IJESD 2025 Vol.16(3): 202-213
doi: 10.18178/ijesd.2025.16.3.1527

Application of Machine Learning Models to Predict Groundwater Level Change, a Case Study of Vung Tau City, Ba Ria-Vung Tau Province

Trinh Tuyet Vu1, Au Hai Nguyen1,2*, Nhi Tuyet Thi Pham1, Vy Minh Hong Tat1, and Linh Khanh Luu1
1The Institute for Environment and Resources, 142 To Hien Thanh Street, District 10, Ho Chi Minh City, Vietnam
2Vietnam National University Ho Chi Minh City, Linh Trung Ward, Thu Duc District, Ho Chi Minh City, Vietnam
Email: vutuyettrinh82@gmail.com (T.T.V.); haiauvtn@gmail.com (A.H.N.); tuyetnhi.ier@gmail.com (N.T.T.P.); tathongminhvy271@gmail.com (V.M.H.T.); luukhanhlinh1998@gmail.com (L.K.L.)
*Corresponding author
Manuscript received November 11, 2024; revised February 8, 2025; accepted March 24, 2025; published June 6, 2025

Abstract—Groundwater Level (GWL) control is crucial for sustainable water resource management, agricultural planning, and reducing environmental impacts. GWL forecasting is crucial for optimizing water usage, preventing over-extraction, addressing land subsidence, supporting ecological balance, and making informed decisions for economic activities. However, GWL variability is driven by multiple factors, including precipitation, soil properties, seasonal fluctuations, and human activities, which complicate prediction efforts. To address these challenges, advanced machine learning models, such as Long Short-Term Memory (LSTM) networks and Gated Recurrent Units (GRU), have proven effective in capturing temporal dependencies and handling non-linear relationships in time series data. These models are ideal for identifying patterns over extended periods and managing complex interactions influencing GWL. In this study, we utilize LSTM and GRU models to predict GWL, developing four sub-models for each: an hourly, daily, 10-day, and monthly model, all aimed at forecasting GWL one month in advance. The models’ performance is evaluated using statistical metrics such as Mean Square Error (MSE), Mean Absolute Error (MAE), and Rsquared. The daily LSTM and GRU models achieved R-squared values of 69.87% and 71.34%, respectively. Their exceptional predictive capabilities in groundwater forecasting provide valuable insights and contribute significantly to the field of water resource management.

Keywords—groundwater level, LSTM, GRU, machine learning, deep learning

[PDF]

Cite: Trinh Tuyet Vu, Au Hai Nguyen, Nhi Tuyet Thi Pham, Vy Minh Hong Tat, and Linh Khanh Luu, "Application of Machine Learning Models to Predict Groundwater Level Change, a Case Study of Vung Tau City, Ba Ria-Vung Tau Province," International Journal of Environmental Science and Development vol. 16, no. 3, pp. 202-213, 2025.

Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).

Article Metrics in Dimensions