SEARCH · Search USGS

Results for “Scientific Investigations Report”

Search indexed USGS publications on groundwater, aquifers, geologic maps, mineral resources and earthquakes. Explore source records by subject and place.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,135 records · Page 63Linked to original sources

USGS2016

Social-value maps for Arapaho, Roosevelt, Medicine Bow, Routt, and White River National Forests, Colorado and Wyoming

Executive Summary The continued pressures of population growth on the life-sustaining, economic, and cultural ecosystem services provided by our national forests, particularly those located near rapidly growing urban areas, present ongoing challenges to forest managers. Achieving an effective assessment of these ecosystem services includes a proper accounting of the ecological, economic, and social values attributable to them. However, assessments of ecosystem goods and services notably lack information describing the spatial distribution and relative intensity of social values—the perceived, nonmarket values derived particularly from cultural ecosystem services. A geographic information system (GIS) tool developed to fill this need, Social Values for Ecosystem Services (SolVES; http://solves.cr.usgs.gov ), now provides the capability to generate social-value maps at a range of spatial scales. This report presents some of the methods behind SolVES, procedures needed to apply the tool, the first formal map products resulting from its application at a regional scale, and a discussion of the management implications associated with this type of information. In this study, we use SolVES to identify the location and relative intensity of social values as derived from survey responses gathered from residents living in counties adjacent to Arapaho, Roosevelt, Medicine Bow, Routt, and White River National Forests. The results, presented as a series of social-value maps, represent the first publicly available spatial data on social-value intensity for the southern Rocky Mountain region. Our analysis identified high-value areas for social values including aesthetic, biodiversity, and life sustaining within wilderness areas. Other values, like recreation, show high-value areas both within wilderness and throughout the general forest areas, which can be attributed to people using the forests for a diverse set of recreational activities. The economic social-value type was lower within wilderness areas, which was an expected outcome because of the restrictions inside wilderness areas that preclude resource extraction, development, and motorized or mechanized recreation. Providing spatially explicit social-value information collected from residents in counties adjacent to these national forests can assist in facilitating the U.S. Department of Agriculture Forest Service’s Multiple-Use Sustained-Yield Act of 1960 (16 U.S.C. 528) note and the 2012 Forest Service Planning Rule (36 CFR Part 219), which encourage public participation in planning and management. By making these maps available to the public, we hope to encourage exploration of potential uses of these data for resource management and planning.

Colorado, Wyoming↗

USGS2020

Sediment and chemical contaminant loads in tributaries to the Anacostia River, Washington, District of Columbia, 2016–17

A study was conducted by the U.S. Geological Survey (USGS) in cooperation with the Washington, D.C., Department of Energy & Environment to estimate the loads of suspended-sediment-bound chemical compounds in five gaged tributaries and four ungaged tributaries of the Anacostia River (known locally as “Lower Anacostia River”) in Washington, D.C. Tributaries whose discharge is measured by the USGS are the Northeast and Northwest Branches of the Anacostia River, referred to in this report as “Northeast Branch” (NEB) and “Northwest Branch” (NWB), respectively; Watts Branch (WB); and Hickey Run (HR). A USGS streamflow-gaging station was established in 2016 on Beaverdam Creek (known locally as “Lower Beaverdam Creek” [LBDC]) to support this study. The ungaged streams studied include Nash Run; Pope Branch; an unnamed stream at Fort DuPont, referred to in this report as “Fort DuPont Creek”; and an unnamed stream at Fort Stanton, referred to in this report as “Fort Stanton Creek.” The gaged streams were sampled during four to five storms and two low-flow events during January, March, May, and July 2017. The ungaged streams were sampled during one storm and one low-flow event during July 2017. Storm sampling involved collecting large-volume (60- to 70-liter) composite samples, then removing sediment by filtration in the laboratory. Low-flow samples were obtained by filtering streamwater directly in the field. Continuously recording data sondes were deployed throughout the study to measure turbidity and other water-quality characteristics. During sampling, multiple discrete samples of streamwater were collected to determine suspended-sediment concentration (SSC) and particulate organic carbon (POC) concentration. Shortly after each storm, bed sediment was collected for chemical analysis. Sediment samples were analyzed for 209 polychlorinated biphenyl (PCB) congeners; 35 polyaromatic hydrocarbon (PAH) compounds, including 20 nonalkylated and 15 alkylated species; and 20 organochlorine pesticide (OP) compounds. Sediment from one storm was analyzed for 23 metals. Relations were developed among turbidity, discharge, and measured SSC by using multiple linear regression of log-transformed data. These relations were used to estimate SSC from continuous records of discharge and turbidity and were subsequently used to estimate sediment loads for the 2017 calendar year. USGS continuous records of turbidity in NEB, NWB, Watts Branch, and Hickey Run were available for 2013–17, which allowed sediment loads to be calculated for these years. Sediment loads for the ungaged streams were estimated by using loads measured in Watts Branch adjusted on the basis of stream-basin areas. Sediment loads for 2017 total 3.10×10 7 kilograms (kg), with 1.02×107 kg (33 percent of total) from the NEB, 1.55×10 7 kg (50 percent) from the NWB, 4.45×10 6 kg (14 percent) from LBDC, 5.62×10 5 kg (2 percent) from Watts Branch, and 2.82×10 5 kg (1 percent) from Hickey Run. Sediment yields were highest from NWB and LBDC (3.13×10 5 kilograms per year per square mile [kg/yr/mi 2 ] and 3.01 kg/yr/mi 2 , respectively). As a result of gaps in turbidity and discharge data, the load for LBDC reported here was calculated from measurements representing only 88 percent of the year (2017), and thus underestimates the actual load. All other gaged tributaries had datasets covering 100 percent of the year and are considered to fully represent actual loads. Estimated sediment loads for the ungaged streams during 2017 total 3.5×10 5 kg, with 1.2×10 5 kg from Nash Run, 6.2×10 4 kg from Pope Branch, 1.1×10 5 kg from Fort DuPont Creek, and 5.6×10 4 kg from Fort Stanton Creek. Concentrations of PCBs, PAHs, and chlorinated pesticides in streamwater are presented for stormflow and low-flow conditions. Average concentrations (in stormflow and low-flow samples) of total PCBs (sum of all congeners, including coelutions) are 5.9 micrograms per kilogram (µg/kg) for NEB, 6.6 µg/kg for NWB, 130 µg/kg for LBDC, 34 µg/kg for Watts Branch, and 69 µg/kg for Hickey Run. Average concentrations of total PAHs (tPAH) (total of nonalkylated and alkylated species) are 2,000 µg/kg for NEB, 3,300 µg/kg for NWB, 2,200 µg/kg for LBDC, 2,400 µg/kg for Watts Branch, and 18,000 µg/kg for Hickey Run. tPAH concentrations among the ungaged streams were highest in Nash Run (5,500 µg/kg); concentrations in the other ungaged streams were less than (<) 700 µg/kg. The general magnitude of tPCB and tPAH concentrations in streamwater samples was low-flow samples greater than (>) stormflow samples greater than or equal to (≥) bed-sediment samples. PCB congener profiles in the three types of samples were nearly identical in each stream and were similar in all streams except for LBDC, where the dominant PCBs shifted to the lighter di- through tetra- homologs. LBDC showed higher tPCB concentrations and a distinct congener profile from the other streams. The similarity in congener makeup supported that averaging PCB concentrations in stormflow and low-flow samples was appropriate for calculating chemical loads. Loads of tPCB, tPAH (total of alkylated and nonalkylated forms), and pesticides were estimated for each stream by multiplying average contaminant concentrations by the respective sediment loads. Total PCB loads for 2017 were estimated to be 820 grams (g) with 8 percent (60 g) from NEB, 12 percent (95 g) from NWB, 75 percent (590 g) from LBDC, 3 percent (25 g) from Watts Branch, and 2.5 percent (19 g) from Hickey Run. PCB toxicity totaled 3.8×10 −3 µg/kg, with the largest contribution (47 percent) derived from LBDC. Total PAH loads (sum of alkylated and nonalkylated forms) for 2017 were estimated to be 89,000 g, with 23 percent (20,000 g) from NEB, 59 percent (52,000 g) from NWB, 11 percent (9,800 g) from LBDC, 2 percent (1,400 g) from Watts Branch, and 6 percent (5,200 g) from Hickey Run. These results indicate that the largest contributor (75 percent) of PCBs to the Anacostia River is LBDC, although it contributes only 15 percent of the sediment and its basin area represents only 10 percent of the area of the Anacostia River watershed. The majority of the PAH load originates from NWB (59 percent of total) and NEB (22 percent). The ungaged tributaries contribute extremely small loads of PCBs and PAHs, totaling 8.1 g and 765 kg, respectively. More than 94 percent of the total load from the ungaged tributaries is derived from the Nash Run Basin. Various organochlorine pesticides were present in suspended and bed sediment from all gaged and ungaged tributaries; however, elevated detection levels associated with the analytical methods resulted in numerous unquantifiable concentrations in the suspended-sediment samples. Only the pesticide chlordane was found in measurable concentrations in all gaged tributaries. As a result, in this report, a combination of analytical data from suspended-sediment and bed-sediment samples was used to estimate the maximum pesticide loading for each tributary. Chlordane was the principal compound present in the gaged tributaries; the highest average concentration (average of stormflow and low-flow samples from each stream) was 62 µg/kg in sediment from Watts Branch. Chlordane loads for 2017 totaled 1,100 g, of which 7 percent (430 g) was from NEB, 28 percent (320 g) was from NWB, 28 percent (310 g) was from LBDC, 5 percent (56 g) was from Watts Branch, and 1 percent (11 g) was from Hickey Run. Chlordane was not present in suspended or bed sediment from any of the ungaged tributaries. Loads of the other pesticides were estimated by using the highest concentration measured in the combined suspended-sediment and bed-sediment data for each stream. Notable loads include dieldrin (860 g from NWB), methoxychlor (205 g from LBDC), endrin aldehyde (150 g from NWB), and 4,4-DDT (79 g from Watts Branch). Compared with pesticide loads from the gaged streams, those from the ungaged streams were minimal, with only the Pope Branch contribution exceeding 1 gram per year for 4,4-DDE (1.05 g) and 4,4’-DDT (1.3 g). The results of this study show that the dominant source of PCBs and chlordane is LBDC, despite its relatively small basin area. PAHs are ubiquitous throughout the study area, with the largest sources being NEB and NWB; this finding is a result of the large sediment load originating from these basins. The small, ungaged streams supply only minimal PCB and PAH loads, with Nash Run being the largest contributor.

