Michael J. Focazio

A national reconnaissance for pharmaceuticals and other organic wastewater contaminants in the United States - II) Untreated drinking water sources

Numerous studies have shown that a variety of manufactured and natural organic compounds such as pharmaceuticals, steroids, surfactants, flame retardants, fragrances, plasticizers and other chemicals often associated with wastewaters have been detected in the vicinity of municipal wastewater discharges and livestock agricultural facilities. To provide new data and insights about the environmental presence of some of these chemicals in untreated sources of drinking water in the United States targeted sites were sampled and analyzed for 100 analytes with sub-parts per billion detection capabilities. The sites included 25 ground- and 49 surface-water sources of drinking water serving populations ranging from one family to over 8 million people. Sixty-three of the 100 targeted chemicals were detected in at least one water sample. Interestingly, in spite of the low detection levels 60% of the 36 pharmaceuticals (including prescription drugs and antibiotics) analyzed were not detected in any water sample. The five most frequently detected chemicals targeted in surface water were: cholesterol (59%, natural sterol), metolachlor (53%, herbicide), cotinine (51%, nicotine metabolite), beta-sitosterol (37%, natural plant sterol), and 1,7-dimethylxanthine (27%, caffeine metabolite); and in ground water: tetrachloroethylene (24%, solvent), carbamazepine (20%, pharmaceutical), bisphenol-A (20%, plasticizer), 1,7-dimethylxanthine (16%, caffeine metabolite), and tri (2-chloroethyl) phosphate (12%, fire retardant). A median of 4 compounds were detected per site indicating that the targeted chemicals generally occur in mixtures (commonly near detection levels) in the environment and likely originate from a variety of animal and human uses and waste sources. These data will help prioritize and determine the need, if any, for future occurrence, fate and transport, and health-effects research for subsets of these chemicals and their degradates most likely to be found in water resources used for drinking water in the United States.

Discharges of produced waters from oil and gas extraction via wastewater treatment plants are sources of disinfection by-products to receiving streams

Fluids co-produced with oil and gas production (produced waters) are often brines that contain elevated concentrations of bromide. Bromide is an important precursor of several toxic disinfection by-products (DBPs) and the treatment of produced water may lead to more brominated DBPs. To determine if wastewater treatment plants that accept produced waters discharge greater amounts of brominated DBPs, water samples were collected in Pennsylvania from four sites along a large river including an upstream site, a site below a publicly owned wastewater treatment plant (POTW) outfall (does not accept produced water), a site below an oil and gas commercial wastewater treatment plant (CWT) outfall, and downstream of the POTW and CWT. Of 29 DBPs analyzed, the site at the POTW outfall had the highest number detected (six) ranging in concentration from 0.01 to 0.09 μg L(-1) with a similar mixture of DBPs that have been detected at POTW outfalls elsewhere in the United States. The DBP profile at the CWT outfall was much different, although only two DBPs, dibromochloronitromethane (DBCNM) and chloroform, were detected, DBCNM was found at relatively high concentrations (up to 8.5 μg L(-1)). The water at the CWT outfall also had a mixture of inorganic and organic precursors including elevated concentrations of bromide (75 mg L(-1)) and other organic DBP precursors (phenol at 15 μg L(-1)). To corroborate these DBP results, samples were collected in Pennsylvania from additional POTW and CWT outfalls that accept produced waters. The additional CWT also had high concentrations of DBCNM (3.1 μg L(-1)) while the POTWs that accept produced waters had elevated numbers (up to 15) and concentrations of DBPs, especially brominated and iodinated THMs (up to 12 μg L(-1) total THM concentration). Therefore, produced water brines that have been disinfected are potential sources of DBPs along with DBP precursors to streams wherever these wastewaters are discharged.

Arsenic in Ground Water Supplies of the United States

Publisher Summary High arsenic (As) concentrations in ground water have been documented in many areas of the United States. Within the last decade, parts of Maine, Michigan, Minnesota, South Dakota, Oklahoma, and Wisconsin have been found to have widespread As concentrations exceeding 10 μg/L. These high concentrations most commonly result from the upflow of geothermal water, dissolution of, or desorption from, iron-oxide, and dissolution of sulfide minerals. Because the MCL for As is currently being evaluated, estimating the exceedance frequency for different As concentrations in regulated water supplies is particularly timely. The estimates of the frequency of exceedance, which are based on analyses of about 17,000 ground water samples, suggest that about 40% of both large and small regulated water supplies have As concentration greater than 1 μg/L. The frequency of exceedance decreases for greater As concentrations—about 5% of systems are estimated to have As concentration greater than 20 μg/L. Comparison of these estimates with previously published work, based on 275 samples collected from regulated water supplies, shows very good agreement for the United States as a whole, although the two approaches yield somewhat different results for some parts of the nation.

Chemical contaminants in water and sediment near fish nesting sites in the Potomac River basin: determining potential exposures to smallmouth bass (Micropterus dolomieu).

