Geology topics

Dennis R. Helsel

Publications and source records attributed to Dennis R. Helsel.

At least 19 recordsLinked to original sources

USGS2024

Evaluating the reliability of environmental concentration data to characterize exposure in environmental risk assessments

Environmental risk assessments often rely on measured concentrations in environmental matrices to characterize exposure of the population of interest—typically, humans, aquatic biota, or other wildlife. Yet, there is limited guidance available on how to report and evaluate exposure datasets for reliability and relevance, despite their importance to regulatory decision-making. This paper is the second of a four-paper series detailing the outcomes of a Society of Environmental Toxicology and Chemistry Technical Workshop that has developed Criteria for Reporting and Evaluating Exposure Datasets (CREED). It presents specific criteria to systematically evaluate the reliability of environmental exposure datasets. These criteria can help risk assessors understand and characterize uncertainties when existing data are used in various types of assessments and can serve as guidance on best practice for the reporting of data for data generators (to maximize utility of their datasets). Although most reliability criteria are universal, some practices may need to be evaluated considering the purpose of the assessment. Reliability refers to the inherent quality of the dataset and evaluation criteria address the identification of analytes, study sites, environmental matrices, sampling dates, sample collection methods, analytical method performance, data handling or aggregation, treatment of censored data, and generation of summary statistics. Each criterion is evaluated as “fully met,” “partly met,” “not met or inappropriate,” “not reported,” or “not applicable” for the dataset being reviewed. The evaluation concludes with a scheme for scoring the dataset as reliable with or without restrictions, not reliable, or not assignable, and is demonstrated with three case studies representing both organic and inorganic constituents, and different study designs and assessment purposes. Reliability evaluation can be used in conjunction with relevance evaluation (assessed separately) to determine the extent to which environmental monitoring datasets are “fit for purpose,” that is, suitable for use in various types of assessments. Integr Environ Assess Manag 2024;00:1–23. © 2024 Society of Environmental Toxicology & Chemistry (SETAC). This article has been contributed to by U.S. Government employees and their work is in the public domain in the USA.

Integrated Environmental Assessment and Management↗

USGS2020

Statistical methods in water resources

This text began as a collection of class notes for a course on applied statistical methods for hydrologists taught at the U.S. Geological Survey (USGS) National Training Center. Course material was formalized and organized into a textbook, first published in 1992 by Elsevier as part of their Studies in Environmental Science series. In 2002, the work was made available online as a USGS report. The text has now been updated as a USGS Techniques and Methods Report. It is intended to be a text in applied statistics for hydrology, environmental science, environmental engineering, geology, or biology that addresses distinctive features of environmental data. For example, water resources data tend to have many variables with a lower bound of zero, tend to be more skewed than data from many other disciplines, commonly contain censored data (less than values), and assumptions that the data are normally distributed are not appropriate. Computer-intensive methods (bootstrapping and permutation tests) now improve upon and replace the dependence on t-intervals, t-tests, and analysis of variance. A new chapter on sampling design addresses questions such as “How many observations do I need?” The chapter also presents distribution-free methods to help plan sampling efforts. The trends chapter has been updated to include the WRTDS (Weighted Regressions on Time, Discharge, and Season) method for analysis of water-quality data. This new version contains updated graphics and updated guidance on the use of statistical techniques. The text utilizes R, a programming language and open-source software environment, for all exercises and most graphics, and the R code used to generate figures and examples is provided for download.

Techniques and Methods↗

USGS2011

USGS library for S-PLUS for Windows -- Release 4.0

Release 4.0 of the U.S. Geological Survey S-PLUS library supercedes release 2.1. It comprises functions, dialogs, and datasets used in the U.S. Geological Survey for the analysis of water-resources data. This version does not contain ESTREND, which was in version 2.1. See Release 2.1 for information and access to that version. This library requires Release 8.1 or later of S-PLUS for Windows. S-PLUS is a commercial statistical and graphical analysis software package produced by TIBCO corporation(http://www.tibco.com/). The USGS library is not supported by TIBCO or its technical support staff.