District of Columbia↗

USGS2011

Regional groundwater-flow model of the Redwall-Muav, Coconino, and alluvial basin aquifer systems of northern and central Arizona

A numerical flow model (MODFLOW) of the groundwater flow system in the primary aquifers in northern Arizona was developed to simulate interactions between the aquifers, perennial streams, and springs for predevelopment and transient conditions during 1910 through 2005. Simulated aquifers include the Redwall-Muav, Coconino, and basin-fill aquifers. Perennial stream reaches and springs that derive base flow from the aquifers were simulated, including the Colorado River, Little Colorado River, Salt River, Verde River, and perennial reaches of tributary streams. Simulated major springs include Blue Spring, Del Rio Springs, Havasu Springs, Verde River headwater springs, several springs that discharge adjacent to major Verde River tributaries, and many springs that discharge to the Colorado River. Estimates of aquifer hydraulic properties and groundwater budgets were developed from published reports and groundwater-flow models. Spatial extents of aquifers and confining units were developed from geologic data, geophysical models, a groundwater-flow model for the Prescott Active Management Area, drill logs, geologic logs, and geophysical logs. Spatial and temporal distributions of natural recharge were developed by using a water-balance model that estimates recharge from direct infiltration. Additional natural recharge from ephemeral channel infiltration was simulated in alluvial basins. Recharge at wastewater treatment facilities and incidental recharge at agricultural fields and golf courses were also simulated. Estimates of predevelopment rates of groundwater discharge to streams, springs, and evapotranspiration by phreatophytes were derived from previous reports and on the basis of streamflow records at gages. Annual estimates of groundwater withdrawals for agriculture, municipal, industrial, and domestic uses were developed from several sources, including reported withdrawals for nonexempt wells, estimated crop requirements for agricultural wells, and estimated per capita water use for exempt wells. Accuracy of the simulated groundwater-flow system was evaluated by using observational control from water levels in wells, estimates of base flow from streamflow records, and estimates of spring discharge. Major results from the simulations include the importance of variations in recharge rates throughout the study area and recharge along ephemeral and losing stream reaches in alluvial basins. Insights about the groundwater-flow systems in individual basins include the hydrologic influence of geologic structures in some areas and that stream-aquifer interactions along the lower part of the Little Colorado River are an effective control on water level distributions throughout the Little Colorado River Plateau basin. Better information on several aspects of the groundwater flow system are needed to reduce uncertainty of the simulated system. Many areas lack documentation of the response of the groundwater system to changes in withdrawals and recharge. Data needed to define groundwater flow between vertically adjacent water-bearing units is lacking in many areas. Distributions of recharge along losing stream reaches are poorly defined. Extents of aquifers and alluvial lithologies are poorly defined in parts of the Big Chino and Verde Valley sub-basins. Aquifer storage properties are poorly defined throughout most of the study area. Little data exist to define the hydrologic importance of geologic structures such as faults and fractures. Discharge of regional groundwater flow to the Verde River is difficult to identify in the Verde Valley sub-basin because of unknown contributions from deep percolation of excess surface water irrigation.

Scientific Investigations Report↗

USGS2013

Effects of surface applications of biosolids on groundwater quality and trace-element concentrations in crops near Deer Trail, Colorado, 2004-2010

