Determining the Minimum Suitable Number of Water Quality Indicators to Improve the EWQI Water Quality Assessment Model in Baojixia Irrigation District, Northwest China

Selecting a suitable number of parameters for water quality assessment can make the assessment cost-effective. The current investigation involved the collection of 64 groundwater samples from the Baojixia irrigation district located in China. These samples were then analyzed for 17 water quality parameters. Two minimum entropy water quality index (EWQI min ) models were proposed by selecting the key parameters from the analyzed water quality parameters through principal component analysis (PCA) and multiple linear regression analysis (MLR), correspondingly. Then, the two proposed EWQI min models were utilized to evaluate the water quality within the study area. The findings revealed that the EWQI min−MLR model, which comprised 5 key parameters (total dissolved solids (TDS), sodium (Na + ), nitrate (NO 3 − ), total hardness (TH), and fluorine (F − ), exhibited better performance in groundwater quality evaluation. This model demonstrated a higher coefficient of determination (R 2 = 0.953, P < 0.001), coupled with lower values of Root Mean Square Error (RMSE, 4.948) and Percentage Error (PE, 5.823%) when compared to the EWQI min−PCA model consisting of 6 key parameters including TDS, Na + , TH, chloride (Cl − ), nitrite (NO 2 − ) and chemical oxygen demand (COD Mn ). Furthermore, the groundwater quality in the Baojixia irrigation district was considered a moderate quality category, with the eastern region displaying poorer water quality in comparison to the western area. The comparison of EWQI min and EWQI indicated that the developed EWQI min model was a suitable and effective method as its performance in evaluating groundwater quality within the Baojixia irrigation district is excellent. The results of this research have significant implications for the effective management of groundwater and the promotion of sustainable development of water resources in future investigations.