Please login first
Evaluation of the capability of NARX neural network in predicting ground water level changes
* , , , , ,
1  Faculty of Civil Engineering, University of Tabriz, Tabriz, Iran
Academic Editor: Simeone Chianese


Efficient monitoring and tracking of groundwater level changes is critical for sustainable management of water resources, especially in light of population growth, climate change, and increasing water demand. This study evaluates the ability of the Non-linear AutoRegressive with eXogenous input (NARX) model to simulate groundwater level trends in Ajabshir, Iran, using groundwater level data from 2006 to 2019 as a baseline period. The model was trained using time, groundwater levels, and delay times between 1 to 2 as the input training samples. The results indicate that the NARX model performed exceptionally well in simulating historical trends of groundwater levels, achieving a Coefficient of Determination (R2) value of 0.87 and a Root Mean Squared Error (RMSE) of 0.03 (m). The excellent performance can be attributed to the optimal hyperparameters and long-term simulation capabilities of the NARX model. The findings have significant implications for managing groundwater resources in Ajabshir and other regions facing similar challenges. The NARX model can be used to predict future trends in groundwater levels, taking into account current and projected climatic conditions, population growth, and other key factors. Such predictions can inform decision-making and help develop effective water management strategies that promote sustainable use of this vital resource.

Keywords: Groundwater levels; Non-linear AutoRegressive with eXogenous input (NARX); Simulation; Sustainable water management; Trend analysis