Open-File Report↗

USGS2010

Estimation of distributional parameters for censored trace level water quality data: 1. Estimation techniques

A recurring difficulty encountered in investigations of many metals and organic contaminants in ambient waters is that a substantial portion of water sample concentrations are below limits of detection established by analytical laboratories. Several methods were evaluated for estimating distributional parameters for such censored data sets using only uncensored observations. Their reliabilities were evaluated by a Monte Carlo experiment in which small samples were generated from a wide range of parent distributions and censored at varying levels. Eight methods were used to estimate the mean, standard deviation, median, and interquartile range. Criteria were developed, based on the distribution of uncensored observations, for determining the best performing parameter estimation method for any particular data set. The most robust method for minimizing error in censored-sample estimates of the four distributional parameters over all simulation conditions was the log-probability regression method. With this method, censored observations are assumed to follow the zero-to-censoring level portion of a lognormal distribution obtained by a least squares regression between logarithms of uncensored concentration observations and their z scores. When method performance was separately evaluated for each distributional parameter over all simulation conditions, the log-probability regression method still had the smallest errors for the mean and standard deviation, but the lognormal maximum likelihood method had the smallest errors for the median and interquartile range. When data sets were classified prior to parameter estimation into groups reflecting their probable parent distributions, the ranking of estimation methods was similar, but the accuracy of error estimates was markedly improved over those without classification.

Water Resources Research↗

USGS2010

Estimation of distributional parameters for censored trace level water quality data: 2. Verification and applications

Estimates of distributional parameters (mean, standard deviation, median, interquartile range) are often desired for data sets containing censored observations. Eight methods for estimating these parameters have been evaluated by R. J. Gilliom and D. R. Helsel (this issue) using Monte Carlo simulations. To verify those findings, the same methods are now applied to actual water quality data. The best method (lowest root-mean-squared error (rmse)) over all parameters, sample sizes, and censoring levels is log probability regression (LR), the method found best in the Monte Carlo simulations. Best methods for estimating moment or percentile parameters separately are also identical to the simulations. Reliability of these estimates can be expressed as confidence intervals using rmse and bias values taken from the simulation results. Finally, a new simulation study shows that best methods for estimating uncensored sample statistics from censored data sets are identical to those for estimating population parameters. Thus this study and the companion study by Gilliom and Helsel form the basis for making the best possible estimates of either population parameters or sample statistics from censored water quality data, and for assessments of their reliability.

Water Resources Research↗

USGS2010

Estimation of descriptive statistics for multiply censored water quality data

This paper extends the work of Gilliom and Helsel (1986) on procedures for estimating descriptive statistics of water quality data that contain “less than” observations. Previously, procedures were evaluated when only one detection limit was present. Here we investigate the performance of estimators for data that have multiple detection limits. Probability plotting and maximum likelihood methods perform substantially better than simple substitution procedures now commonly in use. Therefore simple substitution procedures (e.g., substitution of the detection limit) should be avoided. Probability plotting methods are more robust than maximum likelihood methods to misspecification of the parent distribution and their use should be encouraged in the typical situation where the parent distribution is unknown. When utilized correctly, less than values frequently contain nearly as much information for estimating population moments and quantiles as would the same observations had the detection limit been below them.

Water Resources Research↗

USGS2009

Multivariate Statistical Models for Predicting Sediment Yields from Southern California Watersheds

Debris-retention basins in Southern California are frequently used to protect communities and infrastructure from the hazards of flooding and debris flow. Empirical models that predict sediment yields are used to determine the size of the basins. Such models have been developed using analyses of records of the amount of material removed from debris retention basins, associated rainfall amounts, measures of watershed characteristics, and wildfire extent and history. In this study we used multiple linear regression methods to develop two updated empirical models to predict sediment yields for watersheds located in Southern California. The models are based on both new and existing measures of volume of sediment removed from debris retention basins, measures of watershed morphology, and characterization of burn severity distributions for watersheds located in Ventura, Los Angeles, and San Bernardino Counties. The first model presented reflects conditions in watersheds located throughout the Transverse Ranges of Southern California and is based on volumes of sediment measured following single storm events with known rainfall conditions. The second model presented is specific to conditions in Ventura County watersheds and was developed using volumes of sediment measured following multiple storm events. To relate sediment volumes to triggering storm rainfall, a rainfall threshold was developed to identify storms likely to have caused sediment deposition. A measured volume of sediment deposited by numerous storms was parsed among the threshold-exceeding storms based on relative storm rainfall totals. The predictive strength of the two models developed here, and of previously-published models, was evaluated using a test dataset consisting of 65 volumes of sediment yields measured in Southern California. The evaluation indicated that the model developed using information from single storm events in the Transverse Ranges best predicted sediment yields for watersheds in San Bernardino, Los Angeles, and Ventura Counties. This model predicts sediment yield as a function of the peak 1-hour rainfall, the watershed area burned by the most recent fire (at all severities), the time since the most recent fire, watershed area, average gradient, and relief ratio. The model that reflects conditions specific to Ventura County watersheds consistently under-predicted sediment yields and is not recommended for application. Some previously-published models performed reasonably well, while others either under-predicted sediment yields or had a larger range of errors in the predicted sediment yields.

