A machine learning approach for prioritizing groundwater testing for per-and polyfluoroalkyl substances (PFAS).

Regulatory agencies are beginning to recognize and regulate per-and polyfluoroalkyl substances (PFAS) as concerning environmental contaminants. In groundwater management, testing and mitigation strategies are desirable, but can be time and cost-intensive processes. As a result, only a fraction of all groundwater wells has been tested for PFAS levels, resulting in potentially extended drinking water exposure to PFAS in the meantime. In this study, we build a series of machine learning models (including linear and random forest regressors) to predict PFAS based on a groundwater dataset from California. These models are used to compare the relative predictive ability of co-contaminant fingerprints, hydrological properties, soil parameters, proximity of airports/military bases, and geospatial data. Additionally, a random forest machine learning model that combines all data types can quantitatively predict the maximum PFAS compound concentration in a well with a Spearman correlation of 0.64 and can discern wells containing concerningly high concentrations of PFAS with an accuracy of 91 % (AUC of 0.90). This approach may have widespread utility for other hazardous anthropogenic compounds in groundwater. Future investigations should evaluate the practicability of using machine learning to prospectively prioritize contaminant testing in groundwater wells.

Phylogenetic diversity in freshwater-dwelling Isochrysidales haptophytes with implications for alkenone production.

Members of the order Isochrysidales are unique among haptophyte lineages in being the exclusive producers of alkenones, long-chain ketones that are commonly used for paleotemperature reconstructions. Alkenone-producing haptophytes are divided into three major groups based largely on molecular ecological data: Group I is found in freshwater lakes, Group II commonly occurs in brackish and coastal marine environments, and Group III consists of open ocean species. Each group has distinct alkenone distributions; however, only Groups II and III Isochrysidales currently have cultured representatives. The uncultured Group I Isochrysidales are distinguished geochemically by the presence of tri-unsaturated alkenone isomers (C37:3b Me, C38:3b Et, C38:3b Me, C39:3b Et) present in water column and sediment samples, yet their genetic diversity, morphology, and environmental controls are largely unknown. Using small-subunit (SSU) ribosomal RNA (rRNA) marker gene amplicon high-throughput sequencing of environmental water column and sediment samples, we show that Group I is monophyletic with high phylogenetic diversity and contains a well-supported clade separating the previously described "EV" clade from the "Greenland" clade. We infer the first partial large-subunit (LSU) rRNA gene Group I sequence phylogeny, which uncovered additional well-supported clades embedded within Group I. Relative to Group II, Group I revealed higher levels of genetic diversity despite conservation of alkenone signatures and a closer evolutionary relationship with Group III. In Group I, the presence of the tri-unsaturated alkenone isomers appears to be conserved, which is not the case for Group II. This suggests differing environmental influences on Group I and II and perhaps uncovers evolutionary constraints on alkenone biosynthesis.