Geology topics

Laura Medalie

Publications and source records attributed to Laura Medalie.

At least 19 recordsLinked to original sources

USGS2025

The National integrated water availability assessment, water years 2010–20

Water availability is fundamentally important to human well-being, economic vitality, and ecosystem health. Because of its central importance, the U.S. Congress tasked the U.S. Geological Survey (USGS) and other Federal agencies with conducting regular, comprehensive assessments of water availability in the United States through the requirements under the SECURE Water Act. In response to this mandate, the USGS has developed the U.S. Geological Survey Integrated Water Availability Assessment—2010–20, which addresses aspects of water supply, quality, and use related to water availability in the United States. This is the first chapter of that report. The major climatic factors affecting water availability are also described. Multiple aspects of water availability are integrated to produce a more comprehensive analysis of water availability in the United States. This chapter enumerates the development, organization, and tools used in the USGS Integrated Water Availability Assessment. SECURE Water Act reports, developed by the Department of Energy Hydropower Climate Change Assessment and the Bureau of Reclamation West-Wide Climate and Hydrology Assessment, are also described. A distilled list of key findings from the overall report is also provided, serving as an introduction to each topic along with the most important high-level information.

conterminous United States↗

USGS2025

Water use across the conterminous United States, water years 2010–20

Withdrawals of water for human use are fundamental to the evaluation of the Nation’s water availability. This chapter provides an analysis of public supply, crop irrigation, and thermoelectric power water use for the conterminous United States (CONUS) during water years 2010–20. These three categories account for about 90 percent of water withdrawals in the Nation. The values presented here are based on modeling approaches that estimate water use at temporal (monthly) and spatial scales (12-digit hydrologic unit code—small watersheds sized 50–100 square kilometers) compatible for integration into a broader national assessment of water availability. Models also provide an understanding of factors that influence water use. An estimated 244,817 million gallons per day (Mgal/d; 28,677 million cubic meters per month [Mm 3 /mo]) were withdrawn on average within the CONUS during water years 2010–20 from fresh water and saline water for crop irrigation, public supply, and thermoelectric power, with shares of 43, 14.5, and 42.5 percent for each of these categories, respectively. In the same period, estimated withdrawals and consumptive use (1) for public supply were 35,400 and 4,219 Mgal/d (4,081 and 486 Mm 3 /mo), respectively; (2) for crop irrigation were 105,497 and 75,698 Mgal/d (12,147 and 8,716 Mm 3 /mo), respectively; and (3) for thermoelectric power from fresh water were 82,656 and 2,904 Mgal/d (9,952 and 345 Mm 3 /mo), respectively. Withdrawals for these categories of water use are highly spatially variable, with western States dominated by crop irrigation and eastern States dominated by thermoelectric-power water use. Public supply accounts for the largest percentage of water use in several heavily populated northeastern States. Reliance on groundwater compared to surface water depends on the availability of water sources and the type of water use. For public supply, withdrawals from groundwater are greater than withdrawals from surface water in the Western aggregated hydrologic regions, whereas the balance shifts to more surface water for the rest of the CONUS. In all aggregated hydrologic regions, the predominant source of water for crop irrigation is groundwater. Most thermoelectric power facilities in the eastern half of the CONUS use surface water from freshwater and saline sources; most thermoelectric power facilities in the western half of the CONUS use groundwater.

conterminous United States↗

USGS2024

Predictability and behavior of water transfers across basin boundaries

Inter-basin water transfers (IBTs) are important components of water balances of basins, and they can have substantial impact on regional water availability. Flow information is often not available at locations with known IBTs, which is a drawback in several published IBT databases. Few, if any, studies examine whether IBT flow behavior can be generalized, and if these behaviors can be predicted at undocumented locations or known IBT locations with no flow information. In this study, we employ a clustering method based on image matching to identify similar classes of flow behavior of IBTs. Machine learning models are used to assess how well IBT flow characteristics (e.g., average flow) associated with these behaviors can be predicted. These evaluations of IBTs are done for two regions in the United States. Three primary classes of IBTs (seasonal, nonseasonal/not mixed, and seasonal/mixed) are identified across the two regions analyzed. The IBT flow characteristics are accurately predicted in the northeast region. In the Colorado region, however, only the flow characteristics related to timing were accurately predicted. These results indicate that the proposed modeling framework can be used to identify generalizable IBT flow characteristics. This framework is shown to predict flow characteristics with a reasonable amount of accuracy to undocumented locations and improves previously published IBT databases by backfilling flow information to locations with a known IBT presence.

