atlas of Natural Disaster Risk, pages 69–102.

Fernandes, E., Rocha, R. L., Ferreira, B., Carvalho, E., Siravenha, A. C., Gomes, A. C. S., Carvalho, S., and de Souza, C. R. (2018). An ensemble of convolutional neural networks for unbalanced datasets: A case study with wagon component inspection. In 2018 International Joint Conference on Neural Networks (IJCNN), pages 1–6. IEEE.

Group, R. S. Re swiss group. http://bit.ly/SwissReFlood. Accessed: 2023-04-18.

McGuffie, K. and Henderson-Sellers, A. (2001). Forty years of numerical climate modelling. International Journal of Climatology: A Journal of the Royal Meteorological Society, 21(9):1067–1109.

Moazzam, M. F. U., Rahman, G., Munawar, S., Tariq, A., Safdar, Q., and Lee, B.-G. (2022). Trends of rainfall variability and drought monitoring using standardized precipitation index in a scarcely gauged basin of northern pakistan. Water, 14(7):1132.

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):243.

NWS. Nws preliminary us flood fatality statistics. https://www.weather.gov/arx/usflood. Accessed: 2023-04-18.

Qerimi, Q. and Sergi, B. S. (2022). The case for global regulation of carbon capture and storage and artificial intelligence for climate change. International Journal of Greenhouse Gas Control, 120:103757.

Rocha, R. L., Silva, C. D., Gomes, A. C. S., Ferreira, B. V., Carvalho, E. C., Siravenha, A. C. Q., and Carvalho, S. R. (2019). Image inspection of railcar structural components: An approach through deep learning and discrete fourier transform. In Anais do VII Symposium on Knowledge Discovery, Mining and Learning, pages 33–40. SBC.

Santos, R. S. and Qin, L. (2019). Risk capital and emerging technologies: innovation and investment patterns based on artificial intelligence patent data analysis. Journal of Risk and Financial Management, 12(4):189.

Su, B., Xiao, C., Zhao, H., Huang, Y., Dou, T., Wang, X., and Chen, D. (2022). Estimated changes in different forms of precipitation (snow, sleet, and rain) across china: 1961–2016. Atmospheric Research, 270:106078.

Yin, K., Cai, F., and Huang, C. (2022). How does artificial intelligence development affect green technology innovation in china? evidence from dynamic panel data analysis. Environmental Science and Pollution Research, pages 1–25.

Zhou, Z., Ren, J., He, X., and Liu, S. (2021). A comparative study of extensive machine learning models for predicting long-term monthly rainfall with an ensemble of climatic and meteorological predictors. Hydrological Processes, 35(11):e14424.