The U.S. Geological Survey (USGS), in cooperation with Metro Wastewater Reclamation District (Metro District), studied biosolids composition and the effects of biosolids applications on groundwater quality and trace-element concentrations in crops of the Metro District properties near Deer Trail, Colorado, during 2004 through 2010. Priority parameters for each monitoring component included the nine trace elements regulated by Colorado for biosolids (arsenic, cadmium, copper, lead, mercury, molybdenum, nickel, selenium, and zinc); other constituents also were analyzed. All concentrations for the priority parameters in monthly biosolids samples were less than Colorado regulatory limits, and the concentrations were relatively consistent. Biosolids likely were the largest source of nitrogen and phosphorus on the Metro District properties. Plutonium isotopes were not detected in the biosolids, but many organic wastewater compounds (organic wastewater compounds: wastewater indicators, pharmaceuticals, and hormones) were detected in substantial concentrations relative to minimum reporting levels and various surface-water concentrations. Bismuth, copper, mercury, nitrogen, phosphorus, silver, biogenic sterols, detergent degradates, disinfectants, fire retardants, fragrances, pharmaceuticals, and plasticizers would be the most likely biosolids signature to indicate the presence of Metro District biosolids in soil or streambed sediment from the study area. Antimony, cadmium, cobalt, copper, molybdenum, nickel, nitrogen, phosphorus, selenium, tungsten, vanadium, zinc, detergent degradates, disinfectants, fire retardants, fragrances, pharmaceuticals or their degradates, and plasticizers would be the most likely biosolids signature for groundwater and surface water in the study area. More biosolids-signature components detected and larger concentration differences from untreated materials, baseline, and blank samples indicate more evidence of biosolids presence or effects. Although the inorganic constituent concentrations were relatively large in samples from one monitoring well, the concentrations of organic wastewater compounds in groundwater samples were not correspondingly large. Concentrations of organic wastewater compounds in the groundwater samples from all five monitoring wells were less than the minimum reporting levels with only a few detections. Some of the organic wastewater compounds detected could have anthropogenic sources that are not biosolids. Concentrations of priority parameters in groundwater varied spatially and temporally but generally were less than Colorado regulatory limits. Concentrations of dissolved nitrate, arsenic, and selenium, in addition to chloride, sulfate, total dissolved solids, boron, iron, manganese, and uranium, in samples from some wells exceeded the Colorado standards. Concentrations of dissolved nitrate (three wells), molybdenum (one well), selenium (two wells), and uranium (one well) in shallow groundwater had significant (alpha = 0.05) upward trends in some parts of the study area. The biosolids-signature results indicate that the aquifers intercepted by the five routinely sampled wells likely have received some recharge through treated (biosolids-applied) fields or biosolids-affected ponds. Adverse effects from this biosolids-related recharge range from few (if any) at one well to large and significantly (alpha = 0.05) increasing nitrate concentrations at another well. A statistical evaluation of five paired wheat-grain samples from treated (biosolids-applied) fields and untreated (control) fields did not indicate any evidence that biosolids applications significantly (alpha = 0.05 or 0.10) increased concentration of any of these constituents in wheat grain. The wheat-grain concentrations from this study were similar to those from other studies for fields in North America where no biosolids were applied. The data for the limited crop samples indicate that biosolids applications are not increasing the concentrations of arsenic, cadmium, copper, lead, mercury, molybdenum, nickel, selenium, sulfur, and zinc in mature wheat grain from the study area.

Colorado↗

USGS2013

An analysis of potential water availability from the Atwood, Leesville, and Tappan Lakes in the Muskingum River Watershed, Ohio

This report presents the results of a study to assess potential water availability from the Atwood, Leesville, and Tappan Lakes, located within the Muskingum River Watershed, Ohio. The assessment was based on the criterion that water withdrawals should not appreciably affect maintenance of recreation-season pool levels in current use. To facilitate and simplify the assessment, it was assumed that historical lake operations were successful in maintaining seasonal pool levels, and that any discharges from lakes constituted either water that was discharged to prevent exceeding seasonal pool levels or discharges intended to meet minimum in-stream flow targets downstream from the lakes. It further was assumed that the volume of water discharged in excess of the minimum in-stream flow target is available for use without negatively impacting seasonal pool levels or downstream water uses and that all or part of it is subject to withdrawal. Historical daily outflow data for the lakes were used to determine the quantity of water that potentially could be withdrawn and the resulting quantity of water that would flow downstream (referred to as “flow-by”) on a daily basis as a function of all combinations of three hypothetical target minimum flow-by amounts (1, 2, and 3 times current minimum in-stream flow targets) and three pumping capacities (1, 2, and 3 million gallons per day). Using both U.S. Geological Survey streamgage data and lake-outflow data provided by the U.S. Army Corps of Engineers resulted in analytical periods ranging from 51 calendar years for the Atwood Lake to 73 calendar years for the Leesville and Tappan Lakes. The observed outflow time series and the computed time series of daily flow-by amounts and potential withdrawals were analyzed to compute and report order statistics (95th, 75th, 50th, 25th, 10th, and 5th percentiles) and means for the analytical period, in aggregate, and broken down by calendar month. In addition, surplus-water mass curve data were tabulated for each of the lakes. Monthly order statistics of computed withdrawals indicated that, for the three pumping capacities considered, increasing the target minimum flow-by amount tended to reduce the amount of water that can be withdrawn. The reduction was greatest in the lower percentiles of withdrawal; however, increasing the flow-by amount had no impact on potential withdrawals during high flow. In addition, for a given target minimum flow-by amount, increasing the pumping rate increased the total amount of water that could be withdrawn; however, that increase was less than a direct multiple of the increase in pumping rate for most flow statistics. Potential monthly withdrawals were observed to be more variable and more limited in some calendar months than others. Monthly order statistics and means of computed daily mean flow-by amounts indicated that flow-by amounts generally tended to be lowest during June–October and February. Increasing the target minimum flow-by amount for a given pumping rate resulted in some small increases in the magnitudes of the mean and 50th percentile and lower order statistics of computed mean flow-by, but had no effect on the magnitudes of the higher percentile statistics. Increasing the pumping rate for a given target minimum flow-by amount resulted in decreases in magnitudes of higher-percentile flow-by statistics by an amount equal to the flow equivalent of the increase in pumping rate; however, some lower percentile statistics remained unchanged.

Ohio↗

USGS2015

Evaluation of mean-monthly streamflow-regression equations for Colorado, 2014

The U.S. Geological Survey, in cooperation with the Colorado Water Conservation Board, evaluated the predictive uncertainty of mean-monthly streamflow-regression equations representative of natural streamflow conditions in Colorado. This study evaluates the predictive uncertainty of mean-monthly streamflow-regression equations developed in a 2009 U.S. Geological Survey study using streamflow data collected over the entire period of record at each streamgage through calendar year 2013. The study area for this report is limited to the Mountain, Northwest, Rio Grande, and Southwest hydrologic regions of Colorado. Data collected from the beginning of the period of record through calendar year 2013 were used to evaluate the mean-monthly streamflow equations using the same basin characteristics as in the 2009 study. U.S. Geological Survey and Colorado Division of Water Resources streamgages with at least 10 years of streamflow record and identified as representative of natural streamflow conditions were selected for this study. During the streamgage selection process, a total of 432 streamgages, composed of 278 from the 2009 study and 154 new streamgages, were identified. The updated standard error of prediction and adjusted coefficient of determination values that correspond to the mean-monthly streamflow equations developed in the 2009 study are in close agreement with the results of this study. The old streamgages performed slightly better than the new streamgages, with approximately 88 and 85 percent of the data within the prediction intervals, respectively. This result was expected because the streamgages used to develop the regression equations should yield a better performance than the new streamgages. For all hydrologic regions, approximately 87 percent of the data are within the 95-percent prediction intervals. The explanation for why fewer than 95 percent of the data are within the prediction intervals is that the data do not conform perfectly to the regression assumptions required to accurately estimate performance metrics. The equations for the Rio Grande hydrologic region had the best fit with the parametric prediction-interval assumptions, with approximately 91.8 percent of the data within the prediction interval (average 12 months). The Mountain, Northwest, and Southwest hydrologic regions had 87.8, 84.9, and 83.5 percent of the data contained within the prediction interval, respectively. Monthly adjusted coefficient of determination values were computed and have the same general pattern for all four hydrologic regions. The largest values usually occur in March or April, and the lowest values usually occur in August or September. Only the Rio Grande hydrologic region deviates from this seasonal pattern, exhibiting a decrease in adjusted coefficient of determination values in August and September, with the lowest values occurring in the winter months (December, January, and February). Generally, the adjusted coefficient of determination values for this report are just slightly less (0.76 compared to 0.79) than the values computed in the 2009 study. The similarity of values, even when tested with data not used to originally develop the mean-monthly streamflow-regression equations, provides confidence that the predictive uncertainty of mean-monthly regression equations in the 2009 study are accurate. The fact that the results for the two datasets are very similar provides assurance that when these equations are applied to locations not used to develop the equations, the standard error of prediction and adjusted-coefficient of determination error metrics should be similar to those established in the 2009 study for locations with natural streamflow. The median absolute differences between the observed and computed mean-monthly streamflow for Mountain, Northwest, and Southwest hydrologic regions are fairly uniform throughout the year, with the exception of late summer and early fall (July, August, and September), when each hydrologic region exhibits a substantial increase in median absolute percent difference. The greatest difference occurs in the Northwest hydrologic region, and the smallest difference occurs in the Mountain hydrologic region. The Rio Grande hydrologic region shows seasonal variation in median absolute percent difference with March, April, August, and September having a median absolute difference near or below 40 percent, and the remaining months of the year having a median absolute difference near or above 50 percent. In the Mountain, Northwest, and Southwest hydrologic regions, the mean-monthly streamflow equations perform the best during spring (March, April, and May). However, in the Rio Grande hydrologic region, the mean-monthly streamflow equations perform the best during late summer and early fall (August and September).