Arizona, Colorado, Delaware, Kansas, Maryland, Mas↗

USGS2023

Assessment of factors that influence human water demand for Providence, Rhode Island

To determine the most relevant climatic and economic factors driving water demand for Providence, Rhode Island, and to further the understanding of human interactions with water availability, linear regression models were developed to estimate single-family and multifamily residential, commercial, and industrial water demand for the service area of Providence Water for 2014–21. Monthly water use delivery data were provided by Providence Water. An array of climatic and economic data, the drought index, and binary variables to represent seasonal water use and the onset of the coronavirus (COVID–19) were investigated as possible explanatory variables for the water demand models. The water demand model with the best fit with the least amount of error was the single-family residential water demand followed in descending order of accuracy by the commercial, multifamily residential, and industrial water demand. Seasonal variables were significant in all models, and the COVID–19 binary variable was significant in the commercial and industrial models. One or two economic variables were significant in all models and one climatic variable was significant in all models except the commercial model. Overall residential, commercial, and industrial water demand in the Providence, Rhode Island, service area has decreased during the study period most likely because of widescale drought conditions and policies designed to improve water efficiencies. The linear regression models developed for single-family and multifamily residential, commercial, and industrial water use explained 94, 85, 91, and 77 percent, respectively, of the variability in monthly water use. Multifamily residential water demand displayed a less distinct seasonal trend than that observed for single-family residential customers, likely because multifamily homes tend to use less water outdoors. The commercial water-demand model included no climatic variables, one economic variable, the COVID–19 pandemic variable, and the high and low water use seasonal variables—the latter two variables indicating the importance of seasonal fluctuations in water use. The COVID–19 pandemic and a concomitant State executive order had the immediate effect of severely reducing commercial water use. The industrial water-demand model did not perform as well as the other models because industrial water delivery data display a greater range of values, both seasonally and for the overall study period.

Rhode Island↗

USGS2021

Statistical methods for simulating structural stormwater runoff best management practices (BMPs) with the Stochastic Empirical Loading and Dilution Model (SELDM)

This report documents statistics for simulating structural stormwater runoff best management practices (BMPs) with the Stochastic Empirical Loading and Dilution Model (SELDM). The U.S. Geological Survey developed SELDM and the statistics documented in this report in cooperation with the Federal Highway Administration to indicate the risk for stormwater flows, concentrations, and loads to exceed user-selected water-quality goals and the potential effectiveness of mitigation measures to reduce such risks. In SELDM, three treatment variables—hydrograph extension, volume reduction, and water-quality treatment—are simulated by using the trapezoidal distribution and the rank correlation with the associated runoff variables. This report describes methods for calculating the trapezoidal distribution statistics and rank correlation coefficients for these treatment variables and methods for estimating the minimum irreducible concentration (MIC), which is the lowest expected effluent concentration from a BMP site or a category of BMPs. These statistics are different from the statistics commonly used to characterize or compare BMPs; they are designed to provide a stochastic transfer function to approximate the quantity, duration, and quality of BMP effluent given the associated inflow values for a population of storm events. Analyses for this study were done with data extracted from a modified copy of the December 2019 version of the International Stormwater Best Management Practices Database. Statistics for volume reduction, hydrograph extension, and water-quality treatment were developed with selected data. The medians of the best-fit statistics for selected constituents were used to construct generalized cumulative distribution functions for the three treatment variables. For volume reduction and hydrograph extension, selection of a Spearman’s rank correlation coefficient (rho) value that is the average of the median and maximum values for the BMP category may help generate realistic simulation results in SELDM. The median rho value may be selected to help generate realistic simulation results for water-quality treatment variables. Water-quality treatment statistics, including trapezoidal ratios and MIC values, were developed for 51 runoff-quality constituents commonly measured in highway and urban runoff studies. Statistics were calculated for water-quality properties, sediment and solids, nutrients, major and trace inorganic elements, organic compounds, and biologic constituents. Analysis of MIC values provides information to guide professional judgement for selecting values for simulating water quality at sites of interest. The MIC is a lower bound for BMP discharge concentrations and will therefore replace simulated BMP discharge concentrations below the selected value. A new method for estimating MIC values, the lognormal variate of inflow concentrations, was developed in this report and these statistics were calculated for individual constituents and constituent categories. Inflow quality is correlated to MIC values for some constituents, but regional soil concentrations were not strongly correlated to MIC values.

Scientific Investigations Report↗

USGS2021

Pesticides in US Rivers: Regional differences in use, occurrence, and environmental toxicity, 2013 to 2017

Pesticides pose a threat to the environment, but because of the substantial number of compounds, a comprehensive assessment of pesticides and an evaluation of the risk that they pose to human and aquatic life is challenging. In this study, improved analytical methods were used to quantify 221 pesticide concentrations in surface waters over the time period from 2013 to 2017. Samples were collected from 74 river sites in the conterminous US (CONUS). Potential toxicity was assessed by comparing surface water pesticide concentrations to standard concentrations that are considered to have adverse effects on human health or aquatic organisms. The majority of pesticide use is related to agriculture, and agricultural production varies across the CONUS. Therefore, our results were summarized by region (Northeast, South, Midwest, West and Pacific), with the expectation that crop production differences would drive variability in pesticide use, detection frequency, and benchmark exceedance patterns. Although agricultural pesticide use was at least 2.5 times higher in the Midwest (49 kg km −2 ) than in any of the other four regions (Northeast, South, West, and Pacific, 3 to 21 kg km −2 ) and the average number of pesticides detected in the Midwest was at least 1.5 higher ( n = 25) than the other four regions ( n = 8 to n = 16), the potential toxicity results were more evenly distributed. At least 50% of the sites within each of the 5 regions had at least 1 chronic benchmark exceedance. Imidacloprid posed the greatest potential threat to aquatic life with a total of 245 benchmark exceedances at 60 of the 74 sites. These results show that pesticides persist in the environment beyond the site of application and expected period of use. Continued monitoring and research are needed to improve our understanding of pesticide effects on aquatic and human life.

