Prioritizing water availability study settings to address geogenic contaminants and related societal factors.

Water availability for human and ecological uses depends on both water quantity and water quality. The U.S. Geological Survey (USGS) is developing strategies for prioritizing regional-scale and watershed basin-scale studies of water availability across the nation. Previous USGS ranking processes for basin-scale studies incorporated primarily water quantity factors but are now considering additional water quality factors. This study presents a ranking based on the potential impacts of geogenic constituents on water quality and consideration of societal factors related to water quality. High-concentration geogenic constituents, including trace elements and radionuclides, are among the most prevalent contaminants limiting water availability in the USA and globally. Geogenic constituents commonly occur in groundwater because of subsurface water-rock interactions, and their distributions are controlled by complex geochemical processes. Geogenic constituent mobility can also be affected by human activities (e.g., mining, energy production, irrigation, and pumping). Societal factors and relations to drinking water sources and water quality information are often overlooked when evaluating research priorities. Sociodemographic characteristics, data gaps resulting from historical data-collection disparities, and infrastructure condition/age are examples of factors to consider regarding environmental justice. This paper presents approaches for ranking and prioritizing potential basin-scale study areas across the contiguous USA by considering a suite of conventional physical and geochemical variables related to geogenic constituents, with and without considering variables related to societal factors. Simultaneous consideration of societal and conventional factors could provide decision makers with more diverse, interdisciplinary tools to increase equity and reduce bias in prioritizing focused research areas and future water availability studies.

Machine-Learning Predictions of High Arsenic and High Manganese at Drinking Water Depths of the Glacial Aquifer System, Northern Continental United States.

Globally, over 200 million people are chronically exposed to arsenic (As) and/or manganese (Mn) from drinking water. We used machine-learning (ML) boosted regression tree (BRT) models to predict high As (>10 µg/L) and Mn (>300 µg/L) in groundwater from the glacial aquifer system (GLAC), which spans 25 states in the northern United States and provides drinking water to 30 million people. Our BRT models' predictor variables (PVs) included recently developed three-dimensional estimates of a suite of groundwater age metrics, redox condition, and pH. We also demonstrated a successful approach to significantly improve ML prediction sensitivity for imbalanced data sets (small percentage of high values). We present predictions of the probability of high As and high Mn concentrations in groundwater, and uncertainty, at two nonuniform depth surfaces that represent moving median depths of GLAC domestic and public supply wells within the three-dimensional model domain. Predicted high likelihood of anoxic condition (high iron or low dissolved oxygen), predicted pH, relative well depth, several modeled groundwater age metrics, and hydrologic position were all PVs retained in both models; however, PV importance and influence differed between the models. High-As and high-Mn groundwater was predicted with high likelihood over large portions of the central part of the GLAC.

Months-long spike in aqueous arsenic following domestic well installation and disinfection: Short- and long-term drinking water quality implications.

Exposure to high concentration geogenic arsenic via groundwater is a worldwide health concern. Well installation introduces oxic drilling fluids and hypochlorite (a strong oxidant) for disinfection, thus inducing geochemical disequilibrium. Well installation causes changes in geochemistry lasting 12 + months, as illustrated in a recent study of 250 new domestic wells in Minnesota, north-central United States. One study well had extremely high initial arsenic (1550 µg/L) that substantially decreased after 15 months (5.2 µg/L). The drilling and development of the study well were typical and ordinary; nothing observable indicated the very high initial arsenic concentration. We hypothesized that oxidation of arsenic-containing sulfides (which lowers pH) combined with low pH dissolution of arsenic-bearing Fe (oxyhydr)oxides caused the very high arsenic concentration. Geochemical equilibrium considerations and modeling supported our hypothesis. Groundwater equilibrium redox conditions are poised at the Fe(III)(s)/Fe(II)(aq) stability boundary, indicating arsenic-bearing Fe (oxyhydr)oxide mineral sensitivity to pH and redox changes. Changing groundwater geochemistry can have negative implications for home water treatment (e.g., reduced arsenic removal efficiency, iron fouling), which can lead to ongoing but unrecognized hazard of arsenic exposure from domestic well water. Our results may inform arsenic mobilization processes and geochemical sensitivity in similarly complex aquifers in Southeast Asia and elsewhere.

Machine Learning Predictions of pH in the Glacial Aquifer System, Northern USA.