Arizona, Colorado, New Mexico, Utah, Wyoming↗

USGS2013

Groundwater contributions of flow, nitrate, and dissolved organic carbon to the lower San Joaquin River, California, 2006-08

The influence of groundwater on surface-water quality in the San Joaquin River, California, was examined for a 59-mile reach from the confluence with Salt Slough to Vernalis. The primary objective of this study was to quantify the rate of groundwater discharged to the lower San Joaquin River and the contribution of nitrate and dissolved organic carbon concentrations to the river. Multiple lines of evidence from four independent approaches were used to characterize groundwater contributions of nitrogen and dissolved organic carbon. Monitoring wells (in-stream and bank wells), streambed synoptic surveys (stream water and shallow groundwater), longitudinal profile surveys by boat (continuous water-quality parameters in the stream), and modeling (MODFLOW and VS2DH) provided a combination of temporal, spatial, quantitative, and qualitative evidence of groundwater contributions to the river and the associated quality. Monitoring wells in nested clusters in the streambed (in-stream wells) and on both banks (bank wells) along the river were monitored monthly from September 2006 to January 2009. Nitrate concentrations in the bank wells ranged from less than detection—that is, less than 0.01 milligrams per liter (mg/L) as nitrogen (N)—to approximately 13 mg/L as N. Nitrate was not detected at 17 of 26 monitoring wells during the study period. Dissolved organic carbon concentrations among monitoring wells were highly variable, but they generally ranged from 1 to 4 mg/L. In a previous study, 14 bank wells were sampled once in 1988 following their original installation. With few exceptions, specific conductivity and nitrate concentrations measured in this study were virtually identical to those measured 20 years ago. Streambed synoptic measurements were made by using a temporarily installed drive-point piezometer at 113 distinct transects across the stream during 4 sampling events. Nitrate concentrations exceeded the detection limit of 0.01 mg/L as N in 5 percent of groundwater samples collected from the in-stream wells as part of the synoptic surveys. Only 7 of the 113 cross-sectional transects had nitrate concentrations greater than 1 mg/L as N. In contrast, surface waters in the San Joaquin River tended to have nitrate concentrations in the 1–3 mg/L as N range. A zone of lower oxygen (less than 2 mg/L) in the streambed could limit nitrate contributions from regional groundwater flow because nitrate can be converted to nitrogen gas within this zone. Appreciable concentrations of ammonium (average concentration was 1.92 mg/L as N, and 95th percentile was 10.34 mg/L as N) in the shallow groundwater, believed to originate from anoxic mineralization of streambed sediments, could contribute nitrogen to the overlying stream as nitrate following in-stream nitrification, however. Dissolved organic carbon concentrations were highly variable in the shallow groundwater below the river (1 to 6 ft below streambed) and generally ranged between 1 and 5 mg/L, but had maximum concentrations in the 15–25 mg/L range. The longitudinal profile surveys were not particularly useful in identifying groundwater discharge areas. However, the longitudinal approach described in this report was useful as a baseline survey of measured water-quality parameters and for identifying tributary inflows that affect surface-water concentrations of nitrate. Results of the calibrated MODFLOW model indicated that the simulated groundwater discharge rate was approximately 1.0 cubic foot per second per mile (cfs/mi), and the predominant horizontal groundwater flow direction between the deep bank wells was westward beneath the river. The modeled (VS2DH) flux values (river gain versus river loss) were calculated for the irrigation and non-irrigation season, and these fluxes were an order of magnitude less than those from MODFLOW. During the irrigation season, the average river gain was 0.11 cfs/mi, and the average river loss was −0.05 cfs/mi. During the non-irrigation season, the average river gain was 0.10 cfs/mi, and the average river loss was -0.08 cfs/mi. Information on groundwater interactions and water quality collected for this study was used to estimate loads of nitrate and dissolved organic carbon from the groundwater to the San Joaquin River. Estimated loads of dissolved inorganic nitrogen and dissolved organic carbon were calculated by using concentrations measured during four streambed synoptic surveys and the estimated groundwater discharge rate to the San Joaquin River from MODFLOW of 1 cfs/mi. The estimated groundwater loads to the San Joaquin River for dissolved inorganic nitrogen and dissolved organic carbon were 300 and 350 kilograms per day, respectively. These loads represent 9 and 7 percent, respectively, of the estimated instantaneous surface-water loads for dissolved inorganic nitrogen and dissolved organic carbon at the most downstream site, Vernalis, measured during the four streambed synoptic surveys.

California↗

USGS2015

Groundwater chemistry in the vicinity of the Puna Geothermal Venture Power Plant, Hawai‘i, after two decades of production

We report chemical data for selected shallow wells and coastal springs that were sampled in 2014 to determine whether geothermal power production in the Puna area over the past two decades has affected the characteristics of regional groundwater. The samples were analyzed for major and minor chemical species, trace metals of environmental concern, stable isotopes of water, and two organic compounds (pentane and isopropanol) that are injected into the deep geothermal reservoir at the power plant. Isopropanol was not detected in any of the groundwaters; confirmed detection of pentane was restricted to one monitoring well near the power plant at a low concentration not indicative of source. Thus, neither organic compound linked geothermal operations to groundwater contamination, though chemical stability and transport velocity questions exist for both tracers. Based on our chemical analysis of geothermal fluid at the power plant and on many similar results from commercially analyzed samples, we could not show that geothermal constituents in the groundwaters we sampled came from the commercially developed reservoir. Our data are consistent with a long-held view that heat moves by conduction from the geothermal reservoir into shallow groundwaters through a zone of low permeability rock that blocks passage of geothermal water. The data do not rule out all impacts of geothermal production on groundwater. Removal of heat during production, for example, may be responsible for minor changes that have occurred in some groundwater over time, such as the decline in temperature of one monitoring well near the power plant. Such indirect impacts are much harder to assess, but point out the need for an ongoing groundwater monitoring program that should include the coastal springs down-gradient from the power plant.

