Application of Nontarget High Resolution Mass Spectrometry Data to Quantitative Source Apportionment

Abstract

High resolution mass spectrometry (HRMS) analyses provide expansive chemical characterizations of environmental samples. To date, most research efforts have developed tools to expedite labor- and time-intensive contaminant identification efforts. However, even without chemical identity, the richness of nontarget HRMS data sets represents a significant opportunity to chemically differentiate samples and delineate source contributions. To develop this potential, we evaluated the use of unidentified HRMS detections to define sample uniqueness and provide additional statistical resolution for quantitative source apportionment, overcoming a critical limitation of existing approaches based on targeted contaminants. By creating a series of sample mixtures that mimic pollution sources in a representative watershed, we assessed the fidelity of HRMS source fingerprints during dilution and mixing. This approach isolated 8-447 nontarget compounds per sample for source apportionment and yielded accurate source concentration estimates (between 0.82 and 1.4-fold of actual values), even in multisource systems with <1% source contributions. Furthermore, we mined the nontarget data to identify five source-specific chemical end-members amenable to apportionment. While additional development studies are needed to fully evaluate the myriad factors affecting method accuracy and capabilities, this study provides a conceptual foundation for novel applications of nontarget HRMS data to confidently distinguish and quantify source impacts in complex systems.