Estimation of groundwater quality using an integration of water quality index, artificial intelligence methods and GIS: Case study, Central Mediterranean Region of Turkey

Creative Commons License

Taşan S.

APPLIED WATER SCIENCE, cilt.13, sa.1, 2023 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 13 Sayı: 1
  • Basım Tarihi: 2023
  • Doi Numarası: 10.1007/s13201-022-01810-4
  • Dergi Adı: APPLIED WATER SCIENCE
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, IBZ Online, CAB Abstracts, Compendex, Environment Index, Food Science & Technology Abstracts, Geobase, INSPEC, Veterinary Science Database, Directory of Open Access Journals
  • Anahtar Kelimeler: Artificial neural networks, Adaptive network-based fuzzy inference system, Irrigation suitability, Principal component analysis, Spatial distribution, ANFIS MODELS, IRRIGATION, PREDICTION, NETWORKS, AREA, ANN
  • Ondokuz Mayıs Üniversitesi Adresli: Evet

Özet

Groundwater is one of the most important natural resources in the world and is widely used for irrigation purposes. Groundwater quality is affected by various natural heterogeneities and anthropogenic activities. Consequently, monitoring groundwater quality and assessing its suitability are crucial for sustainable agricultural irrigation. In this study, the suitability of groundwater for irrigation was determined by using sodium adsorption ratio (SAR), residual sodium carbonate (RSC), Kelly index (KI), percentage of sodium (Na%), magnesium ratio (MR), potential salinity (PS) and permeability index (PI). The groundwater samples were collected and analyzed from 37 different sampling stations for this purpose. Along with suitability analysis, artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models were used to predict irrigation water quality parameters. The models were evaluated by comparing the measured values and the predicted values using the statistical criteria [coefficient of determination (R-2), mean absolute error (MAE), root mean square error (RMSE) and Nash-Sutcliffe efficiency (NS)]. In the estimation of all irrigation water quality parameters, the ANN model has performed much higher compared with the ANFIS model. Spatial distribution maps were generated for measured and ANN model-estimated irrigation water quality indices using the IDW interpolation method. Spatial distributions of groundwater quality indices revealed that MR was higher than the allowable limits in most of the study areas and the other quality criteria were within the permissible limits. It has been determined that the interpolation maps obtained as a result of artificial intelligence methods have appropriate sensitivity when compared with the observed maps. Based on the present findings, ANN models could be used as an efficient tool for estimating groundwater quality indices in unsampled sections of the study area and the other regions with similar conditions.