Open-File Report↗

USGS2008

Using logistic regression to predict a probability of debris flows in areas burned by wildfires, southern California, 2003-2006

Logistic regression was used to develop statistical models that can be used to predict the probability of debris flows in areas recently burned by wildfires by using data from 14 wildfires that burned in southern California during 2003-2006. Twenty-eight independent variables describing the basin morphology, burn severity, rainfall, and soil properties of 306 drainage basins located within those burned areas were evaluated. The models were developed as follows: (1) Basins that did and did not produce debris flows soon after the 2003 to 2006 fires were delineated from data in the National Elevation Dataset using a geographic information system; (2) Data describing the basin morphology, burn severity, rainfall, and soil properties were compiled for each basin. These data were then input to a statistics software package for analysis using logistic regression; and (3) Relations between the occurrence or absence of debris flows and the basin morphology, burn severity, rainfall, and soil properties were evaluated, and five multivariate logistic regression models were constructed. All possible combinations of independent variables were evaluated to determine which combinations produced the most effective models, and the multivariate models that best predicted the occurrence of debris flows were identified. Percentage of high burn severity and 3-hour peak rainfall intensity were significant variables in all models. Soil organic matter content and soil clay content were significant variables in all models except Model 5. Soil slope was a significant variable in all models except Model 4. The most suitable model can be selected from these five models on the basis of the availability of independent variables in the particular area of interest and field checking of probability maps. The multivariate logistic regression models can be entered into a geographic information system, and maps showing the probability of debris flows can be constructed in recently burned areas of southern California. This study demonstrates that logistic regression is a valuable tool for developing models that predict the probability of debris flows occurring in recently burned landscapes.

California↗

USGS2006

Computer program for the Kendall family of trend tests

The Seasonal Kendall (SK) test for trend was developed by the U.S. Geological Survey and has become the most frequently used test for trend in the environmental sciences. Recently the test was modified to form the Regional Kendall (RK) test for trend. In this form, trends at numerous locations within a region are tested to determine whether the direction of trend is consistent across the entire region. Computer code developed at the USGS in the 1980s to perform the SK test is no longer widely available. Other versions written by other scientists may or may not be easily available, and may require commercial software in order to be run. These other versions do not explicitly compute the RK test. Therefore, the original code for computing the SK test has been repackaged into a program that runs under the Windows operating system. This program may be used to verify that other implementations of the test give the same results as the original. The program also provides a means for computing the RK test and the simpler Mann-Kendall test for trend.

Scientific Investigations Report↗

USGS2005

Arsenic in ground water of the United States: occurrence and geochemistry

Concentrations of naturally occurring arsenic in ground water vary regionally due to a combination of climate and geology. Although slightly less than half of 30,000 arsenic analyses of ground water in the United States were 1 μg/L, about 10% exceeded 10 μg/L. At a broad regional scale, arsenic concentrations exceeding 10 μg/L appear to be more frequently observed in the western United States than in the eastern half. Arsenic concentrations in ground water of the Appalachian Highlands and the Atlantic Plain generally are very low ( 1 μg/L). Concentrations are somewhat greater in the Interior Plains and the Rocky Mountain System. Investigations of ground water in New England, Michigan, Minnesota, South Dakota, Oklahoma, and Wisconsin within the last decade suggest that arsenic concentrations exceeding 10 μg/L are more widespread and common than previously recognized. Arsenic release from iron oxide appears to be the most common cause of widespread arsenic concentrations exceeding 10 μg/L in ground water. This can occur in response to different geochemical conditions, including release of arsenic to ground water through reaction of iron oxide with either natural or anthropogenic (i.e., petroleum products) organic carbon. Iron oxide also can release arsenic to alkaline ground water, such as that found in some felsic volcanic rocks and alkaline aquifers of the western United States. Sulfide minerals are both a source and sink for arsenic. Geothermal water and high evaporation rates also are associated with arsenic concentrations 10g/L in ground and surface water, particularly in the west. Arsenic release from iron oxide appears to be the most common cause of widespread arsenic concentrations exceeding 10 µg/L a ground water. This can occur in response to different geochemical conditions, including release of arsenic to ground water through reaction of iron oxide with either natural or anthropogenic (i.e., petroleum products) organic carbon. Iron oxide also can release arsenic to alkaline ground water, such as that found in some felsic volcanic rocks and alkaline aquifers of the Western United States. Sulfide minerals are both a source and sink for arsenic. Geothermal water and high evaporation rates also are associated with arsenic concentrations ≥ 10g/L in ground and surface water, particularly in the west.