A boosted regression tree model was developed to predict pH conditions in three dimensions throughout the glacial aquifer system of the contiguous United States using pH measurements in samples from 18,386 wells and predictor variables that represent aspects of the hydrogeologic setting. Model results indicate that the carbonate content of soils and aquifer materials strongly controls pH and, when coupled with long flowpaths, results in the most alkaline conditions. Conversely, in areas where glacial sediments are thin and carbonate-poor, pH conditions remain acidic. At depths typical of drinking-water supplies, predicted pH >7.5-which is associated with arsenic mobilization-occurs more frequently than predicted pH <6-which is associated with water corrosivity and the mobilization of other trace elements. A novel aspect of this model was the inclusion of numerically based estimates of groundwater flow characteristics (age and flowpath length) as predictor variables. The sensitivity of pH predictions to these variables was consistent with hydrologic understanding of groundwater flow systems and the geochemical evolution of groundwater quality. The model was not developed to provide precise estimates of pH at any given location. Rather, it can be used to more generally identify areas where contaminants may be mobilized into groundwater and where corrosivity issues may be of concern to prioritize areas for future groundwater monitoring.

Time scales of arsenic variability and the role of high-frequency monitoring at three water-supply wells in New Hampshire, USA.

Groundwater geochemistry, redox process classification, high-frequency physicochemical and hydrologic measurements, and climate data were analyzed to identify controls on arsenic (As) concentration changes. Groundwater was monitored in two public-supply wells (one glacial aquifer and one bedrock aquifer), and one bedrock-aquifer domestic well in New Hampshire, USA, from 2014 to 2018 to identify time scales of and controls on As concentration changes. Concentrations of As and other geochemical constituents were measured bimonthly. Specific conductance (SC), pH, dissolved oxygen, and pumping rate/water level were measured at high frequency (every 5 to 15 min). Median (and 95% confidence interval) As concentrations at the three wells were 4.1 (3.7-4.6), 18.9 (17.2-23.6), and 37.5 (30.4-42.9) µg/L. Arsenic variability in each of the three wells, in relative standard deviation, ranged from 9 to 12%. Median quarterly As concentrations were highest in all wells in the spring. The bedrock-aquifer public-supply well As concentration increased over the period of study while pumping rate decreased. In the public-supply wells, As variability was correlated with SC and pH, and As species were related to SC, pH, pumping, precipitation, and changes in redox process. Specific conductance also had a seasonal pattern in the two public-supply wells and was correlated with Na and Cl. Excess Na in water samples suggests possible ion exchange with dissolved Ca, creating more capacity to dissolve CaCO3 from calcareous rocks, which can increase pH and in turn, As concentrations in wells. High-frequency monitoring data are cost effective to collect, which could be advantageous in other parts of the United States and in the many parts of the world where glacial aquifers are in direct contact with other water supply aquifers or where water from different aquifers have potential to mix.

Arsenic concentrations after drinking water well installation: Time-varying effects on arsenic mobilization.

Chronic exposure to geogenic arsenic via drinking water is a worldwide health concern. However, effects of well installation and operation on arsenic concentrations and mobilization are not well understood. This knowledge gap impacts both reliable detection of arsenic in drinking water and effective public health recommendations to reduce exposure to arsenic. This study examines changes in arsenic and redox geochemistry over one year following installation of 254 new domestic water wells in three regions of the north-central USA that commonly have elevated arsenic concentrations. Our regions' geologic settings share some important characteristics with other high-arsenic aquifers: igneous bedrock aquifers; or late Pleistocene-age glacial sand and gravel aquifers interbedded with aquitards. Over the study, arsenic concentrations increased by 16% or more in 25% of wells in glacial aquifer regions, and the redox conditions changed towards more reducing. In wells in the bedrock region, there was no significant change in arsenic concentrations, and redox conditions changed towards more oxidizing. Our findings illustrate the importance of understanding short- to moderate-term impacts of well installation and operation on arsenic and aqueous chemistry, as it relates to human exposure. Our study informs water quality sampling requirements, which currently do not consider the implications sampling timing with respect to well installation. Evaluating arsenic concentrations in samples from new wells in the context of general regional pH and redox conditions can provide information regarding the degree of disequilibrium created by well drilling. Our analysis approach may be transferable and scalable to similar aquifer settings across the globe.

Concentrations of pharmaceuticals and other micropollutants in groundwater downgradient from large on-site wastewater discharges.