The Potomac River basin is an area where a high prevalence of abnormalities such as testicular oocytes (TO), skin lesions, and mortality has been observed in smallmouth bass (SMB, Micropterus dolomieu). Previous research documented a variety of chemicals in regional streams, implicating chemical exposure as one plausible explanation for these biological effects. Six stream sites in the Potomac basin (and one out-of-basin reference site) were sampled to provide an assessment of chemicals in these streams. Potential early life-stage exposure to chemicals detected was assessed by collecting samples in and around SMB nesting areas. Target chemicals included those known to be associated with important agricultural and municipal wastewater sources in the Potomac basin. The prevalence and severity of TO in SMB were also measured to determine potential relations between chemistry and biological effects. A total of 39 chemicals were detected at least once in the discrete-water samples, with atrazine, caffeine, deethylatrazine, simazine, and iso-chlorotetracycline being most frequently detected. Of the most frequently detected chemicals, only caffeine was detected in water from the reference site. No biogenic hormones/sterols were detected in the discrete-water samples. In contrast, 100 chemicals (including six biogenic hormones/sterols) were found in a least one passive-water sample, with 25 being detected at all such samples. In addition, 46 chemicals (including seven biogenic hormones/sterols) were found in the bed-sediment samples, with caffeine, cholesterol, indole, para-cresol, and sitosterol detected in all such samples. The number of herbicides detected in discrete-water samples per site had a significant positive relation to TO(rank) (a nonparametric indicator of TO), with significant positive relations between TO(rank) and atrazine concentrations in discrete-water samples and to total hormone/sterol concentration in bed-sediment samples. Such significant correlations do not necessarily imply causation, as these chemical compositions and concentrations likely do not adequately reflect total SMB exposure history, particularly during critical life stages.

Comparing wastewater chemicals, indicator bacteria concentrations, and bacterial pathogen genes as fecal pollution indicators

The objective of this study was to compare fecal indicator bacteria (FIB) (fecal coliforms, Escherichia coli [EC], and enterococci [ENT]) concentrations with a wide array of typical organic wastewater chemicals and selected bacterial genes as indicators of fecal pollution in water samples collected at or near 18 surface water drinking water intakes. Genes tested included esp (indicating human-pathogenic ENT) and nine genes associated with various animal sources of shiga-toxin-producing EC (STEC). Fecal pollution was indicated by genes and/or chemicals for 14 of the 18 tested samples, with little relation to FIB standards. Of 13 samples with <50 EC 100 mL(-1), human pharmaceuticals or chemical indicators of wastewater treatment plant effluent occurred in six, veterinary antibiotics were detected in three, and stx1 or stx2 genes (indicating varying animal sources of STEC) were detected in eight. Only the EC eaeA gene was positively correlated with FIB concentrations. Human-source fecal pollution was indicated by the esp gene and the human pharmaceutical carbamazepine in one of the nine samples that met all FIB recreational water quality standards. Escherichia coli rfbO157 and stx2c genes, which are typically associated with cattle sources and are of potential human health significance, were detected in one sample in the absence of tested chemicals. Chemical and gene-based indicators of fecal contamination may be present even when FIB standards are met, and some may, unlike FIB, indicate potential sources. Application of multiple water quality indicators with variable environmental persistence and fate may yield greater confidence in fecal pollution assessment and may inform remediation decisions.

Genes Indicative of Zoonotic and Swine Pathogens Are Persistent in Stream Water and Sediment following a Swine Manure Spill

ABSTRACT Manure spills into streams are relatively frequent, but no studies have characterized stream contamination with zoonotic and veterinary pathogens, or fecal chemicals, following a spill. We tested stream water and sediment over 25 days and downstream for 7.6 km for the following: fecal indicator bacteria (FIB), the fecal indicator chemicals cholesterol and coprostanol, 20 genes for zoonotic and swine-specific bacterial pathogens by presence/absence PCR for viable cells, one swine-specific Escherichia coli toxin gene (STII gene) by quantitative PCR (qPCR), and nine human and animal viruses by qPCR or reverse transcription-qPCR. Twelve days postspill, and 4.2 km downstream, water concentrations of FIB, cholesterol, and coprostanol were 1 to 2 orders of magnitude greater than those detected before, or above, the spill, and genes indicating viable zoonotic or swine-infectious Escherichia coli were detected in water or sediment. STII gene levels increased from undetectable before or above the spill to 105 copies/100 ml of water 12 days postspill. Thirteen of 14 water (8/9 sediment) samples had viable STII-carrying cells postspill. Eighteen days postspill, porcine adenovirus and teschovirus were detected 5.6 km downstream. FIB concentrations (per gram [wet weight]) in sediment were greater than in water, and sediment was a continuous reservoir of genes and chemicals postspill. Constituent concentrations were much lower, and detections less frequent, in a runoff event (200 days postspill) following manure application, although the swine-associated STII and stx 2e genes were detected. Manure spills are an underappreciated pathway for livestock-derived contaminants to enter streams, with persistent environmental outcomes and the potential for human and veterinary health consequences.

Contamination with bacterial zoonotic pathogen genes in U.S. streams influenced by varying types of animal agriculture.