Ground Water↗

USGS2005

Evaluating regional trends in ground-water nitrate concentrations of the Columbia Basin ground water management area, Washington

Trends in nitrate concentrations in water from 474 wells in 17 subregions in the Columbia Basin Ground Water Management Area (GWMA) in three counties in eastern Washington were evaluated using a variety of statistical techniques, including the Friedman test and the Kendall test. The Kendall test was modified from its typical 'seasonal' version into a 'regional' version by using well locations in place of seasons. No statistically significant trends in nitrate concentrations were identified in samples from wells in the GWMA, the three counties, or the 17 subregions from 1998 to 2002 when all data were included in the analysis. For wells in which nitrate concentrations were greater than 10 milligrams per liter (mg/L), however, a significant downward trend of -0.4 mg/L per year was observed between 1998 and 2002 for the GWMA as a whole, as well as for Adams County (-0.35 mg/L per year) and for Franklin County (-0.46 mg/L per year). Trend analysis for a smaller but longer-term 51-well dataset in Franklin County found a statistically significant upward trend in nitrate concentrations of 0.1 mg/L per year between 1986 and 2003. The largest increase of nitrate concentrations occurred between 1986 and 1991. No statistically significant differences were observed in this dataset between 1998 and 2003 indicating that the increase in nitrate concentrations has leveled off.

Scientific Investigations Report↗

USGS2002

Statistical methods in water resources

Preface This book began as class notes for a course we teach on applied statistical methods to hydrologists of the Water Resources Division, U. S. Geological Survey (USGS). It reflects our attempts to teach statistical methods which are appropriate for analysis of water resources data. As interest in this course has grown outside of the USGS, incentive grew to develop the material into a textbook. The topics covered are those we feel are of greatest usefulness to the practicing water resources scientist. Yet all topics can be directly applied to many other types of environmental data. This book is not a stand-alone text on statistics, or a text on statistical hydrology. For example, in addition to this material we use a textbook on introductory statistics in the USGS training course. As a consequence, discussions of topics such as probability theory required in a general statistics textbook will not be found here. Derivations of most equations are not presented. Important tables included in all general statistics texts, such as quantiles of the normal distribution, are not found here. Neither are details of how statistical distributions should be fitted to flood data -- these are adequately covered in numerous books on statistical hydrology. We have instead chosen to emphasize topics not always found in introductory statistics textbooks, and often not adequately covered in statistical textbooks for scientists and engineers. Tables included here, for example, are those found more often in books on nonparametric statistics than in books likely to have been used in college courses for engineers. This book points the environmental and water resources scientist to robust and nonparametric statistics, and to exploratory data analysis. We believe that the characteristics of environmental (and perhaps most other 'real') data drive analysis methods towards use of robust and nonparametric methods. Exercises are included at the end of chapters. In our course, students compute each type of analysis (t-test, regression, etc.) the first time by hand. We choose the smaller, simpler examples for hand computation. In this way the mechanics of the process are fully understood, and computer software is seen as less mysterious. We wish to acknowledge and thank several other scientists at the U. S. Geological Survey for contributing ideas to this book. In particular, we thank those who have served as the other instructors at the USGS training course. Ed Gilroy has critiqued and improved much of the material found in this book. Tim Cohn has contributed in several areas, particularly to the sections on bias correction in regression, and methods for data below the reporting limit. Richard Alexander has added to the trend analysis chapter, and Charles Crawford has contributed ideas for regression and ANOVA. Their work has undoubtedly made its way into this book without adequate recognition. Professor Ken Potter (University of Wisconsin) and Dr. Gary Tasker (USGS) reviewed the manuscript, spending long hours with no reward except the knowledge that they have improved the work of others. For that we are very grateful. We also thank Madeline Sabin, who carefully typed original drafts of the class notes on which the book is based. As always, the responsibility for all errors and slanted thinking are ours alone.