Large subsurface treatment systems (LSTS) and rapid infiltration basins (RIB) are preferred onsite wastewater treatments compared to direct discharge of treated wastewater to streams and adjacent facilities. Discharge of these wastewater treatments may result in contaminant loading to aquifers that also serve as drinking water sources downgradient from the discharge site. Until recently, few studies have characterized the contribution of micropollutants (e.g. pharmaceuticals, fragrances, flame retardants, etc.) to receiving aquifers. We conducted a pilot project to characterize the occurrence of micropollutants in groundwater downgradient from 7 on-site treatment systems in Minnesota, USA: 5 community LSTS and 2 municipal RIB. One downgradient monitoring well was sampled three times at each facility over one year. Of 223 micropollutants analyzed, 35 were detected. Total sample concentrations ranged from 90 to 4,039 ng/L. Sulfamethoxazole (antibiotic) was detected in all samples at concentrations from 7 to 965 ng/L. Other pharmaceuticals (0.12-1,000 ng/L), organophosphorus flame retardants (10-500 ng/L), and other anthropogenic chemicals (4-775 ng/L) were also detected. The numbers and concentrations of micropollutants detected were inversely related to dissolved oxygen and depth to water. Ratios of pharmaceutical concentrations to human-health screening values were <0.10 for most samples. However, concentrations of carbamazepine and sulfamethoxazole exceeded screening values at two sites. Study results illustrate that large on-site wastewater systems designed to discharge to permeable soil or shallow groundwater effectively deliver pharmaceuticals and other micropollutants to groundwater aquifers and could contribute micropollutants to drinking water via water supply wells.

How or When Samples Are Collected Affects Measured Arsenic Concentration in New Drinking Water Wells.

Naturally occurring arsenic can adversely affect water quality in geologically diverse aquifers throughout the world. Chronic exposure to arsenic via drinking water is a human health concern due to risks for certain cancers, skin abnormalities, peripheral neuropathy, and other negative health effects. Statewide in Minnesota, USA, 11% of samples from new drinking water wells have arsenic concentrations exceeding 10 µg/L; in certain counties more than 35% of tested samples exceed 10 µg/L arsenic. Since 2008, Minnesota well code has required testing water from new wells for arsenic. Sample collection protocols are not specified in the well code, so among 180 well drillers there is variability in sampling methods, including sample collection point and sample collection timing. This study examines the effect of arsenic sample collection protocols on the variability of measured arsenic concentrations in water from new domestic water supply wells. Study wells were drilled between 2014 and 2016 in three regions of Minnesota that commonly have elevated arsenic concentrations in groundwater. Variability in measured arsenic concentration at a well was reduced when samples were (1) filtered, (2) collected from household plumbing instead of from the drill rig pump, or (3) collected several months after well construction (instead of within 4 weeks of well installation). Particulates and fine aquifer sediments entrained in groundwater samples, or other artifacts of drilling disturbance, can cause undesirable variability in measurements. Establishing regulatory protocols requiring sample filtration and/or collection from household plumbing could improve the reliability of information provided to well owners and to secondary data users.

Detecting sulfamethoxazole and carbamazepine in groundwater: Is ELISA a reliable screening tool?

In recent years, numerous studies have reported the prevalence of organic micropollutants in natural waters. There is an increasing interest in assessing the occurrence and transport of these contaminants in groundwater because a large number of people in the United States rely on groundwater for their drinking water. However, commonly used mass-spectrometry-based analytical methods are expensive and time-consuming. The enzyme-linked immunosorbent assay (ELISA) method offers an inexpensive analytical alternative that provides semi-quantitative results in a relatively quick timeframe. We investigated the use of ELISA for two commonly detected micropollutants, sulfamethoxazole (SMX) and carbamazepine (CBZ), in groundwater collected as part of two different studies, one in Minnesota and the other in Iowa. The ELISA results were compared with two mass-spectrometry-based methods: (1) direct aqueous injection-high performance liquid chromatography/tandem mass spectrometry (HPLC) and (2) online solid-phase extraction with liquid chromatography/electrospray ionization-mass spectrometry (SPE LC). Differences in SMX and CBZ observations between ELISA and both HPLC and SPE LC were analyzed using the Paired Prentice-Wilcoxon test. Estimates of bias and limits of agreement between paired observations also were calculated. The SMX determinations by ELISA yielded results that were 30 and 14% greater than HPLC and SPE LC, respectively. The CBZ determinations by ELISA yielded results that were 25 and 9% greater than HPLC and SPE LC, respectively. The ELISA determinations were in presence-absence agreement with HPLC for 83% of samples for SMX and CBZ; and with SPE LC for 76 and 80% of samples for SMX and CBZ, respectively. Results indicate that ELISA for SMX and CBZ is a reliable and cost effective screening-tool alternative to more commonly used mass spectrometry-based analytical methods.

Crude Oil Metabolites in Groundwater at Two Spill Sites.