Science of the Total Environment↗

USGS2020

Considerations for incorporating quality control into water quality sampling strategies for the U.S. Geological Survey

This report describes considerations for incorporating routine quality-assessment and quality-control evaluations into U.S. Geological Survey discrete water-sampling programs and projects. U.S. Geological Survey water-data science in 2020 is characterized by robustness, external reproducibility, collaborative large-volume data analysis, and efficient delivery of water-quality data. Confidence in data, or robustness, can be increased by supplementing traditional field-based quality-control data with laboratory quality control (QC) data, such as third-party blind spikes and blind blanks, laboratory blanks, and laboratory-reagent spikes. Laboratory quality-control data can provide additional information about bias and variability, method performance, and false-positive and false-negative rates that are not available from field QC data alone. Reproducibility is supported by means of standardizing metadata and documentation. Collaborative analysis brings together disparate elements of various types of quality-control review and communicates persistent data quality issues for compounds to data users internal and external to the U.S. Geological Survey. Efficient delivery of water-quality data is achieved when quality-control review is accomplished in the same expedited (near real-time) time frame as distribution of environmental results to the public and might be improved with consideration given to data versioning or to a system of alerting data users to data interpretation that might differ from originally published data.

Open-File Report↗

USGS2020

Quality of data from the U.S. Geological Survey National Water Quality Network for water years 2013–17

Water samples from 122 sites in the U.S. Geological Survey National Water Quality Network were collected in 2013–17 to document ambient water-quality conditions in surface water of the United States and to determine status and trends of loads and concentrations for nutrients, contaminants, and sediment to estuaries and streams. Quality-control (QC) samples collected in the field with environmental samples were combined with QC samples from laboratory processing to provide information and documentation about the quality of the environmental data. Quality assurance for inorganic and organic compounds assessed in the National Water Quality Network includes collection of field blanks to determine contamination bias and field replicates to determine variability bias. No contamination bias was found for 6 of the 13 nutrient compounds analyzed, and some potential contamination bias for some years was found for the other 7 nutrient compounds. Contamination bias was not found for carbon compounds or ultraviolet-absorbance measurements and was not assessed for sediment. All major ions and trace elements except potassium and lithium showed moderate contamination bias for at least 1 water year; generally, this bias was not at environmentally relevant concentrations. All compounds in the nutrient, carbon, and sediment group and in the major ions and trace elements group had low variability both in detection frequency and in concentration. Exceptions to this low variability were total particulate inorganic carbon and sediment for 2015, both of which are particulate substances with intrinsically high sampling variability. The risk of contamination bias for pesticides in National Water Quality Network samples was low, as indicated by very few detections in field blanks. Sixteen pesticide compounds showed potential contamination bias based on unexpected detections in third-party blind spikes (false-positive results for compounds that are not included in the spike mixture of a sample, where the identity as a QC sample is unknown to the analyst), and 47 different compounds (out of 225 pesticide compounds) showed potential contamination bias from laboratory blanks. However, when timing and relative magnitudes of detections in blank samples, environmental samples, and benchmark concentrations are considered, most of this potential contamination is not relevant to interpretation of published pesticide results. Overall variability in detection frequency for pesticides from field replicates was low or moderate. Also based on field replicates, 55 pesticides had overall high variability in concentrations for at least 1 water year, although these assessments likely overestimate high variability. At least 1 QC issue was found for 87 pesticides; however, most of the QC issues had no or little effect on the interpretation of environmental results because the U.S. Geological Survey National Water Quality Laboratory addressed the QC issue before publishing the environmental results, environmental results were almost entirely nondetections, concentrations of environmental results were higher than potential contamination bias, or benchmark concentrations were orders of magnitude higher than all environmental results. Eight compounds affected by two QC issues had a benchmark less than 100 nanograms per liter and warranted careful consideration of timing and magnitude of QC results in relation to surface-water results before interpretive use.

Scientific Investigations Report↗

USGS2020

Quality of pesticide data for groundwater analyzed for the National Water-Quality Assessment Project, 2013–18

