Classification of surface water quality using data-driven methods

Abstract

Access to clean and quality water resources has been one of the main concerns of human beings for a long time. Therefore, determining the quality of water for various uses, including irrigation is very important. River pollution is one of the most important problems in the world today, especially in developing countries. In the present study, using data related to water quality parameters of Bagh Kalayeh hydrometric station in the 23-year statistical period, first the WQI index was calculated, then using data mining technique, factors affecting water quality were determined. Finally, the results of data mining methods were compared with the results obtained from the qualitative index. Quantitative results of the models were evaluated by Correlation Coefficient (R), Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and qualitative results of the models were evaluated by Kappa, RMSE and MAE statistics. The results showed that in quantitative modeling, scenario 5 including TH, K, SO4, TDS and EC and in qualitative modeling, scenario 3 including TH, K and SO4 were selected as the superior scenario.