Hawaii↗

USGS2022

Potential effects of energy development on environmental resources of the Williston Basin in Montana, North Dakota, and South Dakota—Water resources

The Williston Basin has been a leading oil and gas producing area for more than 50 years. While oil production initially peaked within the Williston Basin in the mid-1980s, production rapidly increased in the mid-2000s, largely because of improved horizontal (directional) drilling and hydraulic fracturing methods. In 2012, energy development associated with the Bakken Formation was identified as a priority requiring collaboration toward improved timeliness of issuing permits for new wells combined with reasonable measures to maintain environmental quality. Shortly thereafter, the Bakken Federal Executive Group was created to address common challenges associated with energy development. The Bakken Federal Executive Group partner agencies identified a gap in current understanding of the cumulative environmental challenges attributed to energy development throughout the area, resulting in an effort to aggregate scientific data and identify additional research and information needs related to natural resources within areas of energy development in the Williston Basin. As part of this effort, water resources in the area (including groundwater; streams and rivers; and lakes, reservoirs, and wetlands) were characterized and described in terms of physical occurrence, flow characteristics, recharge, water quality, and water use. Similarly, waters produced during energy-development activities also were characterized even though these waters are not considered usable resources within the area. Groundwater resources were characterized by the major hydrogeologic units, or aquifers, identifying the units that supply most groundwater used for domestic, stock, agricultural, and industrial purposes. The groundwater characterization included other deeper hydrogeologic units in the Williston Basin that may be a useable source of water with treatment, have utility as a reservoir for reinjection of produced waters, or be a source of minerals and energy resources. A generalized groundwater budget and flow system identifying the sources of recharge (stream infiltration, precipitation, and movement [leakage] from other aquifers) and the general groundwater flow direction is included for each of the major hydrogeologic units. Rivers and streams within the Williston Basin with 10 or more years of continuous streamflow data were identified. For a subset of these sites, streamflow characteristics, including the monthly and annual mean flow, were generated to identify seasonal and interannual changes in streamflow and thus provide information on the drivers and reliability of streamflow at the seasonal or multiyear scale. Daily streamflow and annual extreme flows (peak and low flow) also were estimated for the subset of sites. The daily streamflow and annual extreme flow values provide information on short-term or extreme events that are relevant to infrastructure design and evaluating spills, leaks, or accidental discharges of water or petroleum products. Surface-water features (lakes, ponds, and wetlands) were classified using the Cowardin system and identified on the National Wetlands Inventory maps generated by the U.S. Fish and Wildlife Service. The spatial distribution of the surface-water features was analyzed by State, county, and specifically in comparison to the Prairie Pothole Region. The proximity of the surface-water features to energy development infrastructure (specifically oil or gas well pads) was evaluated. It was determined that, although oil or gas wells are often near a surface-water feature, most surface-water features do not have wells nearby, with the exception of wells in the Prairie Pothole Region. Water-quality data were aggregated from two data sources: (1) the Water-Quality Portal, sponsored by the U.S. Geological Survey (USGS), U.S. Environmental Protection Agency (EPA), and National Water Quality Monitoring Council; and (2) a data compilation completed as part of the USGS National Water-Quality Assessment project. The Water-Quality Portal integrates publicly available water-quality data from databases maintained by the USGS, EPA, and U.S. Department of Agriculture, including water-quality data from Tribal, State, and local databases. Water-quality data for 15 commonly measured water-quality constituents were aggregated for groundwater, rivers and streams, and lakes and reservoirs. For each aggregated dataset (groundwater, rivers and streams, and lakes and reservoirs), analyses of the water-quality data included summary statistics, maps of spatial distribution of constituent values, boxplots of constituent values by timeframe or hydrogeologic unit, spatial comparisons of site locations and constituent values to petroleum well density, and comparisons of the constituent values measured to EPA drinking-water standards/guidelines. Produced water includes all fluids brought to the surface along with the targeted hydrocarbons as part of the oil and gas exploration and extraction processes. These fluids may include formation water (waters that co-exist with rock/oil/gas), hydraulic fracturing fluids, and other combinations of water and chemicals used during oil and gas well drilling, development, treatments, recompletions, and workovers. Produced water datasets were aggregated from two sources: the USGS National Produced Waters Geochemical database (ver. 2.1) and a series of projects focused specifically on sampling produced water in the Williston Basin from 2010 to 2014. The National Produced Waters Geochemical database was useful for a general understanding of produced-water chemistry. Produced waters are characterized by extreme salinity and contain elevated concentrations of other constituents (including arsenic, barium, cadmium, lead, zinc, radium-226/radium-228, and ammonium) that could negatively affect water and aquatic resources if released. Produced waters also have a generally unique chemical (isotopic) signature that may be useful in tracking water from different geologic units; for example, the oxygen/deuterium and strontium ratio values measured in brine waters from the Bakken Formation are distinct from brines collected from other geologic units in the Williston Basin. Water-use information related to energy production in the area also was aggregated and summarized. The summary of water use is not limited to oil and gas production but includes water used to produce all types of energy resources in the Williston Basin, including coal/lignite, thermoelectric power, oil and gas, hydropower, biomass and biofuels, wind, geothermal, and solar. Each State has its own methods for regulating and reporting water usage within its jurisdiction. These methods can introduce problems when examining water use from sources, such as the Missouri River or Fox Hills aquifer, that are shared across political boundaries. Without the one-to-one match for usage types and amounts used from a water source, it is difficult to develop a comprehensive water budget for the water source being evaluated. A large amount of freshwater is required to prepare a well for oil and gas well production; in some cases, 3 to 7 million gallons of water are needed per well. The EPA estimates that hydraulic fracturing in the Williston Basin uses between 70 to 140 billion gallons per year. Water also is used for myriad other purposes related to ancillary oil and gas extraction. In addition to water used for immediate energy development, the expanded human workforce migrating into the area and other support staff who have moved into the area during the development also use water. Research and information needs were identified that could be relevant in the evaluation of the effects of energy development on water resources. Information needs related to the evaluation of groundwater resources include the following: improved potentiometric-surface maps for glacial units; availability of a uniform stream network digital geographic coverage that spans the international boundary with Canada; enhanced surface-water use information with regards to the gain and loss of streamflow to shallow groundwater, which would increase understanding groundwater and surface-water interactions; and expanded geophysical assessments. Gaps in the availability of streamflow data include the lack of information on ice-jam flooding despite potential for effects to infrastructure (pipelines, roads, and facilities) and an understanding of the cumulative effects of largely undocumented stock and diversion dams. Although this study resulted in the aggregation of a large quantity of water-quality data, the availability of consistently collected, systematically processed and reported data over large parts of the Williston Basin is sparse. Few samples have been analyzed for constituents that may indicate the effect of energy development on water resources. Constituents that could be considered include boron, chloride, bromide, iodine, fluoride, manganese, lithium, radium, strontium isotopes, volatile organic compounds, and isotopes of inorganic ions (such as hydrogen and carbon). Collaboration between Tribal, Federal, State, and local entities to identify a common study design, common monitoring constituents, and consistent sampling locations would generate datasets with broad utility and would likely result in overall cost savings for monitoring over time. Similarly, there is a need for standardized sample collection, processing, laboratory analytical methods, and the collection of ancillary data for produced waters sampling. Additional characterization of the range of chemical, microbial, and isotopic compositions and quantities of “end-member” produced waters, and the collection of time-series datasets to document the changes in produced waters during and after well development also were needs identified during this study. Water-use estimates would be improved through the implementation of comprehensive studies of water use from groundwater and surface-water sources using consistent methodologies across the Williston Basin. The submission of chemical and water data related to hydraulic fracturing collected by the oil and gas industry would add to the quantity of available data. Consistent implementation of regulations and monitoring controls across political boundaries (State, county, and international) would further improve the consistency of data available for the estimates of water use.

