Monthly Groundwater Level Prediction in the Nahavand Aquifer Using Deep Learning and Support Vector Machine (SVM) Models

Abdi, Somaye; Fathian, Hossein; Asadi Lour, Mehdi; Igdernejad, Aslan; Asareh, Ali

Journal of Wetland Ecobiology

EN FA

Volume 18, Issue 4 (3-2026) 2026, 18(4): 37-54 | Back to browse issues page

Mendeley

Zotero

RefWorks

Abdi S, Fathian H, Asadi Lour M, Igdernejad A, Asareh A. Monthly Groundwater Level Prediction in the Nahavand Aquifer Using Deep Learning and Support Vector Machine (SVM) Models. Wetland Ecobiology 2026; 18 (4) : 4
URL: http://jweb.ahvaz.iau.ir/article-1-1121-en.html

Monthly Groundwater Level Prediction in the Nahavand Aquifer Using Deep Learning and Support Vector Machine (SVM) Models

Somaye Abdi

, Hossein Fathian

, Mehdi Asadi Lour

, Aslan Igdernejad

, Ali Asareh

Abstract: (15 Views)

Decline in rainfall and excessive exploitation of groundwater resources have led to a significant drop in groundwater levels in many regions of the world. One of the aquifers facing groundwater depletion is the Nahavand aquifer, located in Hamadan Province in western Iran. To model groundwater level fluctuations, Long Short-Term Memory (LSTM) and Support Vector Machine (SVM) models were employed. Sensitivity analysis of groundwater level prediction for the following month, with respect to variations in the input variables of the two models, indicated that precipitation, air temperature, evapotranspiration, and groundwater levels up to two months prior are the most influential input variables. Modeling results using the SVM model showed that the radial basis function (RBF) kernel outperformed linear and polynomial kernels. Comparison of the intelligent models demonstrated that the SVM model achieved better performance than the LSTM model. The results revealed that, for the test period, the LSTM model yielded R² = 0.87 and RMSE = 0.043, while the SVM model achieved R² = 0.89 and RMSE = 0.042. Overall, it can be concluded that the SVM model is highly suitable for predicting groundwater levels in the Nahavand aquifer and has the potential to be applied to other aquifers as well.

Article number: 4

Keywords: Prediction, Groundwater level, Nahavand aqufer, LSTM, SVM.

Full-Text [PDF 1213 kb] (7 Downloads)

Type of Study: Research | Subject: Special
Received: 2025/12/30 | Accepted: 2026/06/26 | Published: 2026/06/26

Send email to the article author

Rights and permissions
	This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.