Machine learning predictions of nitrate in groundwater used for drinking supply in the conterminous United States.

Groundwater is an important source of drinking water supplies in the conterminous United State (CONUS), and presence of high nitrate concentrations may limit usability of groundwater in some areas because of the potential negative health effects. Prediction of locations of high nitrate groundwater is needed to focus mitigation and relief efforts. A three-dimensional extreme gradient boosting (XGB) machine learning model was developed to predict the distribution of nitrate. Nitrate was predicted at a 1 km resolution for two drinking water zones, each of variable depth, one for domestic supply and one for public supply. The model used measured nitrate concentrations from 12,082 wells and included predictor variables representing well characteristics, hydrologic conditions, soil type, geology, land use, climate, and nitrogen inputs. Predictor variables derived from empirical or numerical process-based models were also included to integrate information on controlling processes and conditions. The model provided accurate estimates at national and regional scales: the training (R2 of 0.83) and hold-out (R2 of 0.49) data fits compared favorably to previous studies. Predicted nitrate concentrations were less than 1 mg/L across most of the CONUS. Nationally, well depth, soil and climate characteristics, and the absence of developed land use were among the most influential explanatory factors. Only 1% of the area in either water supply zone had predicted nitrate concentrations greater than 10 mg/L; however, about 1.4 M people depend on groundwater for their drinking supplies in those areas. Predicted high concentrations of nitrate were most prevalent in the central CONUS. In areas of predicted high nitrate concentration, applied manure, farm fertilizer, and agricultural land use were influential predictor variables. This work represents the first application of XGB to a three-dimensional national-scale groundwater quality model and provides a significant milestone in the efforts to document nitrate in groundwater across the CONUS.

Comparison of a novel passive sampler to standard water-column sampling for organic contaminants associated with wastewater effluents entering a New Jersey stream.

Four water samples collected using standard depth and width water-column sampling methodology were compared to an innovative passive, in situ, sampler (the polar organic chemical integrative sampler or POCIS) for the detection of 96 organic wastewater-related contaminants (OWCs) in a stream that receives agricultural, municipal, and industrial wastewaters. Thirty-two OWCs were identified in POCIS extracts whereas 9-24 were identified in individual water-column samples demonstrating the utility of POCIS for identifying contaminants whose occurrence are transient or whose concentrations are below routine analytical detection limits. Overall, 10 OWCs were identified exclusively in the POCIS extracts and only six solely identified in the water-column samples, however, repetitive water samples taken using the standard method during the POCIS deployment period required multiple trips to the sampling site and an increased number of samples to store, process, and analyze. Due to the greater number of OWCs detected in the POCIS extracts as compared to individual water-column samples, the ease of performing a single deployment as compared to collecting and processing multiple water samples, the greater mass of chemical residues sequestered, and the ability to detect chemicals which dissipate quickly, the passive sampling technique offers an efficient and effective alternative for detecting OWCs in our waterways for wastewater contaminants.

Frequently co-occurring pesticides and volatile organic compounds in public supply and monitoring wells, southern New Jersey, USA.

One or more pesticides were detected with one or more volatile organic compounds (VOCs) in more than 95% of samples collected from 30 public supply and 95 monitoring wells screened in the unconsolidated surficial aquifer system of southern New Jersey, USA. Overall, more than 140,000 and more than 3,000 unique combinations of pesticides with VOCs were detected in two or more samples from the supply and monitoring wells, respectively. More than 400 of these combinations were detected in 20% or more of the samples from the supply wells, whereas only 17 were detected in 20% or more of the samples from the monitoring wells. Although many constituent combinations detected in water from the supply and monitoring wells are similar, differences in constituent combinations also were found and can be attributed, in part, to differences in the characteristics of these two well types. The monitoring wells sampled during this study yield water that typically was recharged beneath a single land-use setting during a recent, discrete time interval and that flowed along relatively short paths to the wells. Public supply wells, in contrast, yield large volumes of water and typically have contributing areas that are orders of magnitude larger than those of the monitoring wells. These large contributing areas generally encompass multiple land uses; moreover, because flow paths that originate in these areas vary in length, these wells typically yield water that was recharged over a large temporal interval. Water withdrawn from public supply wells, therefore, contains a mixture of waters of different ages that were recharged beneath various land-use settings. Because public supply wells intercept water flowing along longer paths with longer residence times and integrate waters from a larger source area than those associated with monitoring wells, they are more likely to yield water that contains constituents that were used in greater quantities in the past, that were introduced from point sources, and/or that are derived from the degradation of parent compounds along extended flow paths.