Montana, North Dakota, South Dakota↗

USGS2023

Development of the North Carolina stormwater-treatment decision-support system by using the Stochastic Empirical Loading and Dilution Model (SELDM)

The Federal Highway Administration and State departments of transportation nationwide need an efficient method to assess potential adverse effects of highway stormwater runoff on receiving waters to optimize stormwater-treatment decisions. To this end, the U.S. Geological Survey, in cooperation with the Federal Highway Administration and the North Carolina Department of Transportation (NCDOT), developed a decision-support software tool based on a statewide version of the Stochastic Empirical Loading and Dilution Model (SELDM). This decision-support tool is designed to identify potential adverse effects of highway runoff by using a criterion based on a measurable change in water quality from a surrogate pollutant. The NCDOT worked with the North Carolina Department of Environmental Quality to select a 25-percent change in suspended sediment concentration as the decision-rule criterion for identifying measurable downstream water-quality change; this selection was based on available data and widely accepted stormwater monitoring uncertainties. Development of the statewide tool and its application to the Piedmont ecoregion are described in this report. Because SELDM can be applied to build a similar decision-support tool in any State, this report describes practice-ready methods that other State departments of transportation and municipal permittees can use to streamline environmental permitting and project delivery while protecting the environment. Hydraulic design engineers can use this decision-support tool to establish stormwater-treatment goals for highway construction or improvement projects without having to learn SELDM or interpret its statistical output. The tool is a spreadsheet that determines if a selected highway segment can directly discharge highway runoff, if the highway segment can discharge runoff following treatment using a basic vegetated conveyance best management practice (BMP), or if treatment using an advanced BMP is needed to minimize effects of discharges on downstream water quality. To use the tool, hydraulic design engineers obtain upstream-basin characteristics from the U.S. Geological Survey StreamStats application and highway-site characteristics from preliminary design plans. They then enter these characteristics in the decision-support tool, which identifies the necessary stormwater-treatment goal. The Piedmont ecoregion was used as a case study to demonstrate the type of information the decision-support tool can provide. In this ecoregion, 100 percent of direct discharges meet the water-quality criterion when the drainage-area ratio is less than about 0.007 acres of highway per square mile of upstream basin. Advanced BMPs are needed in 100 percent of basins with drainage-area ratios greater than about 50 acres per square mile. Between these drainage-area ratios, the selection of direct discharge, a basic vegetated conveyance BMP, or an advanced BMP is a function of highway-site and upstream-basin properties.

North Carolina↗

USGS2024

Dissolved arsenic concentrations in surface waters within the upper portions of the Klamath River Basin, Oregon and California

Arsenic toxicity is an environmental health problem. Levels of arsenic in surface waters at some locations in the Klamath River Basin in southern Oregon and northern California can exceed the U.S. Environmental Protection Agency (EPA) standard for drinking water. There are both anthropogenic and natural sources of arsenic. The Klamath River Basin consists primarily of volcanic deposits and contains an underground geothermal system with hot springs and warm water wells, all known natural sources of arsenic. Anthropogenic sources of arsenic are related to the agricultural use of herbicides, fungicides, and insecticides. Surface water arsenic levels can also be affected by fertilizer amendments, evaporative concentration, oxygen-level depletion, and various geochemical transformations that can increase arsenic mobilization. In this study by the U.S. Geological Survey and the Bureau of Reclamation, dissolved concentrations of arsenic, copper, and lead were measured in surface waters at 39 unique sites within the upper portions of the Klamath River Basin between 2018 and 2022. In every year, except 2022, sites were sampled four times between April and November. Surface-water arsenic concentrations varied up to four-orders of magnitude among sites. Median arsenic concentration was lowest at Cherry Creek (0.03 micrograms per liter [μg/L]) and highest at Wood Kimball Spring (36.7 μg/L), two sites located north of Upper Klamath Lake. The highest arsenic concentrations (17.4±4.9 μg/L, n =3) were found in drain sites (defined here as a waterbody returning used irrigation water) while the lowest arsenic concentrations were found in an artesian well (0.8 μg/L, n =1). The elevated arsenic concentrations of the drain sites suggest that arsenic might be concentrated or mobilized by agricultural activities, water re-use practices, and (or) by geochemical processes occurring around water stored in drains (that is, in the water column and across sediment water boundaries). A source of arsenic in drain water in the Klamath Strait Drain area includes water used for irrigation originating from Ady Canal. Other potential sources include groundwater, geothermal water, and local soils and sediments. Seasonal differences in surface-water arsenic concentrations were detected at 13 sites, 10 of which had higher arsenic concentrations in summer than in either spring or fall. The sites sampled around Upper Klamath Lake, the impounded rivers, one of the two canal sites, and 5 of the 14 river sites had higher surface-water arsenic concentrations in the summer than in either spring or fall. Surface-water arsenic concentrations from groundwater sources (that is, springs and in the artesian well) did not vary significantly among seasons (p-values greater than 0.1). Median surface-water concentrations of copper and lead ranged from 0.03 to 3.7 μg/L, and from 0.013 to 0.175 μg/L ( n =2–18), respectively. Dissolved concentrations of both metals were below acute toxicity endpoints reported by the EPA for freshwater animals. Surface-water arsenic concentrations varied independently from corresponding changes in surface-water lead or copper concentrations. However, arsenic concentrations measured in bed-sediment samples collected from a subset of sites located north of Upper Klamath Lake correlated strongly and significantly with the corresponding sedimentary lead concentrations ( p =0.015). Aqueous arsenic speciation measured in a subset of sites in 2019 and 2022 showed that all the arsenic existed as arsenic (V), the most oxidized arsenic species, and presumably, the least toxic. The highest proportions of arsenite (As(III)), the presumably most toxic arsenic species, relative to total arsenic concentrations were found at drain sites. Our assessment of dissolved arsenic concentrations in various surface-water bodies in the Upper Klamath River Basin reveals geographical areas of consistently low (below 2.1 μg/L), moderate (below 10 μg/L) and high (above 10 μg/L) surface-water arsenic concentrations. South of Upper Klamath Lake, surface-water arsenic concentrations were consistently higher than 20 μg/L at two drain sites located in an area of predominant agricultural land use with extensive water re-use practices. North of Upper Klamath Lake, surface-water arsenic concentrations greater than 20 μg/L were consistently measured at sites with limited nearby agricultural activities, suggesting a geogenic source. The consistently high arsenic levels from the Wood River at Jackson F. Kimball State Park, Fort Creek, and Crooked Creek, which are sites located at or near headwater spring sources, suggest a natural background source of arsenic. Water flowing downstream from this area could be a potential source of arsenic to Upper Klamath Lake and the Upper Klamath River.

