Salinity forecasting in the Vietnamese Mekong Delta: Evaluating the predictive power of machine learning approaches using multitemporal lag features
Tập 134 Số 1S-1 (2025), Special Issue: IFGTM 2025
Tập 134 Số 1S-1 (2025)

Salinity forecasting in the Vietnamese Mekong Delta: Evaluating the predictive power of machine learning approaches using multitemporal lag features

Special Issue: IFGTM 2025
https://doi.org/10.26459/hueunijns.v134i1S-1.7876
Đã xuất bản 2025-12-29
Van Quyen Le
Faculty of Engineering, Vietnamese-German University, Ho Chi Minh City, Vietnam
Tuan Cuong Nguyen
Faculty of Engineering, Vietnamese-German University, Ben Cat City, Binh Duong Province, Vietnam
Sameh A. Kantoush
Water Resources Research Center, Disaster Prevention Research Institute, Kyoto University, Goka-sho, Uji City, Kyoto, Japan
Cao Duong Phan
Ireland's Centre for AI, School of Computer Science, University of Dublin, Belfield, Dublin 4, Ireland
Thi Phuong Mai Nguyen
Department of Civil Engineering, Thuyloi University, 175 Tay Son, Dong Da, Hanoi, Vietnam
Doan Van Binh
Faculty of Engineering, Vietnamese-German University, Ben Cat City, Binh Duong Province, Vietnam

Tóm tắt

Salinity intrusion poses a threat to water security, agriculture, and livelihoods in the Vietnamese Mekong Delta (VMD), particularly under the combined pressures of climate change and upstream hydrological developments. Accurate short- to mid-term salinity forecasts are essential for proactive water resource management. This study evaluates the performance of two machine learning models—Random Forest Regression (RFR) and Support Vector Regression (SVR)—for salinity forecasting using long-term observational data (1996–2023) from 44 monitoring stations in the VMD. To capture temporal dynamics, multitemporal lag features (1, 10, 20, 30, and 60 days) were generated from observed salinity records. Bayesian optimization and time-series cross-validation were used for model tuning. Results show that SVR performs best for short-term forecasts (1–3 days), achieving R2 and NSE up to 0.927–0.928, MAE ≈ 0.824 g/L, and RMSE ≈ 1.858 g/L, while RFR provides more stable predictions over longer horizons (4–7 days), maintaining R2/NSE values of 0.627 to 0.766 with lower errors. Additionally, the 20-day lag windows yielded the most accurate results, likely reflecting the influence of tidal cycles. These findings highlight the importance of selecting appropriate models and temporal features for various forecast horizons, providing a data-driven framework to enhance early warning systems and support adaptive water resource management in the VMD.

https://doi.org/10.26459/hueunijns.v134i1S-1.7876
PDF (English)

