Groundwater Level Simulation Using Soft Computing Methods with Emphasis on Major Meteorological Components

• Published in: Water resources management 2022-08, Vol.36 (10), p.3627-3647
• Main Authors: Samani, Saeideh, Vadiati, Meysam, Azizi, Farahnaz, Zamani, Efat, Kisi, Ozgur
• Format: Article
• Language: English
• Subjects: Adaptive systems; Aquifer characteristics; Aquifers; Artificial neural networks; Atmospheric models; Atmospheric Sciences; Civil Engineering; Components; Earth and Environmental Science; Earth Sciences; Environment; Evapotranspiration; Fuzzy logic; Geology; Geotechnical Engineering & Applied Earth Sciences; Groundwater; Groundwater levels; Group method of data handling; Hydrogeology; Hydrology/Water Resources; Mathematical models; Methods; Neural networks; Numerical models; Predictions; Soft computing; Software; Support vector machines; Sustainable development; Sustainable Development Goals; Unconfined aquifers; Water resources; Water resources management
• ISSN: 0920-4741; 1573-1650
• DOI: 10.1007/s11269-022-03217-x

Precise estimation of groundwater level (GWL) might be of great importance for attaining sustainable development goals and integrated water resources management. Compared with alternative numerical models, soft computing methods are promising tools for GWL prediction, which need more hydrogeological and aquifer characteristics. The central aim of this research is to explore the performance of such well-accepted data-driven models to predict monthly GWL with emphasis on major meteorological components, including; precipitation (P), temperature (T), and evapotranspiration (ET). Artificial neural network (ANN), fuzzy logic (FL), adaptive neuro-fuzzy inference system (ANFIS), group method of data handling (GMDH), and least-square support vector machine (LSSVM) are used to predict one-, two-, and three-month ahead GWL in an unconfined aquifer. The main meteorological components (T t , ET t , P t , P t-1 ) and GWL for one, two, and three lag-time (GWL t-1 , GWL t-2 , GWL t-3 ) are used as input to attain precise prediction. The results show that all models could have the best prediction for one month ahead in scenario 5, comprising inputs of GWL t-1 , GWL t-2 , GWL t-3 , T t , ET t , P t , T t-1 , ET t-1 , P t-1 . Based on different evaluation criteria, all employed models could predict the GWL with a desirable accuracy, and the results of LSSVM are the superior one.