The National Water-Quality Assessment (NAWQA) Project of the U.S. Geological Survey (USGS) submitted nearly 1,900 samples collected from groundwater sites across the United States in 2013–18 for analysis of 225 pesticide compounds (pesticides and pesticide degradates, hereafter referred to as “pesticides”) by USGS National Water Quality Laboratory schedule 2437 (S2437). For the associated NAWQA study of pesticide occurrence and concentration in groundwater, and for other studies using pesticide results determined by S2437, it is necessary to assess the ability of reported results to meet data-quality requirements that will allow study objectives to be achieved. This assessment of the quality of S2437 results reported in 2013–18 examined data from field and laboratory quality-control samples, along with third-party performance assessment samples, to estimate bias and variability and to identify their potential sources, with an emphasis on implications for the interpretation of pesticide data for groundwater. Results indicate that measurements produced by the S2437 method for most pesticides have bias and variability that would be considered acceptable for many interpretative studies, which could therefore use the results without qualification or censoring. However, the reported data for a subset of pesticides have the potential for unacceptable contamination bias, high or low recovery bias, or high variability as a consequence of method performance and (or) nonlaboratory factors that could preclude their use for certain common objectives or could necessitate adjustment or qualification to meet those objectives. Based on data for laboratory blanks, censoring of some detections for a subset of pesticides reported by the laboratory in environmental samples might be necessary or desirable to avoid an unacceptably high likelihood of a false-positive result caused by laboratory contamination. The 90-percent upper confidence limit for the 95th percentile of laboratory blank concentration equals or exceeds the minimum reported groundwater concentration in at least 1 water year for 28 pesticides. During at least 1 water year, this upper confidence limit exceeds the maximum laboratory detection limit for 17 pesticides and exceeds the maximum laboratory reporting limit for 3 pesticides (ametryn, atrazine, and diazinon). The level of contamination indicated by this upper confidence limit should not substantially affect the suitability of reported environmental concentrations for any compound for comparison with corresponding human-health benchmarks. Despite being subjected to the same laboratory processes as laboratory blanks, field blanks indicated little evidence of contamination bias. This observation could largely be the consequence of data-reporting practices, which utilize detections in laboratory blanks to censor results in associated field samples (including blanks and environmental samples) when relative concentrations indicate that a result could have a substantial contribution from laboratory contamination. Laboratory censoring appears likely to reduce the risk of false-positive results in environmental samples below the level that laboratory blank results alone would imply. Whereas data available for third-party blind blank samples analyzed in 2018 indicate that only propoxur had any false-positive results, data for pesticides that were not spiked into blind spike samples analyzed in 2013–18 indicate that the false-positive rates for 31 pesticides exceeded 1 percent when considering only detections reported at concentrations greater than the maximum detection limit. Although about half of these pesticides lack substantial supporting evidence of contamination bias based on laboratory blank or field blank detections, indicating that spiking issues or degradation of parent compounds within the spiked samples might be a contributing factor to some false-positive results, these results indicate the need to closely examine detections reported for some pesticides in environmental samples analyzed during a similar period for possible contributions from contamination bias. Data for blind spike samples that were spiked at concentrations above the maximum reporting limit indicate that false-negative rates for eight pesticides exceed 10 percent; substantial low bias could affect results reported for these pesticides in environmental samples analyzed during a similar period. Data for laboratory reagent spikes, which measure recovery of pesticides in blank water, show little evidence for unacceptable recovery bias for S2437 pesticides. However, field matrix spikes, which measure recovery of pesticides in environmental matrices, indicate that degradation and (or) matrix effects could result in moderate to substantial low bias for groundwater results for several pesticides. Low bias could cause some reported concentrations to be categorized as being below a benchmark when the actual concentration in groundwater is greater than the benchmark. Occurrence and concentrations in groundwater could be substantially underrepresented for six pesticides with benchmarks (1H-1,2,4-triazole, asulam, bifenthrin, cis-permethrin, fenbutatin oxide, and naled) that have median recoveries between zero and 50 percent in field matrix spikes. Two compounds (didealkylatrazine and 2-hydroxy-6-ethylamino-4-amino-s-triazine) have median recoveries near or greater than 150 percent in field matrix spikes, indicating a substantial high bias. Plots of data for all spike types show clear changes in the typical recovery with time for some pesticides, which would require further examination for evaluation of temporal trends in environmental concentrations. Data for laboratory reagent spikes indicate that nearly all S2437 pesticides have acceptable variability resulting from random measurement error. Only two compounds (fenbutatin oxide and naled) have F-pseudosigma values greater than 30 percent for recovery, which implies the potential for relatively high variability in reported concentrations and could affect comparison of concentrations to benchmarks and determination of whether concentrations for samples collected at separate locations or times are truly different with a specified level of confidence. Data for third-party blind spike samples show relatively high variability for a greater number of pesticides, although these results likely reflect the influence of degradation and (or) differences in the magnitude and variability of concentrations used for blind spikes relative to laboratory reagent spikes. Detailed analysis of variability using field replicate data is possible for only 12 pesticides on S2437; low variability in analyte detection and concentration is indicated for most of these pesticides in groundwater.

Scientific Investigations Report↗

USGS2020

Influence of land use and region on glyphosate and aminomethylphosphonic acid in streams in the USA

Glyphosate is the most widely used herbicide in the United States for agricultural and non-agricultural weed control. Many studies demonstrate possible effects of glyphosate and its degradate AMPA on human and ecological health. Although glyphosate is thought to have limited mobility in soil, it is found year-round in many rivers and streams throughout the world in both agricultural and developed environments. It is vitally important to continue to increase the knowledge base of glyphosate use, distribution, transport, and impacts on human health and the environment. Here we show that glyphosate and AMPA are found in nearly all of 70 streams throughout the United States at concentrations far below human health or ecological benchmarks, with less occurrence in the Northeast and that undeveloped land, classified as such by land use near the sampling station, has lower concentrations compared to other types of land. Results also show that sites with large watersheds tend to have more AMPA than glyphosate and the opposite is true for small watersheds. Travel times and opportunity for glyphosate to degrade to AMPA and for reservoirs of AMPA to grow are greater in large watersheds. Factors that promoted quick movement of glyphosate to streams, such as subsurface tile or storm drains, sewers, overland flow from developed landscapes, and arid landscapes were associated with sites that had greater concentrations of glyphosate compared to AMPA. These results contribute contemporary information and generalized interpretations adding to the knowledge base of the fate of glyphosate on a national scale and provide a springboard for further exploration of technical processes controlling transport to streams.