Techniques of Water-Resources Investigations↗

USGS2002

Assessing ground-water vulnerability to contamination: Providing scientifically defensible information for decision makers

Throughout the United States increasing demands for safe drinking water and requirements to maintain healthy ecosystems are leading policy makers to ask complex social and scientific questions about how to assess and manage our water resources. This challenge becomes particularly difficult as policy and management objectives require scientific assessments of the potential for ground-water resources to become contaminated from anthropogenic, as well as natural sources of contamination. Assessments of the vulnerability of ground water to contamination range in scope and complexity from simple, qualitative, and relatively inexpensive approaches to rigorous, quantitative, and costly assessments. Tradeoffs must be carefully considered among the competing influences of the cost of an assessment, the scientific defensibility, and the amount of acceptable uncertainty in meeting the objectives of the water-resource decision maker.

Circular↗

USGS2001

A national pilot study of mercury contamination of aquatic ecosystems along multiple gradients" Bioaccumulation in fish

Water, sediment, and fish were sampled in the summer and fall of 1998 at 106 sites from 20 U.S. watershed basins to examine relations of mercury (Hg) and methylmercury (MeHg) in aquatic ecosystems. Bioaccumulation of Hg in fish from these basins was evaluated in relation to species, Hg and MeHg in surficial sediment and water, and watershed characteristics. Bioaccumulation was strongly (positively) correlated with MeHg in water (r = 0.63, p < 0.001) but only moderately with the MeHg in sediment (r = 0.33, p < 0.001) or total Hg in water (r = 0.28, p < 0.01). Of the other significantly measured parameters, pH, DOC, sulfate, sediment LOI, and the percent wetlands of each basin were also significantly correlated with Hg bioaccumulation in fish. The best model for predicting Hg bioaccumulation included Me Hg in water, PH of the water, % wetlands in the basin, and the AVS content of the sediment. These four variables accounted for 45% of the variability of the fish fillet Hg concentration normalized (divided) by total length; however, the majority was described by MeHg in water. A MeHg water concentration 0.12 ng/L was on average, associated with a fish fillet Hg concentration of 0.3 mg/kg wet weight for an age-3 fish when all species were considered. For age-3 largemouth bass, a MeHg water concentration of 0.058 ng/L was associated with the 0.3 mg/kg fillet concentration. Based on rankings for Hg in sediment, water, and fish, sampling sites from the following five study basins had the greatest Hg contamination: Nevada Basin and Range, South Florida Basin, Sacramento River Basin (California), Santee River Basin and Caostal Drainages (South Carolina), and the Long Island and New Jersey Coastal Drainags. A sampling and analysis strategy based on this pilot study is planned for all USGS/NAWQA study units over the next decade.

Biological Science Report↗

USGS2000

Occurrence and distribution of microbiological indicators in groundwater and stream water

A total of 136 stream water and 143 groundwater samples collected in five important hydrologic systems of the United States were analyzed for microbiological indicators to test monitoring concepts in a nationally consistent program. Total coliforms were found in 99%, Escherichia coli in 97%, and Clostridium perfringens in 73% of stream water samples analyzed for each bacterium. Total coliforms were found in 20%, E. coli in less than 1%, and C. perfringens in none of the groundwater samples analyzed for each bacterium. Although coliphage analyses were performed on many of the samples, contamination in the laboratory and problems discerning discrete plaques precluded quantification. Land use was found to have the most significant effect on concentrations of bacterial indicators in stream water. Presence of septic systems on the property near the sampling site and well depth were found to be related to detection of coliforms in groundwater, although these relationships were not statistically significant. A greater diversity of sites, more detailed information about some factors, and a larger dataset may provide further insight to factors that affect microbiological indicators.

Water Environment Research↗

USGS2000

Arsenic in Ground-Water Resources of the United States

Arsenic is a naturally occurring element in rocks, soils, and the waters in contact with them. Recognized as a toxic element for centuries, arsenic today also is a human health concern because it can contribute to skin, bladder, and other cancers (National Research Council, 1999). Recently, the National Research Council (1999) recommended lowering the current maximum contaminant level (MCL) allowed for arsenic in drinking water of 50 ?g/L (micrograms per liter), citing risks for developing bladder and other cancers. The U.S. Environmental Protection Agency (USEPA) will propose a new, and likely lower, arsenic MCL during 2000 (U.S. Environmental Protection Agency, 2000). This fact sheet provides information on where and to what extent natural concentrations of arsenic in ground water exceed possible new standards. The U.S. Geological Survey (USGS) has collected and analyzed arsenic in potable (drinkable) water from 18,850 wells in 595 counties across the United States during the past two decades. These wells are used for irrigation, industrial purposes, and research, as well as for public and private water supply. Arsenic concentrations in samples from these wells are similar to those found in nearby public supplies (see Focazio and others, 1999). The large number of samples, broad geographic coverage, and consistency of methods produce a more accurate and detailed picture of arsenic concentrations than provided by any previous studies.

