Enhancing Precipitation Forecasting with Time Series Models: A Comparative Study of Prophet and Other Machine Learning Algorithms

Volume: 10    Issue: 2

Year of Publication: 2025

Pages: (18-29)

Authors: Rathnayaka Pathiranage Ravindi Kawushini, Danso Solomon Danquah, Rathnayaka Pathiranage Tharindu Kaushan

DOI: https://doi.org/10.5281/zenodo.15575203

Download Full Text (PDF)

Abstract

The exact precipitation forecasting is important for different climate-related things like agriculture planning, water resource management, and disaster preparedness. This particular study aims at time series forecasting using a typical model such as the Prophet model from Facebook, particularly for forecasting the precipitation levels in Sri Lanka. Moreover, it compares the performance of Prophet with various machine learning algorithms such as ARIMA, Random Forest, and XGBoost for discovering the model that is most successful in improving the accuracy of precipitation prediction. The models will be evaluated on several metrics such as Mean Absolute Percentage Error (MAPE), Root Mean Squared Error (RMSE), and R2 score. Results show that for handling seasonal trends and long-term patterns, Prophet beats all other models under study; thus, it\"s the most favorable for precipitation forecasting.

References

1. Vasselin, H. (2022, March 17). Is Facebook prophet suited for doing good predictions in a real-world project? Medium. Retrieved March 3, 2023, from https://medium.com/artefact-engineering-and-data-science/is-facebook-prophet-suited-for-doing-good-predictions-in-a-real-world-project-44be1fe4ce91

2. R. J. , & A. G. (2018) Hyndman, Forecasting: principles and practice (2nd ed.). OTexts.

3. Md. A. Saleh and H. M. Rasel, ‘Performance evaluation of Machine Learning based regression models for rainfall forecasting’, Jan. 25, 2024. doi: 10.21203/rs.3.rs-3856741/v1.

4. S. J. Taylor and B. Letham, ‘Forecasting at Scale’, Am Stat, vol. 72, no. 1, pp. 37-45, Jan. 2018, doi: 10.1080/00031305.2017.1380080.

5. Y. Zhou, C. Matyas, H. Li, and J. Tang, ‘Conditions associated with rain field size for tropical cyclones landfalling over the Eastern United States’, Atmos Res, vol. 214, pp. 375-385, Dec. 2018, doi: 10.1016/j.atmosres.2018.08.019.

6. J. Kim and T. J. Kim, ‘Application of Facebook’s Prophet Model for Forecasting Meteorological Data’, Journal of the Korean Society of Hazard Mitigation, vol. 21, no. 2, pp. 53-58, Apr. 2021, doi: 10.9798/KOSHAM.2021.21.2.53.

7. F. R. Aderyani, S. Jamshid Mousavi, and F. Jafari, ‘Short-term rainfall forecasting using machine learning-based approaches of PSO-SVR, LSTM and CNN’, J Hydrol (Amst), vol. 614, p. 128463, Nov. 2022, doi: 10.1016/j.jhydrol.2022.128463.

8. T. Chen and C. Guestrin, ‘XGBoost’, in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, New York, NY, USA: ACM, Aug. 2016, pp. 785-794. doi: 10.1145/2939672.2939785.

9. LEOBREIMAN (Y. Freund & R. Schapire), ‘Random Forests’, in Machine Learning: Proceedings of the Thirteenth International conference, 2001, pp. 1-28.

10. Y. Liang, D. Feng, and Z. Sun, ‘Comparison of four machine learning models for forecasting daily reference evaporation based on public weather forecast data’, Jul. 31, 2023. doi: 10.5194/hess-2023-158.

11. P. Li and J.-S. Zhang, ‘A new hybrid method for China’s energy supply security forecasting based on ARIMA and XGBoost’, Energies (Basel), vol. 11, no. 7, p. 1687, 2018.

12. S. Dörterler, S. Arslan, and D. Özdemir, ‘Unlocking the potential: A review of artificial intelligence applications in wind energy’, Expert Syst, p. e13716, 2024.

13. A. Yamagami, ‘Atmospheric Science Letters’, Atmospheric Science Letters, vol. 22, no. 8, p. e1038, 2021.

14. S. García, S. Ramírez-Gallego, J. Luengo, J. M. Benítez, and F. Herrera, ‘Big data preprocessing: methods and prospects’, Big Data Anal, vol. 1, pp. 1-22, 2016.

15. M. Kuhn and K. Johnson, Feature engineering and selection: A practical approach for predictive models. Chapman and Hall/CRC, 2019.

16. H. Balti et al., ‘Big data-based architecture for drought forecasting using LSTM, ARIMA, and Prophet: Case study of the Jiangsu Province, China’, in 2021 International Congress of Advanced Technology and Engineering (ICOTEN), IEEE, 2021, pp. 1-8.

17. M. Yeshwanth, P. R. S. Kumar, and Dr. G. M. M. E. , Ph.D, ‘Comparative Study of Machine Learning Algorithms for Rainfall Prediction’, International Journal of Trend in Scientific Research and Development, vol. Volume-3, no. Issue-3, pp. 677-681, Apr. 2019, doi: 10.31142/ijtsrd22961.

18. V. I. Kontopoulou, A. D. Panagopoulos, I. Kakkos, and G. K. Matsopoulos, ‘A review of ARIMA vs. machine learning approaches for time series forecasting in data driven networks’, Future Internet, vol. 15, no. 8, p. 255, 2023.

19. M. Yeshwanth, P. R. S. Kumar, and Dr. G. M. M. E. , Ph.D, ‘Comparative Study of Machine Learning Algorithms for Rainfall Prediction’, International Journal of Trend in Scientific Research and Development, vol. Volume-3, no. Issue-3, pp. 677-681, Apr. 2019, doi: 10.31142/ijtsrd22961.

20. J. Kihoro, R. Otieno, and C. Wafula, ‘Seasonal time series forecasting: a comparative study of arima and ann models’, African Journal of Science and Technology, vol. 5, no. 2, May 2006, doi: 10.4314/ajst.v5i2.15330.

21. Monigatti, L. (2022, August 23). Time series problems simply explained as fast food combo meals. Medium. Retrieved March 3, 2023, from https://towardsdatascience.com/time-series-problems-simply-explained-as-fast-food-combo-meals-70c6eb9bdef

22. Merwe, R. van der. (2018, November 15). Implementing facebook prophet efficiently. Medium. Retrieved March 3, 2023, from https://towardsdatascience.com/implementing-facebook-prophet-efficiently-c241305405a3

Keywords

Precipitation Forecasting, Time Series Modeling, Prophet, Machine Learning Algorithms