Two groundwater plumes in north central Minnesota with residual crude oil sources have 20 to 50 mg/L of nonvolatile dissolved organic carbon (NVDOC). These values are over 10 times higher than benzene and two to three times higher than Diesel Range Organics in the same wells. On the basis of previous work, most of the NVDOC consists of partial transformation products from the crude oil. Monitoring data from 1988 to 2015 at one of the sites located near Bemidji, MN show that the plume of metabolites is expanding toward a lakeshore located 335 m from the source zone. Other mass balance studies of the site have demonstrated that the plume expansion is driven by the combined effect of continued presence of the residual crude oil source and depletion of the electron accepting capacity of solid phase iron oxide and hydroxides on the aquifer sediments. These plumes of metabolites are not covered by regulatory monitoring and reporting requirements in Minnesota and other states. Yet, a review of toxicology studies indicates that polar metabolites of crude oil may pose a risk to aquatic and mammalian species. Together the results suggest that at sites where residual sources are present, monitoring of NVDOC may be warranted to evaluate the fates of plumes of hydrocarbon transformation products.

Arsenic Cycling in Hydrocarbon Plumes: Secondary Effects of Natural Attenuation.

Monitored natural attenuation is widely applied as a remediation strategy at hydrocarbon spill sites. Natural attenuation relies on biodegradation of hydrocarbons coupled with reduction of electron acceptors, including solid phase ferric iron (Fe(III)). Because arsenic (As) adsorbs to Fe-hydroxides, a potential secondary effect of natural attenuation of hydrocarbons coupled with Fe(III) reduction is a release of naturally occurring As to groundwater. At a crude-oil-contaminated aquifer near Bemidji, Minnesota, anaerobic biodegradation of hydrocarbons coupled to Fe(III) reduction has been well documented. We collected groundwater samples at the site annually from 2009 to 2013 to examine if As is released to groundwater and, if so, to document relationships between As and Fe inside and outside of the dissolved hydrocarbon plume. Arsenic concentrations in groundwater in the plume reached 230 µg/L, whereas groundwater outside the plume contained less than 5 µg/L As. Combined with previous data from the Bemidji site, our results suggest that (1) naturally occurring As is associated with Fe-hydroxides present in the glacially derived aquifer sediments; (2) introduction of hydrocarbons results in reduction of Fe-hydroxides, releasing As and Fe to groundwater; (3) at the leading edge of the plume, As and Fe are removed from groundwater and retained on sediments; and (4) downgradient from the plume, patterns of As and Fe in groundwater are similar to background. We develop a conceptual model of secondary As release due to natural attenuation of hydrocarbons that can be applied to other sites where an influx of biodegradable organic carbon promotes Fe(III) reduction.

Glacial sediment causing regional-scale elevated arsenic in drinking water.

In the upper Midwest, USA, elevated arsenic concentrations in public drinking water systems are associated with the lateral extent of northwest provenance late Wisconsin-aged drift. Twelve percent of public water systems located within the footprint of this drift (212 of 1764) exceed 10 microg/L arsenic, which is the U.S. EPA's drinking water standard. Outside of the footprint, only 2.4% of public water systems (52 of 2182) exceed 10 microg/L arsenic. Both glacial drift aquifers and shallow bedrock aquifers overlain by northwest provenance late Wisconsin-aged sediment are affected by arsenic contamination. Evidence suggests that the distinct physical characteristics of northwest provenance late Wisconsin-aged drift--its fine-grained matrix and entrained organic carbon that fosters biological activity--cause the geochemical conditions necessary to mobilize arsenic via reductive mechanisms such as reductive desorption and reductive dissolution of metal oxides. This study highlights an important and often unrecognized phenomenon: high-arsenic sediment is not necessary to cause arsenic-impacted ground water--when "impacted" is now defined as >10 microg/L. This analysis also demonstrates the scientific and economic value of using existing large but imperfect statewide data sets to observe and characterize regional-scale environmental problems.

Well characteristics influencing arsenic concentrations in ground water.

Naturally occurring arsenic contamination is common in ground water in the upper Midwest. Arsenic is most likely to be present in glacial drift and shallow bedrock wells that lie within the footprint of northwest provenance Late Wisconsinan glacial drift. Elevated arsenic is more common in domestic wells and in monitoring wells than it is in public water system wells. Arsenic contamination is also more prevalent in domestic wells with short screens set in proximity to an upper confining unit, such as glacial till. Public water system wells have distinctly different well-construction practices and well characteristics when compared to domestic and monitoring wells. Construction practices such as exploiting a thick, coarse aquifer and installing a long well screen yield good water quantity for public water system wells. Coincidentally, these construction practices also often yield low arsenic water. Coarse aquifer materials have less surface area for adsorbing arsenic, and thus less arsenic available for potential mobilization. Wells with long screens set at a distance from an upper confining unit are at lower risk of exposure to geochemical conditions conducive to arsenic mobilization via reductive mechanisms such as reductive dissolution of metal hydroxides and reductive desorption of arsenic.