California, Oregon↗

USGS2006

Watershed characteristics and pre-restoration surface-water hydrology of Minebank Run, Baltimore County, Maryland, water years 2002-04

Stream restoration efforts have been ongoing in Maryland since the early 1990s. Physical stream restoration often involves replacement of lost sediments to elevate degraded streambeds, re-establishment of riffle-pool sequences along the channel profile, planting vegetation in riparian zones, and re-constructing channel banks, point bars, flood plains, and stream-meanders. The primary goal of many restoration efforts is to re-establish geomorphic stability of the stream channel and reduce erosive energy from urban runoff. Monitoring streams prior to and after restoration could help quantify other possible benefits of stream restoration, such as improved water quality and biota. This report presents general watershed characteristics associated with the Minebank Run watershed; a small, urban watershed in the south-central section of Baltimore County, Maryland that was physically restored in phases during 1999, 2004, and 2005. The physiography, geology, hydrology, land use, soils, and pre-restoration geomorphic setting of the unrestored stream channel are discussed. The report describes a reach of Minebank Run that was selected for the purpose of collecting several types of environmental data prior to restoration, including continuous-record and partial-record stage and streamflow data, precipitation, and ground-water levels. Examples of surface-water data that were collected in and near the study reach during water years 2002 through 2004, including continuous-record streamflow, partial-record stage and discharge, and precipitation, are described. These data were used in analyses of several characteristics of surface-water hydrology in the watershed, including (1) rainfall totals, storm duration, and intensity, (2) instantaneous peak discharge and daily mean discharge, (3) stage-discharge ratings, (4) hydraulic-geometry relations, (5) water-surface slope, (6) time of concentration, (7) flood frequency, (8) flood volume, and (9) rainfall-runoff relations. Several hydrologic characteristics that are typical of urban environments were quantified by these analyses. These include (1) large ratios of peak discharge to daily mean discharge as an indicator of flashiness, (2) consistent shifting of the stage-discharge rating over short periods of time that indicates instability of the stream channel, (3) analyses of hydraulic-geometry relations that indicate mean velocities of 11 feet per second or more while the flow is contained in the stream channel, (4) discharges that are 4 to 5 times larger in Minebank Run for corresponding flood frequency recurrence intervals than in Slade Run, which is a Piedmont watershed of similar size with smaller percentages of urban development, and (5) flood waves that can travel through the stream channel at a velocity of 412 feet per minute, or 6.9 feet per second.

Maryland↗

USGS2017

A process to estimate net infiltration using a site-scale water-budget approach, Rainier Mesa, Nevada National Security Site, Nevada, 2002–05

This report documents a process used to estimate net infiltration from precipitation, evapotranspiration (ET), and soil data acquired at two sites on Rainier Mesa. Rainier Mesa is a groundwater recharge area within the Nevada National Security Site where recharged water flows through bedrock fractures to a deep (450 meters) water table. The U.S. Geological Survey operated two ET stations on Rainier Mesa from 2002 to 2005 at sites characterized by pinyon-juniper and scrub-brush vegetative cover. Precipitation and ET data were corrected to remove measurement biases and gap-filled to develop continuous datasets. Net infiltration (percolation below the root zone) and changes in root-zone water storage were estimated using a monthly water-balance model. Site-scale water-budget results indicate that the heavily-fractured welded-tuff bedrock underlying thin (<40 centimeters) topsoil is a critical water source for vegetation during dry periods. Annual precipitation during the study period ranged from fourth lowest (182 millimeters [mm]) to second highest (708 mm) on record (record = 55 years). Annual ET exceeded precipitation during dry years, indicating that the fractured-bedrock reservoir capacity is sufficient to meet atmospheric-evaporative demands and to sustain vegetation through extended dry periods. Net infiltration (82 mm) was simulated during the wet year after the reservoir was rapidly filled to capacity. These results support previous conclusions that preferential fracture flow was induced, resulting in an episodic recharge pulse that was detected in nearby monitoring wells. The occurrence of net infiltration only during the wet year is consistent with detections of water-level rises in nearby monitoring wells that occur only following wet years.

Nevada↗

USGS2017

Hydrogeologic and geochemical characterization and evaluation of two arroyos for managed aquifer recharge by surface infiltration in the Pojoaque River Basin, Santa Fe County, New Mexico, 2014–15

In order to provide long-term storage of diverted surface water from the Rio Grande as part of the Aamodt water rights settlement, managed aquifer recharge by surface infiltration in Pojoaque River Basin arroyos was proposed as an option. The initial hydrogeologic and geochemical characterization of two arroyos located within the Pojoaque River Basin was performed in 2014 and 2015 in cooperation with the Bureau of Reclamation to evaluate the potential suitability of these two arroyos as sites for managed aquifer recharge through surface infiltration. The selected reaches were high-gradient (average 3.0–3.5 percent) braided channels filled with unconsolidated sand and gravel-sized deposits that were generally 30–50 feet thick. Saturation was not observed in the unconsolidated channel sands in four subsurface borings but was found at 7–60 feet below the contact between the unconsolidated channel sands and the bedrock. The poorly to well-cemented alluvial deposits that make up the bedrock underlying the unconsolidated channel material is the Tesuque Formation. The individual beds of the Tesuque Formation are reported to be highly heterogeneous and anisotropic, and the bedrock at the site was observed to have variable moisture and large changes in lithology. Surface electrical-resistivity geophysical survey methods showed a sharp contrast between the electrically resistive unconsolidated channel sands and the highly conductive bedrock; however, because of the high conductivity, the resistivity methods were not able to image the water table or preferential flow paths (if they existed) in the bedrock. Infiltration rates measured by double-ring and bulk infiltration tests on a variety of channel morphologies in the study reaches were extremely large (9.7–94.5 feet per day), indicating that the channels could potentially accommodate as much as 6.6 cubic feet per second of applied water without generating surface runoff out of the reach; however, the small volume available for storage in the unconsolidated channel sands (about 410 acre-feet in the east arroyo and about 190 acre-feet in the west arroyo) and the potential for the infiltrating water to preferentially flow over the bedrock contact and out of the reach present a challenge for storing water. Although a detailed assessment of the infiltration rate of the Tesuque Formation is beyond the scope of this investigation, one double-ring infiltrometer test was conducted on an outcrop, resulting in an estimated infiltration rate of about 4 feet per day. The shallow groundwater observed in this investigation was determined to be recharged locally on the basis of groundwater elevations and geochemical and isotopic signatures. The channel sands and shallow bedrock were observed to be weathered, indicating contact with oxic groundwater following deposition. This observation was supported by whole-rock elemental analysis and mineralogy of several core samples. The downward groundwater gradient between the shallow wells and those wells screened at greater depths suggests that the shallow groundwater is recharged by local precipitation and has the potential to migrate to the deeper aquifer units. The two age-dating tracers measured in this investigation, however, demonstrate that the shallow groundwater flow paths are very slow and that the deeper flow paths are likely part of a larger regional system. The composition of the shallow, native groundwater suggests that storing water diverted from the Rio Grande is not likely to leach constituents of concern that would cause the stored water to exceed health-based U.S. Environmental Protection Agency Maximum Contaminant Levels.