Science of the Total Environment↗

USGS2019

Use of set blanks in reporting pesticide results at the U.S. Geological Survey National Water Quality Laboratory, 2001-15

Executive Summary Background .—Pesticide results from the U.S. Geological Survey (USGS) National Water Quality Laboratory (NWQL) are used for water-quality assessments by many agencies and organizations. The USGS is committed to providing data of the highest possible quality to the consumers of its data. A cooperator’s inquiries about specific pesticide detections in water revealed potential laboratory contamination issues for some results. Consequently, the USGS conducted an extensive evaluation of potential low-level contamination related to processing or analysis of water-quality samples at NWQL for 21 pesticide compounds of interest to the cooperator. This is the most comprehensive study of NWQL pesticide quality-control (QC) results to date. Purpose and scope .—The purpose of this study was to document protocols used by the NWQL to censor pesticide results and to determine the effects of laboratory contamination—as determined from detections in laboratory set blanks—on pesticide detections in groundwater and surface-water samples. More than 30,000 pesticide results from 113 selected batches of samples (2 percent or less of total batches) analyzed by the NWQL during the 15 years from 2001 to 2015 were reviewed. All laboratory results from the selected batches, including results from environmental (surface water and groundwater) and QC (set-blank, blind-blank, and blind-spike) samples, were evaluated. The study includes results for 21 pesticide compounds analyzed in groundwater and surface-water samples collected across the United States. Eleven pesticide compounds were analyzed by a gas chromatography/mass spectrometry method and 10 compounds by a liquid chromatography/mass spectrometry method. Objectives and methods .—The objectives of this study were to (1) determine the characteristics of laboratory contamination over time, (2) compare distributions of pesticide results in set blanks with distributions in environmental samples, (3) evaluate the potential for false-positive and false-negative reporting of results, and (4) evaluate the effects of reevaluating historical pesticide results using 2017 compound identification protocols on detections of pesticides in groundwater and surface-water samples. The 113 instrument batches selected for this study contained detections of one or more of the 21 pesticide compounds in set blanks or were among those batches with the highest pesticide detection frequencies in set blanks. As a result, the dataset for this study was targeted toward pesticides and batches with laboratory contamination. The objectives were addressed by statistically comparing environmental and set-blank results; computing moving averages of set-blank detection frequencies to identify periods of episodic contamination; and using summary statistics, tabular summaries, and graphical approaches, such as time-series plots and cumulative distribution functions. Results .— Objective 1 : Laboratory contamination, as determined by pesticide detections in set blanks, was found in 13 percent of set-blank results from the 113 targeted batches included in this study (as compared to 6 percent of set-blank results from all 7,620 batches analyzed during the study period). It is estimated that 92 percent of the laboratory contamination during the study period was episodic, meaning that it occurred during discrete periods of time. All 21 of the targeted pesticide compounds had periods of episodic contamination, with most episodes ranging in duration from about 1 to 8 months. The remaining 8 percent of laboratory contamination was random or from a known source (deterministic). Objective 2 : For some compounds, graphs of cumulative distribution functions of the entire distributions of set-blank and environmental samples overlap, suggesting that there is no difference in the distributions of the two types of samples. However, time-series graphs show that detections in set blanks often occur at different times (sometimes separated by years) than detections in environmental samples, indicating clear differences in those distributions, and indicating the importance of evaluating the timing of detections in all sample types. For most compounds detected in set-blank and environmental samples, detection frequencies were significantly greater in set blanks than in groundwater or surface-water samples ( p <0.05). There are several explanations for this finding, including that the 113 batches of samples chosen for this study targeted batches with detections in set blanks or that detections in set-blank samples were historically determined with less stringent identification criteria than for environmental samples (groundwater and surface-water samples). Objective 3 : The false-positive and false-negative rates from blind samples submitted during the study period by the USGS Quality Systems Branch generally were less than 1 and 5 percent, respectively, for the 21 pesticides. The only compound with a false-positive rate greater than 1 percent was flumetsulam (2.6 percent), indicating that there is a higher likelihood of flumetsulam being reported as a detection when it is not present in an environmental sample compared with the reporting of other compounds. Objective 4 : Altogether, for data in targeted batches, NWQL would have reported 0.1 percent of results from groundwater samples and 1.4 percent of results from surface-water samples differently if 2017 identification protocols were applied to historical pesticide results. In most of these cases, detections observed in historical results would change to nondetections. The small percentages of changes that would occur if historical data were reevaluated indicate that historical protocols used by the NWQL to identify detections in environmental samples were robust and produced results that are predominantly consistent with current [2017] practices. Conclusions .—The NWQL produces high-quality pesticide results at environmentally relevant concentrations. NWQL identification protocols and censoring practices are largely effective at minimizing the reporting of false-positive and false-negative results. Laboratory contamination, when it occurred, tended to occur in episodes; thus, evaluating the timing and magnitude of detections in set blanks relative to detections in environmental samples was determined to be an important consideration for analysis of environmental results. Because NWQL censoring practices do not address all types and occurrences of laboratory contamination, options for additional censoring practices are provided for data users with more specific or stringent data-quality objectives. The methods used to analyze the 21 compounds for this report can similarly be applied to all 173 pesticide compounds that were analyzed by the NWQL during the same time period. This study also has helped to identify potential improvements in reporting USGS data, such as conducting more frequent review of set-blank datasets.