Fact Sheet↗

USGS2000

Microbiological monitoring for the U.S. Geological Survey National Water-Quality Assessment Program

Data to characterize the microbiological quality of the Nation's fresh, marine, and estuarine waters are usually collected for local purposes, most often to judge compliance with standards for protection of public health in swimmable or drinkable waters. Methods and procedures vary with the objectives and practices of the parties collecting data and are continuously being developed or modified. Therefore, it is difficult to provide a nationally consistent picture of the microbial quality of the Nation's waters. Study objectives and guidelines for a national microbiological monitoring program are outlined in this report, using the framework of the U.S. Geological Survey (USGS) National Water-Quality Assessment (NAWQA) program. A national program is designed to provide long-term data on the presence of microbiological pathogens and indicators in ground water and surface water to support effective water policy and management. Three major groups of waterborne pathogens affect the public health acceptability of waters in the United States—bacteria, protozoa, and viruses. Microbiological monitoring in NAWQA would be designed to assess the occurrence, distribution, and trends of pathogenic organisms and indicators in surface waters and ground waters; relate the patterns discerned to factors that help explain them; and improve our understanding of the processes that control microbiological water quality.

Water-Resources Investigations Report↗

USGS2000

A Retrospective Analysis on the Occurrence of Arsenic in Ground-Water Resources of the United States and Limitations in Drinking-Water-Supply Characterizations

The Safe Drinking Water Act, as amended in 1996, requires the U.S. Environmental Protection Agency (USEPA) to review current drinking-water standards for arsenic, propose a maximum contaminant level for arsenic by January 1, 2000, and issue a final regulation by January, 2001. Quantification of the national occurrence of targeted ranges in arsenic concentration in ground water used for public drinking-water supplies is an important component of USEPA's regulatory process. Data from the U.S. Geological Survey (USGS) National Water Information System (NWIS) were used in a retrospective analysis of arsenic in the ground-water resources of the United States. The analysis augments other existing sources of data on the occurrence of arsenic collected in ground water at public water-supply systems.The USGS, through its District offices and national programs, has been compiling data for many years on arsenic concentrations collected from wells used for public water supply, research, agriculture, industry, and domestic water supply throughout the United States. These data have been collected for a variety of purposes ranging from simple descriptions of the occurrence of arsenic in local or regional ground-water resources to detailed studies on arsenic geochemistry associated with contamination sites. A total of 18,864 sample locations were selected from the USGS NWIS data base regardless of well type, of which 2,262 were taken from public water-supply sources. Samples with non-potable water (dissolved-solids concentration greater than 2,000 milligrams per liter and water temperature greater than 50o Celsius) were not selected for the retrospective analysis and other criteria for selection included the amount and type of ancillary data available for each sample. The 1,528 counties with sufficient data included 76 percent of all large public water-supply systems (serving more than 10,000 people) and 61 percent of all small public water-supply systems (serving more than 1,000 and less than 10,000 people) in the United States. The arsenic data were summarized for the selected counties by associating the arsenic concentrations measured in the ground-water resource with the numbers and sizes of public water-supply systems using ground water in those counties. Targeted arsenic concentrations of 1, 2, 5, 10, 20, and 50 ug/L were exceeded in the ground-water resource associated with 36, 25, 14, 8, 3, and 1 percent respectively of all public water-supply systems accounted for in the analysis.Contributions to uncertainty such as changes in sampling methods and changes in laboratory reporting appear to be less important to the national occurrence estimates than other factors such as temporal variability in arsenic concentrations at a given well, the types of wells sampled, and density and types of sampling locations. In addition, no attempt was made to quantify arsenic concentrations in relation to depth within aquifers. With these qualifications, the USGS data represent the ground-water resource in general and are not restricted to wells currently used for public drinking-water sources. In this way, the broad spatial extent, large number of water samples, and low detection limits used for the USGS data provide a unique source of information to determine where targeted concentrations of arsenic are likely to occur in the ground-water resources within much of the United States.These results indicate USGS data can be effectively used to augment national estimates of arsenic occurrence in the nation's ground-water resources if limitations are recognized. Existing estimates of the occurrence of arsenic in ground water that are used as a source of drinking water can be supplemented with the USGS arsenic concentration data when associated with the public water-supply data base. One such supplementary application is the additional insight gained by establishing relations between arsenic concentration data in the ground-water resource and small public wat

Water-Resources Investigations Report↗