New Mexico↗

USGS2007

Continuous water-quality monitoring and regression analysis to estimate constituent concentrations and loads in the Sheyenne River, North Dakota, 1980-2006

This report presents the results of a study by the U.S. Geological Survey, done in cooperation with the North Dakota State Water Commission, to estimate water-quality constituent concentrations at seven sites on the Sheyenne River, N. Dak. Regression analysis of water-quality data collected in 1980-2006 was used to estimate concentrations for hardness, dissolved solids, calcium, magnesium, sodium, and sulfate. The explanatory variables examined for the regression relations were continuously monitored streamflow, specific conductance, and water temperature. For the conditions observed in 1980-2006, streamflow was a significant explanatory variable for some constituents. Specific conductance was a significant explanatory variable for all of the constituents, and water temperature was not a statistically significant explanatory variable for any of the constituents in this study. The regression relations were evaluated using common measures of variability, including R2, the proportion of variability in the estimated constituent concentration explained by the explanatory variables and regression equation. R2 values ranged from 0.784 for calcium to 0.997 for dissolved solids. The regression relations also were evaluated by calculating the median relative percentage difference (RPD) between measured constituent concentration and the constituent concentration estimated by the regression equations. Median RPDs ranged from 1.7 for dissolved solids to 11.5 for sulfate. The regression relations also may be used to estimate daily constituent loads. The relations should be monitored for change over time, especially at sites 2 and 3 which have a short period of record. In addition, caution should be used when the Sheyenne River is affected by ice or when upstream sites are affected by isolated storm runoff. Almost all of the outliers and highly influential samples removed from the analysis were made during periods when the Sheyenne River might be affected by ice.

North Dakota↗

USGS2008

Occurrence of chlorothalonil, its transformation products, and selected other pesticides in Texas and Oklahoma streams, 2003-2004

The primary purpose of the study described in this report was to determine if the fungicide chlorothalonil (2,4,5,6-tetrachloroisophthalonitrile), three of its transformation products, or selected other pesticides are transported to surface water after use on peanuts or other crops in Texas and Oklahoma. The results summarized here are part of a larger study that includes data from sites in Alabama, Florida, and Georgia. Chlorothalonil is classified as a probable carcinogen, and the 4-hydroxy of chlorothalonil transformation product is more soluble, more stable, and, for some species, more toxic than its parent compound. In 2003, water samples were collected from three surface-water sites in Texas and two surface-water sites in Oklahoma; in 2004, samples were collected from the two Oklahoma sites. Chlorothalonil was not detected in any of the 20 samples analyzed. The 4-hydroxy of chlorothalonil transformation product was detected in three samples collected in 2004, with a maximum concentration of 0.018 microgram per liter (?g/L); the other two transformation products (diamide chlorothalonil and 1-amide-4-hydroxy chlorothalonil) were not detected in any sample. In addition, 19 samples were analyzed for as many as 109 other pesticides and transformation products. Atrazine was detected in 13 samples and had a maximum concentration of 0.122 ?g/L. Deethylatrazine was detected in 10 samples and had a maximum concentration of 0.04 ?g/L. Metolachlor was detected in eight samples and had a maximum concentration of 0.019 ?g/L. Fifteen other pesticides or pesticide transformation products also were detected. In general, concentrations of pesticides were less than concentrations that are commonly observed in Midwestern streams. The results indicate that the use of chlorothalonil on peanut crops has not resulted in substantial contamination of the studied streams in Texas and Oklahoma.

Texas, Oklahoma↗

USGS2015

A comparison of methods to predict historical daily streamflow time series in the southeastern United States

Effective and responsible management of water resources relies on a thorough understanding of the quantity and quality of available water. Streamgages cannot be installed at every location where streamflow information is needed. As part of its National Water Census, the U.S. Geological Survey is planning to provide streamflow predictions for ungaged locations. In order to predict streamflow at a useful spatial and temporal resolution throughout the Nation, efficient methods need to be selected. This report examines several methods used for streamflow prediction in ungaged basins to determine the best methods for regional and national implementation. A pilot area in the southeastern United States was selected to apply 19 different streamflow prediction methods and evaluate each method by a wide set of performance metrics. Through these comparisons, two methods emerged as the most generally accurate streamflow prediction methods: the nearest-neighbor implementations of nonlinear spatial interpolation using flow duration curves (NN-QPPQ) and standardizing logarithms of streamflow by monthly means and standard deviations (NN-SMS12L). It was nearly impossible to distinguish between these two methods in terms of performance. Furthermore, neither of these methods requires significantly more parameterization in order to be applied: NN-SMS12L requires 24 regional regressions—12 for monthly means and 12 for monthly standard deviations. NN-QPPQ, in the application described in this study, required 27 regressions of particular quantiles along the flow duration curve. Despite this finding, the results suggest that an optimal streamflow prediction method depends on the intended application. Some methods are stronger overall, while some methods may be better at predicting particular statistics. The methods of analysis presented here reflect a possible framework for continued analysis and comprehensive multiple comparisons of methods of prediction in ungaged basins (PUB). Additional metrics of comparison can easily be incorporated into this type of analysis. By considering such a multifaceted approach, the top-performing models can easily be identified and considered for further research. The top-performing models can then provide a basis for future applications and explorations by scientists, engineers, managers, and practitioners to suit their own needs.

Scientific Investigations Report↗

USGS2023

Preliminary machine learning models of manganese and 1,4-dioxane in groundwater on Long Island, New York

Manganese and 1,4-dioxane in groundwater underlying Long Island, New York, were modeled with machine learning methods to demonstrate the use of these methods for mapping contaminants in groundwater in the Long Island aquifer system. XGBoost, a gradient boosted, ensemble tree method, was applied to data from 910 wells for manganese and 553 wells for 1,4-dioxane. Explanatory variables included soil properties, groundwater flow, land use, and other features that describe the hydrogeology and geochemistry of the aquifer system. Four models were developed to predict the probability of manganese concentrations greater than a detection level of 10 micrograms per liter (μg/L) and greater than three threshold concentrations (50, 150, and 300 μg/L) relevant to drinking-water quality. One model was developed to predict the probability of 1,4-dioxane concentrations greater than a detection level of 0.07 μg/L. The 1,4-dioxane model was limited geographically to Suffolk County because of data availability. Predictions were made for two layers in the upper glacial aquifer and three layers in the Magothy aquifer, which are the upper two of the three major aquifers of the Long Island aquifer system. The objective of the study described in this report was to demonstrate the application of the methods rather than to develop precise estimates of manganese or 1,4-dioxane concentrations at any given location. The predictive models developed in the study are considered preliminary in the sense that they are an initial effort at developing these kinds of models specifically for Long Island. The models could be improved by the inclusion of additional data, by the use of methods to improve the modeling of infrequent high concentrations of manganese and 1,4-dioxane (above threshold concentrations), and by including more explanatory variables that specifically describe conditions and contaminant sources on Long Island. Nonetheless, the distribution of model predictions and the influence of explanatory variables in the models were consistent with the expected relations between contaminant concentrations and groundwater-flow-system characteristics and the distribution of manmade sources. Mapped predictions indicated that manganese detections were more probable in the upper glacial aquifer and along the southern shore of Long Island, consistent with the distribution of anoxic conditions in groundwater in the Long Island aquifer system. Manganese was infrequently predicted at concentrations greater than thresholds of concern for drinking-water quality in any of the aquifer layers. Detections of 1,4-dioxane were predicted in the western, more highly developed parts of Suffolk County, in the upper glacial aquifer and the top and middle layers of the Magothy aquifer, and in northwestern Suffolk County in the bottom layer of the Magothy aquifer. Although preliminary in nature and based on limited data, these mapped predictions can be used to generally identify areas where manganese and 1,4-dioxane may be present at concentrations of concern to prioritize areas for future monitoring and to guide future modeling and mapping efforts.

New York↗