Scientific Investigations Report↗

USGS2017

Estimating the high-arsenic domestic-well population in the conterminous United States

Arsenic concentrations from 20 450 domestic wells in the U.S. were used to develop a logistic regression model of the probability of having arsenic >10 μg/L (“high arsenic”), which is presented at the county, state, and national scales. Variables representing geologic sources, geochemical, hydrologic, and physical features were among the significant predictors of high arsenic. For U.S. Census blocks, the mean probability of arsenic >10 μg/L was multiplied by the population using domestic wells to estimate the potential high-arsenic domestic-well population. Approximately 44.1 M people in the U.S. use water from domestic wells. The population in the conterminous U.S. using water from domestic wells with predicted arsenic concentration >10 μg/L is 2.1 M people (95% CI is 1.5 to 2.9 M). Although areas of the U.S. were underrepresented with arsenic data, predictive variables available in national data sets were used to estimate high arsenic in unsampled areas. Additionally, by predicting to all of the conterminous U.S., we identify areas of high and low potential exposure in areas of limited arsenic data. These areas may be viewed as potential areas to investigate further or to compare to more detailed local information. Linking predictive modeling to private well use information nationally, despite the uncertainty, is beneficial for broad screening of the population at risk from elevated arsenic in drinking water from private wells.

Environmental Science & Technology↗

USGS2017

Nutrient and pesticide contamination bias estimated from field blanks collected at surface-water sites in U.S. Geological Survey Water-Quality Networks, 2002–12

Potential contamination bias was estimated for 8 nutrient analytes and 40 pesticides in stream water collected by the U.S. Geological Survey at 147 stream sites from across the United States, and representing a variety of hydrologic conditions and site types, for water years 2002–12. This study updates previous U.S. Geological Survey evaluations of potential contamination bias for nutrients and pesticides. Contamination is potentially introduced to water samples by exposure to airborne gases and particulates, from inadequate cleaning of sampling or analytic equipment, and from inadvertent sources during sample collection, field processing, shipment, and laboratory analysis. Potential contamination bias, based on frequency and magnitude of detections in field blanks, is used to determine whether or under what conditions environmental data might need to be qualified for the interpretation of results in the context of comparisons with background levels, drinking-water standards, aquatic-life criteria or benchmarks, or human-health benchmarks. Environmental samples for which contamination bias as determined in this report applies are those from historical U.S. Geological Survey water-quality networks or programs that were collected during the same time frame and according to the same protocols and that were analyzed in the same laboratory as field blanks described in this report. Results from field blanks for ammonia, nitrite, nitrite plus nitrate, orthophosphate, and total phosphorus were partitioned by analytical method; results from the most commonly used analytical method for total phosphorus were further partitioned by date. Depending on the analytical method, 3.8, 9.2, or 26.9 percent of environmental samples, the last of these percentages pertaining to all results from 2007 through 2012, were potentially affected by ammonia contamination. Nitrite contamination potentially affected up to 2.6 percent of environmental samples collected between 2002 and 2006 and affected about 3.3 percent of samples collected between 2007 and 2012. The percentages of environmental samples collected between 2002 and 2011 that were potentially affected by nitrite plus nitrate contamination were 7.3 for samples analyzed with the low-level method and 0.4 for samples analyzed with the standard-level method. These percentages increased to 14.8 and 2.2 for samples collected in 2012 and analyzed using replacement low- and standard-level methods, respectively. The maximum potentially affected concentrations for nitrite and for nitrite plus nitrate were much less than their respective maximum contamination levels for drinking-water standards. Although contamination from particulate nitrogen can potentially affect up to 21.2 percent and that from total Kjeldahl nitrogen can affect up to 16.5 percent of environmental samples, there are no critical or background levels for these substances. For total nitrogen, orthophosphate, and total phosphorus, contamination in a small percentage of environmental samples might be consequential for comparisons relative to impairment risks or background levels. At the low ends of the respective ranges of impairment risk for these nutrients, contamination in up to 5 percent of stream samples could account for at least 23 percent of measured concentrations of total nitrogen, for at least 40 or 90 percent of concentrations of orthophosphate, depending on the analytical method, and for 31 to 76 percent of concentrations of total phosphorus, depending on the time period. Twenty-six pesticides had no detections in field blanks. Atrazine with 12 and metolachlor with 11 had the highest number of detections, mostly occurring in spring or early summer. At a 99-percent level of confidence, contamination was estimated to be no greater than the detection limit in at least 98 percent of all samples for 38 of 40 pesticides. For metolachlor and atrazine, potential contamination was no greater than 0.0053 and 0.0093 micrograms per liter in 98 percent of samples. For 11 of 14 pesticides with at least one detection, the maximum potentially affected concentration of the environmental sample was less than their respective human-health or aquatic-life benchmarks. Small percentages of environmental samples had concentrations high enough that atrazine contamination potentially could account for the entire aquatic-life benchmark for acute effects on nonvascular plants, that dieldrin contamination could account for up to 100 percent of the cancer health-based screening level, or that chlorpyrifos contamination could account for 13 or 12 percent of the concentrations in the aquatic-life benchmarks for chronic effects on invertebrates or the criterion continuous concentration for chronic effects on aquatic life.

