Fall Creek Monitoring Station: Using Environmental Covariates To Predict Micropollutant Dynamics and Peak Events in Surface Water Systems.

This research aimed to further our understanding of how environmental processes control micropollutant dynamics in surface water systems as a means to predict peak concentration events and inform intermittent sampling strategies. We characterized micropollutant concentrations in daily composite samples from the Fall Creek Monitoring Station over 18 months. These data were compiled alongside environmental covariates, including daily measurements of weather, hydrology, and water quality parameters, to generate a novel data set with high temporal resolution. We evaluated the temporal trends of several representative micropollutants, along with cumulative metrics of overall micropollutant contamination, by means of multivariable analyses to determine which combination of covariates best predicts micropollutant dynamics and peak events. Peak events of agriculture-derived micropollutants were best predicted by positive associations with turbidity and UV254 absorbance and negative associations with baseflow index. Peak events of wastewater-derived micropollutants were best predicted by positive associations with alkalinity and negative associations with streamflow rate. We demonstrate that these predictors can be used to inform intermittent sampling strategies aimed at capturing peak events, and we generalize the approach so that it could be applied in other watersheds. Finally, we demonstrate how our approach can be used to contextualize micropollutant data derived from infrequent grab samples.

Fall Creek Monitoring Station: Highly Resolved Temporal Sampling to Prioritize the Identification of Nontarget Micropollutants in a Small Stream.

The goal of this research was to comprehensively characterize the occurrence and temporal dynamics of target and nontarget micropollutants in a small stream. We established the Fall Creek Monitoring Station in March 2017 and collected daily composite samples for one year. We measured water samples by means of high-resolution mass spectrometry and developed and optimized a postacquisition data processing workflow to screen for 162 target micropollutants and group all mass spectral (MS) features into temporal profiles. We used hierarchical clustering analysis to prioritize nontarget MS features based their similarity to target micropollutant profiles and developed a high-throughput pipeline to elucidate the structures of prioritized nontarget MS features. Our analyses resulted in the identification of 31 target micropollutants and 59 nontarget micropollutants with varying levels of confidence. Temporal profiles of the 90 identified micropollutants revealed unexpected concentration-discharge relationships that depended on the source of the micropollutant and hydrological features of the watershed. Several of the nontarget micropollutants have not been previously reported including pharmaceutical metabolites, rubber vulcanization accelerators, plasticizers, and flame retardants. Our data provide novel insights on the temporal dynamics of micropollutant occurrence in small streams. Further, our approach to nontarget analysis is general and not restricted to highly resolved temporal data acquisitions or samples collected from surface water systems.