Effectively managing environmental quality to protect ecosystem and human health requires accurate delineation of historical and ongoing pollutant source contributions to impaired systems. Existing forensic approaches rely on detections of a few legacy contaminants, such as PAHs, to identify original sources (e.g., vehicle exhaust, oil spills). However, these approaches cannot deconvolute contaminant pathways that transport mixtures of original sources (e.g., urban stormwater, eroding contaminated bank sediments), and are limited when similar legacy pollutant profiles confound distinctions between multiple contaminant inputs. Here, we developed a method using non-target high-resolution mass spectrometry (HRMS) to identify, deconvolute, and quantitatively apportion contaminant sources. We created a laboratory-scale model watershed by diluting two road runoffs (i.e., mixed-source contaminant inputs) into surface waters at runoff concentrations ranging from 0.16-30% by volume. After data reduction, we tracked the unique, non-target signature of each road runoff source into the final mixtures, where the peak area (abundance) of each non-target compound corresponded to source concentration. Using the statistical power of the ~10-300 source concentration estimates, we accurately quantified system hydraulics (within a factor of 0.75-1.44) without the need to identify any specific chemicals. The approach also supported identification of source “marker” compounds; we mined the data to identify three road runoff indicators (at a confidence level &gt;S2) that were detectable at source concentrations &lt;1% and whose peak area correlated with source dilution. These data represent the first report of HRMS data applied to quantify the relative contributions of pollutant sources to complex mixtures.

Application of non-target high resolution mass spectrometry for quantitative source apportionment

2019 Spring National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.