Scientific Investigations Report↗

USGS2016

Concentration, flux, and trend estimates with uncertainty for nutrients, chloride, and total suspended solids in tributaries of Lake Champlain, 1990–2014

The U.S. Geological Survey, in cooperation with the New England Interstate Water Pollution Control Commission and the Vermont Department of Environmental Conservation, estimated daily and 9-month concentrations and fluxes of total and dissolved phosphorus, total nitrogen, chloride, and total suspended solids from 1990 (or first available date) through 2014 for 18 tributaries of Lake Champlain. Estimates of concentration and flux, provided separately in Medalie (2016), were made by using the Weighted Regressions on Time, Discharge, and Season (WRTDS) regression model and update previously published WRTDS model results with recent data. Assessment of progress towards meeting phosphorus-reduction goals outlined in the Lake Champlain management plan relies on annual estimates of phosphorus flux. The percent change in annual concentration and flux is provided for two time periods. The R package EGRETci was used to estimate the uncertainty of the trend estimate. Differences in model specification and function between this study and previous studies that used WRTDS to estimate concentration and flux using data from Lake Champlain tributaries are described. Winter data were too sparse and nonrepresentative to use for estimates of concentration and flux but were sufficient for estimating the percentage of total annual flux over the period of record. Median winter-to-annual fractions ranged between 21 percent for total suspended solids and 27 percent for dissolved phosphorus. The winter contribution was largest for all constituents from the Mettawee River and smallest from the Ausable River. For the full record (1991 through 2014 for total and dissolved phosphorus and chloride and 1993 through 2014 for nitrogen and total suspended solids), 6 tributaries had decreasing trends in concentrations of total phosphorus, and 12 had increasing trends; concentrations of dissolved phosphorus decreased in 6 and increased in 8 tributaries; fluxes of total phosphorus decreased in 5 and increased in 10 tributaries; and fluxes of dissolved phosphorus decreased in 4 and increased in 10 tributaries (where the number of increasing and decreasing trends does not add up to 18, the remainder of tributaries had no trends). Concentrations and fluxes of nitrogen decreased in 10 and increased in 4 tributaries and of chloride decreased in 2 and increased in 15 tributaries. Concentrations of total suspended solids decreased in 4 and increased in 8 tributaries, and fluxes of total suspended solids decreased in 3 and increased in 11 tributaries. Although time intervals for the percent changes from this report are not completely synchronous with those from previous studies, the numbers of and specific tributaries with overall negative percent changes in concentration and flux are similar. Concentration estimates of total phosphorus in the Winooski River were used to trace whether changes in trends between a previous study and the current study were due generally to differences in model specifications or differences from 4 years of additional data. The Winooski River analysis illustrates several things: that keeping all model specifications equal, concentration estimates increased from 2010 to 2014; the effects of a smoothing algorithm used in the current study that was not available previously; that narrowing model half-window widths increased year-to-year variations; and that the change from an annual to a 9-month basis by omitting winter estimates changed a few individual points but not the overall shape of the flow-normalized curve. Similar tests for other tributaries showed that the primary effect of differences in model specifications between the previous and current studies was perhaps to increase scatter over time but that changes in trends generally were the result of 4 years of additional data rather than artifacts of model differences.

New York, Vermont↗

USGS2014

Use of acoustic backscatter to estimate continuous suspended sediment and phosphorus concentrations in the Barton River, northern Vermont, 2010-2013

The U.S. Geological Survey, in cooperation with the Vermont Department of Environmental Conservation, investigated the use of acoustic backscatter to estimate concentrations of suspended sediment and total phosphorus at the Barton River near Coventry, Vermont. The hypothesis was that acoustic backscatter—the reflection of sound waves off objects back to the source from which they came—measured by an acoustic Doppler profiler (ADP) and recorded as ancillary data for the calculation of discharge, also could be used to generate a continuous concentration record of suspended sediment and phosphorus at the streamgage, thereby deriving added value from the instrument. Suspended-sediment and phosphorus concentrations are of particular interest in Vermont, where impairment of surface waters by suspended sediments and phosphorus is a major concern. Regression models for estimating suspended-sediment concentrations (SSCs) and total phosphorus concentrations evaluated several independent variables: measured backscatter (MB), water-corrected backscatter (WCB), sediment-corrected backscatter (SCB), discharge, fluid-absorption coefficient, sediment-driven acoustic attenuation coefficient, and discharge hysteresis. The best regression equations for estimating SSC used backscatter as the predictor, reflecting the direct relation between acoustic backscatter and SSC. Backscatter was a better predictor of SSC than discharge in part because hysteresis between SSC and backscatter was less than for SSC and discharge. All three backscatter variables—MB, WCB, and SCB—performed equally as predictors of SSC and phosphorus concentrations at the Barton River site. The similar abilities to predict SSC among backscatter terms may partially be attributed to the low values and narrow range of the sediment-driven acoustic attenuation in the Barton River. The regression based on SCB was selected for estimating SSC because it removes potential bias caused by attenuation and temperature fluctuations. The best regression model for estimating phosphorus concentrations included terms for discharge and discharge hysteresis. The finding that discharge hysteresis was a significant predictor of phosphorus concentrations might be related to preferential sorption of phosphorus to fine-grained sediments, which have been found to be particularly sensitive to hysteresis. Regression models designed to estimate phosphorus concentrations had less predictive power than the models for SSCs. Data from the Barton River did not fully support one of the study’s hypotheses—that backscatter is mostly caused by sands, and attenuation is mostly caused by fines. Sands, fines, and total SSCs in the Barton River all related better to backscatter than to sediment-driven acoustic attenuation. The weak relation between SSC and sediment-driven acoustic attenuation may be related to the low values and narrow range of SSCs and sediment attenuations observed at Barton River. A weak relation between SSC and sediment-driven acoustic attenuation also suggests that the diameters of the fine-sized suspended sediments in the Barton River may be predominantly greater than 20 micrometers (μm). Long-term changes in the particle-size distribution (PSD) were not observed in Barton River; however, some degree of within-storm changes in sediment source and possibly PSD were inferred from the hysteresis between SSC and SCB.

