Employing complementary multivariate methods for a designed nontarget LC-HRMS screening of a wastewater-influenced river

Abstract

Environmental pollution issues, such as the impact of wastewater treatment plants (WWTPs) to contaminate the recipient waters, are multivariate problems in which several variables with varying correlation values contribute. Chemometrics-based methods can play a critical role to study the occurrence and fate of contaminants of emerging concern (CECs) in an environmental pollution question. This study performed nontarget screening (NTS) with liquid chromatography-high resolution mass spectrometry (LC-HRMS) for the investigation of pollution patterns of river water samples affected by the effluent of a WWTP (upstream/downstream) at three sampling times. The complex data sets collected in a designed experiment were analyzed using the ROIMCR approach, using data selection from MS regions of interest (ROI). Then, multivariate curve resolution alternating least-squares (MCR-ALS) was applied for simultaneous resolution of 18 upstream/downstream river water samples. The resolved and cleaned matrix of peak areas were subjected to group-wise ANOVA-simultaneous component analysis (GASCA) for interpretation of the variations induced by location and time factors in a sparse principal component analysis way and to uncover the groups of pollutants co-occurring in the environment. The resolved groups of significant features based on sampling site factor were compared with the prioritized features using partial least squares-discriminant analysis (PLS-DA) least squares-discriminant analysis ures using partial least-squares-discriminant analysis (PLS-DA). Here, we focused on the environmental relevance of groups of compounds with significant variations for a few tentatively identified contaminants. This study clearly shows that employing complementary chemometric tools for NTS can improve the current knowledge on exposure trends of CECs in water bodies for further investigation.