Comparison of Monthly Rainfall Forecasting Methods in Narathiwat Province
Keywords:
rainfall forecasting, holt-winter’s, ARIMA, time series analysisAbstract
Narathiwat Province exhibits pronounced seasonal variability in average monthly rainfall, which significantly impacts water resource management and disaster preparedness. This study aims to analyze time series patterns and compare the forecasting accuracy of Holt-Winters’ Additive Exponential Smoothing: HW-AES and the Box-Jenkins ARIMA: ARIMA models. The analysis is based on 132 observations of monthly average rainfall from January 2013 to December 2023. Model performance was evaluated using Mean Absolute Percentage Error: MAPE, Root Mean Square Error: RMSE, and the smallest absolute differences between observed and forecasted rainfall values. The results revealed that the Holt-Winters’: HW model outperformed the ARIMA model, with a lower MAPE (83.86%), RMSE (46.27 millimeters: mm), and forecast-observation difference (116.65 mm), compared to ARIMA’s MAPE (198.26%) and RMSE (100.89 mm). The forecasted rainfall for 2024 showed the lowest value in March (46.03 mm) and the highest in November (402.42 mm). These findings suggest that the HW-AES model is more appropriate for forecasting in regions with distinct seasonal rainfall patterns.
References
Aini, N. N., Iriany, A., Nugroho, W. H., & Wibowo, F. L. (2022). Comparison of adaptive Holt-Winters exponential smoothing and recurrent neural network model for forecasting rainfall in Malang City. ComTech: Computer, Mathematics and Engineering Applications, 13(2), 87−96. https://doi.org/10.21512/comtech.v13i2.7570
Brito, G. R. A., Villaverde, A. R., Quan A. L., & Pérez, M. E. R. (2021). Comparison between SARIMA and Holt–Winters models for forecasting monthly streamflow in the western region of Cuba. SN Applied Sciences, 3, 671. https://doi.org/10.1007/s42452-021-04667-5
Department of Disaster Prevention and Mitigation. (2023). Flood, windstorm, and landslide disaster response plan for Narathiwat Province, 2023. Retrieved from https://backoffice minisitedisaster.go.th/apiv1/apps/minisite_e17/204/content/3107/download?filename=87ddea98ba65edeb4970125534ceff22.pdf (in Thai)
Ghayyib, M. N., Abdulrazzaq, A. A., & Adnan, F. A. (2024). Rainfall forecasting with time series model in Karbala, Iraq. Al-Kut University College Journal, Special Issue, 505-514. Retrieved from https://js.alkutcollege.edu.iq/article_24250_dbe336300b4afc34f03f5b06a8da18f0.pdf
Gowri, L. Manjula, K. R., Sasireka, K., & Deepa, D. (2022). Assessment of statistical models for rainfall forecasting using machine learning technique. Journal of Soft Computing in Civil Engineering, 6(2), 51-67. https://doi.org/10.22115/SCCE.2022.304260.1363
Homdee, T., Senawang, W., Maneeakul, P., Paengnawin, T., & Krutsaeng-anant, P. (2022). Forecasting monthly rainfall in Nakhon Phanom Province at the Lower Namkam Project Station. 27th National Civil Engineering Conference, August 24-26, 2022 (pp. 1-6). Chiang Rai: National Civil Engineering. (in Thai)
Hydro - Informatics Institute. (2023). Rainfall data. Retrieved from https://data.hii.or.th/ dataset/hii-rainfall (in Thai)
Khan, M.M.H., Mustafa, M.R.U., Hossain, M.S., Shams, S. & Julius, A.D. (2023). Short-Term and Long-Term Rainfall Forecasting Using ARIMA Model. International Journal of Environmental Science and Development, 14(5), 292-298. Retrieved from https://pure.hw.ac.uk/ws/portalfiles/portal/103987196/IJESD-V14N5-1447.pdf
Narathiwat Provincial Office. (2023). Narathiwat Provincial development plan 2023-2027 (Rev. ed.). Retrieved from http://www2.narathiwat.go.th/nara2016/files/ com_news_develop_plan/2023-01_09cea4479f3f3b8.pdf (in Thai)
Poh, L. C., & Norrulashikin, S. M. (2022). Forecasting amount of rainfall in peninsular malaysia using holt-winter method and Seasonal Autoregressive Integrated Moving Average (SARIMA). Proceedings of Science and Mathematics, 10, 142 – 152. Retrieved from https://science.utm.my/procscimath/wp-content/uploads/sites/605/2022/10/142-152-Lee-Chi-Poh_Siti-Mariam-Norrulashikin.pdf
Sopipan, N. (2014). Forecasting rainfall in Thailand: A case study of Nakhon Ratchasima Province. International Journal of Environmental, Ecological, Geological and Marine Engineering, 8(11), 712-716. Retrieved from https://www.researchgate.net/publication/275713804_Forecasting_Rainfall_in_Thailand_ A_Case_ Study_of _Nakhon_Ratchasima_Province (in Thai)
Tan, T. J., & Kamisan, N. A. B. (2022). Rainfall forecasting with time series model in Senai, Johor. Proceedings of Science and Mathematics, 11, 169–178. Retrieved from https://science.utm.my/procscimath/wp-content/uploads/sites/605/2022/10/169-178-Tan-Theng-Joe_Nur-Arina-Bazilah-Kamisan.pdf (in Thai)
Tee H.Y. & Mansor R. (2024). Forecasting rainfall volume in Selangor with combined ARIMA model. Journal of Computational Innovation and Analytics, 3(1), 83-103. https://doi.org/10.32890/ jcia2024.3.1.5
Thai Meteorological Department. (2024). Forecast of the impacts from the ENSO phenomenon in 2024. Retrieved from https://www.tmd.go.th/climate/ El-Nino-La-Nina (in Thai)
Thai Meteorological Department. (2024). Rainfall outlook for Thailand 2024. Retrieved from https://www.tmd.go.th/ (in Thai)
Waqas, M., Wannasingha Humphries, U., & Hlaing, P. T. (2024). Time series trend analysis and forecasting of climate variability using deep learning in Thailand. Results in Engineering, 24, 102997. Retrieved from https://www.sciencedirect.com/science/article/pii/S2590123024012520
Wiguna, I. K. A. G., Utami, N. L. P. A. C., Parwita, W. G. S., Udayana, I. P. A. E. D., & Sudipa, I. G. I. (2023). Rainfall forecasting using the Holt-Winters exponential smoothing method conceptual framework. Journal Info Sains : Informatikan dan Sains, 13(1), 15-23. Retrieved from https://ejournal.seaninstitute.or.id/index.php/InfoSains/article/download/2656/2007
Wutthiwongyothin, S., Srijumpa, K., & Sopha, K. (2020). Estimation of missing daily rainfall data using the quantile method. 25th National Civil Engineering Conference (p. 1-7). Chonburi: National Civil Engineering (in Thai)
