Assessing Pollution with Heavy Metals and Its Impact on Population Health.

Abstract

Pollution is one of the most important issues currently affecting the global population and environment. Therefore, determining the zones where stringent measures should be taken is necessary. In this study, Principal Component Analysis (PCA), Factor Analysis (FA), and t-distributed Stochastic Neighbor Embedding (t-SNE) were utilized for dimensionality reduction and clustering of data series containing the concentration of 10 heavy metals collected at 14 locations. The Hazard Quotient (HQ) and Hazard Index (HI) were utilized to determine the non-carcinogenic risk to the population in the studied zones. The highest concentrations of metals in the samples were those of Fe, Zn, Mn, and Cr. PCA indicated that Fe and Zn (Co and Cd) had the highest contribution on the first (second) Principal Component (PC). FA showed that the three-factor model is adequate for explaining the variability of pollutant concentrations. The factor loadings revealed the strength of association between variables and factors, e.g., 0.97 for Zn, 0.83 for Cr, and 0.99 for Co. HQ for ingestion, HQing, was the highest for Fe (between 6.10 × 10^-5 and 2.57 × 10^-4). HQ for inhalation, HQinh, was the biggest for Mn (from 1.41 × 10^-3 to 1.95 × 10^-3). HI varied in the interval [0.172, 0.573], indicating the absence of a non-carcinogenic risk. However, since values above 0.5 were determined at four sites, continuous monitoring of the pollution in the sampling locations is necessary.