| Abstrakt: |
Determining the controlling factors of the quality of groundwater can help decision-makers in preparing optimal management plans and a sustainable supply of water for drinking and agriculture. The results of factor analysis and hierarchical cluster analysis show that the chemical composition of groundwater in the Fereidan region (Iran) is controlled by both geogenic and anthropogenic factors. The natural background level of major variables and heavy metals including Cu, Zn, Hg, Cd, Pb, and As have been estimated using probabilistic and deterministic methods including the iterative 2σ outlier test, calculated distribution function (CDF), 95th percentile, and iterative Grubbs test (IGT). The results show that the pre-selection process and removal of outliers play a significant role in calculating NBLs and threshold values (TVs). Although heavy metals had different datasets and it was impossible to perform the pre-selection process for them, nevertheless, CDF, iterative 2σ, and 95th percentile methods were also used to calculate their NBLs. As the results show, Δ% between NBLs calculated by CDF, iterative 2σ, 95th percentile methods, and IGT is greater than that for complete datasets. This is because contaminated samples were not removed by the pre-selection process and some of these data remained in the final dataset. This shows itself as a skewed dataset of some heavy elements including Cu, Zn, and Cd. Therefore, if the goal is strict planning to take away the freedom of farmers and local planners to pollute the aquifer by providing smaller NBLs and TVs, it is better to use the smaller NBLs and TV determined by the IGT method after the pre-selection process. [ABSTRACT FROM AUTHOR] |