Animal waste, stream water, and streambed sediment from 19 small (<32km(2)) watersheds in 12U.S. states having either no major animal agriculture (control, n=4), or predominantly beef (n=4), dairy (n=3), swine (n=5), or poultry (n=3) were tested for: 1) cholesterol, coprostanol, estrone, and fecal indicator bacteria (FIB) concentrations, and 2) shiga-toxin producing and enterotoxigenic Escherichia coli, Salmonella, Campylobacter, and pathogenic and vancomycin-resistant enterococci by polymerase chain reaction (PCR) on enrichments, and/or direct quantitative PCR. Pathogen genes were most frequently detected in dairy wastes, followed by beef, swine and poultry wastes in that order; there was only one detection of an animal-source-specific pathogen gene (stx1) in any water or sediment sample in any control watershed. Post-rainfall pathogen gene numbers in stream water were significantly correlated with FIB, cholesterol and coprostanol concentrations, and were most highly correlated in dairy watershed samples collected from 3 different states. Although collected across multiple states and ecoregions, animal-waste gene profiles were distinctive via discriminant analysis. Stream water gene profiles could also be discriminated by the watershed animal type. Although pathogen genes were not abundant in stream water or streambed samples, PCR on enrichments indicated that many genes were from viable organisms, including several (shiga-toxin producing or enterotoxigenic E. coli, Salmonella, vancomycin-resistant enterococci) that could potentially affect either human or animal health. Pathogen gene numbers and types in stream water samples were influenced most by animal type, by local factors such as whether animals had stream access, and by the amount of local rainfall, and not by studied watershed soil or physical characteristics. Our results indicated that stream water in small agricultural U.S. watersheds was susceptible to pathogen gene inputs under typical agricultural practices and environmental conditions. Pathogen gene profiles may offer the potential to address both source of, and risks associated with, fecal pollution.

Regional variability in bed-sediment concentrations of wastewater compounds, hormones and PAHs for portions of coastal New York and New Jersey impacted by hurricane Sandy

Bed sediment samples from 79 coastal New York and New Jersey, USA sites were analyzed for 75 compounds including wastewater associated contaminants, PAHs, and other organic compounds to assess the post-Hurricane Sandy distribution of organic contaminants among six regions. These results provide the first assessment of wastewater compounds, hormones, and PAHs in bed sediment for this region. Concentrations of most wastewater contaminants and PAHs were highest in the most developed region (Upper Harbor/Newark Bay, UHNB) and reflected the wastewater inputs to this area. Although the lack of pre-Hurricane Sandy data for most of these compounds make it impossible to assess the effect of the storm on wastewater contaminant concentrations, PAH concentrations in the UHNB region reflect pre-Hurricane Sandy conditions in this region. Lower hormone concentrations than predicted by the total organic carbon relation occurred in UHNB samples, suggesting that hormones are being degraded in the UHNB region.

Resetting the bar: Establishing baselines for persistent contaminants after Hurricane Sandy in the coastal environments of New Jersey and New York, USA

In the immediate aftermath of natural disasters, public health officials and other first responders engage in many activities to protect the public and ecosystems in the affected area. These activities include critical tasks designed to minimize adverse consequences resulting from chemical and microbial contaminant exposures, such as acute disease incidence and transmission. However, once these urgent priorities have been met and situations requiring immediate attention have been stabilized, questions regarding the potential longer term threats to humans and ecosystems associated with persistent contaminant exposures remain. Research conducted to address these questions is frequently challenged by the lack of available baseline contaminant information collected before the event for comparison and perspective. In addition, deployments of field crews and collection of environmental samples typically occur days, weeks, or months after the event. Consequently, during and in the aftermath of disasters, public health agencies commonly advise the public to disinfect water, avoid contact with disturbed infrastructure (such as sewer lines), and (or) refrain from use of recreational waters, with the general focus on acute health threats; however, the persisting effects of such releases on local recreational waters, fisheries, and other estuarine habitats are often undetermined. Potential contaminant sources are prevalent in developed coastal areas and include combined sewer overflows; debris from buildings, automobiles, and boats; inundated infrastructure such as gas stations, landfills, chemical storage facilities, and hazardous waste sites; and saline water intrusion in estuaries and on shorelines. The vulnerability of nearshore environments to storm-derived contamination depends on factors associated with the storm (such as wind speed, precipitation, path, and timing) and the contaminant sources it disturbs. These drivers, which are often the basis of predictive models such as ARkStorm (Plumlee et al., 2015), are not commonlymonitored together or within the context of the potential environmental health threats associated with long-term contaminant exposures after storms. Consequently, the requisite data for determination and (or) modeling of contaminant impacts effects resulting from large coastal storms are typically lacking or minimal. For example, contaminant (chemical and microbial) vulnerability of an estuary to a storm that generates largescale interior flooding will differ from that of a storm that generates tidal surges. Hurricane Sandy (commonly referred to informally as “Superstorm Sandy”) made landfall in the United States (US) near Brigantine, New Jersey (NJ) on October 29, 2012 and was the second costliest hurricane