Vermont↗

USGS2014

Concentration and flux of total and dissolved phosphorus, total nitrogen, chloride, and total suspended solids for monitored tributaries of Lake Champlain, 1990-2012

Annual and daily concentrations and fluxes of total and dissolved phosphorus, total nitrogen, chloride, and total suspended solids were estimated for 18 monitored tributaries to Lake Champlain by using the Weighted Regressions on Time, Discharge, and Seasons regression model. Estimates were made for 21 or 23 years, depending on data availability, for the purpose of providing timely and accessible summary reports as stipulated in the 2010 update to the Lake Champlain “Opportunities for Action” management plan. Estimates of concentration and flux were provided for each tributary based on (1) observed daily discharges and (2) a flow-normalizing procedure, which removed the random fluctuations of climate-related variability. The flux bias statistic, an indicator of the ability of the Weighted Regressions on Time, Discharge, and Season regression models to provide accurate representations of flux, showed acceptable bias (less than ±10 percent) for 68 out of 72 models for total and dissolved phosphorus, total nitrogen, and chloride. Six out of 18 models for total suspended solids had moderate bias (between 10 and 30 percent), an expected result given the frequently nonlinear relation between total suspended solids and discharge. One model for total suspended solids with a very high bias was influenced by a single extreme value; however, removal of that value, although reducing the bias substantially, had little effect on annual fluxes.

Lake Champlain↗

USGS2013

High-water marks from flooding in Lake Champlain from April through June 2011 and Tropical Storm Irene in August 2011 in Vermont

The U.S. Geological Survey, in cooperation with the Federal Emergency Management Agency, identified high-water marks after two floods in Vermont during 2011. Following a snowy winter, new monthly precipitation records were set in Burlington, Vermont, in April and May 2011, causing extensive flooding from April through June. The spring 2011 flooding resulted in a new record for stage (103.27 feet, referenced to the National Geodetic Vertical Datum of 1929) at the Lake Champlain at Burlington, Vt., gaging station (04294500). During August 28 and 29, 2011, tropical storm Irene delivered rainfall totals of 3 to more than 7 inches throughout Vermont, which resulted in extensive flooding and new streamflow records at nine streamgaging stations. Four presidential declarations of disaster were made following the 2011 flood events in Vermont. Thirty-nine high-water marks were identified and flagged to mark the highest levels of Lake Champlain from the May 2011 flooding, and 1,138 high-water marks were identified and flagged along Vermont rivers after flooding from tropical storm Irene in August 2011. Seventy-four percent of the high-water marks that were flagged were later found and surveyed to the North American Vertical Datum of 1988.

Vermont↗

USGS2013

High-water marks from tropical storm Irene for selected river reaches in northwestern Massachusetts, August 2011

A Presidential Disaster Declaration was issued for Massachusetts, with a focus on the northwestern counties, following flooding from tropical storm Irene on August 28–29, 2011. Three to 10 inches of rain fell during the storm on soils that were susceptible to flash flooding because of wet antecedent conditions. The gage height at one U.S. Geological Survey (USGS) streamgage rose nearly 20 feet in less than 4 hours because of the combination of saturated soils and intense rainfall. Eight of 16 USGS long-term streamgages in western Massachusetts set new peaks of record on August 28 or 29, 2011. To document the historic water levels of the streamflows from tropical storm Irene, the USGS identified, flagged, and surveyed 323 high-water marks in the Deerfield and Hudson- Hoosic River basins in northwestern Massachusetts. Areas targeted for high-water marks were generally upstream and downstream from structures along selected river reaches. Elevations from high-water marks can be used to confirm peak river stages or help compute peak streamflows, to calibrate hydraulic models, or to update flood-inundation and recovery maps. For areas in western Massachusetts that flooded as a result of tropical storm Irene, high-water marks surveyed for this study have helped to confirm or determine instantaneous peak river gage heights at several USGS streamgages.

Massachusetts↗