Tài liệu tham khảo

  1. De Costa GS, Kojiri T, Porter M. Salinity intrusion: its characteristics and impact - cases in the Asia Pacific region. In: Lee JHW, Lam KM, editors. 4th International Symposium on Environmental Hydraulics. Hong Kong, China; 2005. p. 2027-32.
  2. Mahmuduzzaman M, Uddin Z, Nuruzzaman A, Rabbi F, Ahmed S. Causes of Salinity Intrusion in Coastal Belt of Bangladesh. International Journal of Plant Research. 2014;2014:8-13.
  3. Sherin VR, Durand F, Papa F, Islam AKMS, Gopalakrishna VV, Khaki M, et al. Recent salinity intrusion in the Bengal delta: Observations and possible causes. Continental Shelf Research. 2020;202:104142.
  4. Biswas KK, Sarder BC, Saika U, Hoque MA-A. Environmental and Socio-economic Impacts of Salinity Intrusion in the Coastal Area: A Case Study on Munshigong Union, Shymnagor, Satkhira. Jahangirnagar University Environmental Bulletin. 2013;2:41-9.
  5. Dung TD. Drought and salinity intrusion in the Mekong Delta: What are the fundamental solutions? Health & Agricultural Policy Research Institute. 2024.
  6. Minderhoud P, Coumou L, Erban L, Middelkoop H, Stouthamer E, Addink E. The relation between land use and subsidence in the Vietnamese Mekong Delta. Science of the Total Environment. 2018;634:715-26.
  7. Qiu C, Zhu J-R. Influence of seasonal runoff regulation by the Three Gorges Reservoir on saltwater intrusion in the Changjiang River Estuary. Continental Shelf Research. 2013;71:16-26.
  8. Andrews SW, Gross ES, Hutton PH. Modeling salt intrusion in the San Francisco Estuary prior to anthropogenic influence. Continental Shelf Research. 2017;146:58-81.
  9. Gong W, Lin Z, Zhang H, Lin H. The response of salt intrusion to changes in river discharge, tidal range, and winds, based on wavelet analysis in the Modaomen estuary, China. Ocean & Coastal Management. 2022;219.
  10. Ahmed AA, Sayed S, Abdoulhalik A, Moutari S, Oyedele L. Applications of machine learning to water resources management: A review of present status and future opportunities. Journal of Cleaner Production. 2024;441:140715.
  11. Tran TT, Pham NH, Pham QB, Pham TL, Ngo XQ, Nguyen DL, et al. Performances of Different Machine Learning Algorithms for Predicting Saltwater Intrusion in the Vietnamese Mekong Delta Using Limited Input Data: A Study from Ham Luong River. Water Resources. 2022;49(3):391-401.
  12. Dang LTT, Ishidaira H, Nguyen KP, Souma K, Magome J. Short-term salinity prediction for coastal areas of the Vietnamese Mekong Delta using various machine learning algorithms: a case study in Soc Trang Province. Applied Water Science. 2025;15(4):79.
  13. Hoai PN, Quoc PB, Thanh Thai T. Apply Machine Learning to Predict Saltwater Intrusion in the Ham Luong River, Ben Tre Province. VNU Journal of Science: Earth and Environmental Sciences. 2022;38(3).
  14. Hai T, Vu Van N, Hùng H, Tuan N, Dang T, Lam, et al. Assessing and Forecasting Saline Intrusion in the Vietnamese Mekong Delta Under the Impact of Upstream flow and Sea Level Rise. Journal of Environmental Science and Engineering B. 2019;8:174-85.
  15. Cgiar Research Program on Climate Change A, Food Security - Southeast A. Assessment Report: The drought and salinity intrusion in the Mekong River Delta of Vietnam. Hanoi: CGIAR; 2016.
  16. Loc HH, Low Lixian M, Park E, Dung TD, Shrestha S, Yoon Y-J. How the saline water intrusion has reshaped the agricultural landscape of the Vietnamese Mekong Delta, a review. Science of The Total Environment. 2021;794:148651.
  17. Breiman L. Random forests. Machine learning. 2001;45:5-32.
  18. Geurts P, Ernst D, Wehenkel L. Extremely randomized trees. Machine learning. 2006;63:3-42.
  19. Liaw A, Wiener M. Classification and regression by randomForest. R news. 2002;2(3):18-22.
  20. Wu J, Chen X-Y, Zhang H, Xiong L-D, Lei H, Deng S-H. Hyperparameter Optimization for Machine Learning Models Based on Bayesian Optimizationb. Journal of Electronic Science and Technology. 2019;17(1):26-40.
  21. Hyndman RJ, Athanasopoulos G. Forecasting: principles and practice: OTexts; 2018.
  22. Drucker H, Burges CJ, Kaufman L, Smola A, Vapnik V. Support vector regression machines. Advances in neural information processing systems. 1996;9.
  23. Vapnik V. The nature of statistical learning theory: Springer science & business media; 1999.
  24. Smola AJ, Schölkopf B. A tutorial on support vector regression. Statistics and computing. 2004;14:199-222.
  25. Ramedani Z, Omid M, Keyhani A, Shamshirband S, Khoshnevisan B. Potential of radial basis function based support vector regression for global solar radiation prediction. Renewable and Sustainable Energy Reviews. 2014;39:1005-11.
  26. Li M, Liu Y-H, editors. Learning interaction force model for endodontic shaping with support vector regression. Proceedings 2006 IEEE International Conference on Robotics and Automation, 2006 ICRA 2006; 2006: IEEE.
  27. Yu P-S, Chen S-T, Chang IF. Support vector regression for real-time flood stage forecasting. Journal of Hydrology. 2006;328(3):704-16.
  28. Guillou N, Chapalain G, Petton S. Predicting sea surface salinity in a tidal estuary with machine learning. Oceanologia. 2023;65(2):318-32.
  29. Nash JE, Sutcliffe JV. River flow forecasting through conceptual models part I—A discussion of principles. Journal of hydrology. 1970;10(3):282-90.
  30. Zheng R, Sun Z, Jiao J, Ma Q, Zhao L. Salinity Prediction Based on Improved LSTM Model in the Qiantang Estuary, China. Journal of Marine Science and Engineering. 2024;12(8).
  31. Moges DM, Virro H, Kmoch A, Cibin R, Rohith RAN, Martínez-Salvador A, et al. Streamflow Prediction with Time-Lag-Informed Random Forest and Its Performance Compared to SWAT in Diverse Catchments. Water. 2024;16(19).
  32. Ma K, He D, Liu S, Ji X, Li Y, Jiang H. Novel time-lag informed deep learning framework for enhanced streamflow prediction and flood early warning in large-scale catchments. Journal of Hydrology. 2024;631.
  33. Chen Y-C, Yeh H-C, Kao S-P, Wei C, Su P-Y. Water Level Forecasting in Tidal Rivers during Typhoon Periods through Ensemble Empirical Mode Decomposition. Hydrology. 2023;10(2).
  34. Bargam B, Boudhar A, Kinnard C, Bouamri H, Nifa K, Chehbouni A. Evaluation of the support vector regression (SVR) and the random forest (RF) models accuracy for streamflow prediction under a data-scarce basin in Morocco. Discover Applied Sciences. 2024;6(6).
  35. Sharma B, Goel NK. Streamflow prediction using support vector regression machine learning model for Tehri Dam. Applied Water Science. 2024;14(5).
  36. Abbes A, Inoubli R, Rhif M, Farah I. Combining deep learning methods and multi-resolution analysis for drought forecasting modeling. Earth Science Informatics. 2023;16:1-10.
Creative Commons License

công trình này được cấp phép theo Creative Commons Ghi công-Chia sẻ tương tự 4.0 License International .

Bản quyền (c) 2025 Array