This data release is focused on the analysis of surface water concentration data associated with 12 elements of concern from three hydrologic basins. Data is analyzed with respect to: a) reporting limits, b) the extent of censored data, c) co-location with USGS real-time sensor data, and d) median concentrations at the catchment spatial scale. The Proxies Project (under the Water Quality Program of the USGS Water Mission Area) is a multi-year effort designed to develop rapid and cost-effective approaches for monitoring and risk assessment of a range of aquatic contaminants in riverine surface waters at multiple spatial scales. One component of this project is focused on 12 Elements of Concern (EoC; Al, As, Cd, Cr, Cu, Fe, Hg, Mn, Pb, Se, U and Zn) in three primary hydrologic basins: Delaware River Basin (DRB), the Illinois River Basin (ILRB) and the Upper Colorado (UCOL) River Basin (USGS, 2023). Two modeling approaches being explored as part of the Proxies Project rely on the analysis of previously published EoC concentration data retrieved from the multi-agency supported Water Quality Portal (www.waterqualitydata.us/). This basin-specific retrieved data, covering the 1900-2022 timeframe, was subsequently screened, harmonized and published as part of an earlier USGS Data Release (Marvin-DiPasquale and others, 2022). The two distinct modeling approaches that leverage this previously published data are: a) machine learning statistical analysis of EoC concentration distributions as a function of geospatial attributes; and b) time series analysis in support of estimating EoC concentrations in (near)real-time at a sub-set of USGS real-time stations using discharge in combination with a range of deployed in-situ sensors. Prior to the final stages of model development, there were several data analysis steps required to further define which elements and aquatic fractions (i.e. filtered, unfiltered, and particulate) best lend themselves to further model exploration and development. These intermediate data analyses include: a) an analysis of the change in detection quantitation limits, by element and methods over time (DR_Table _1); b) an analysis of data censoring, by study basin, element, and fraction (DR_Table_2); c) a calculation of median EoC concentrations at the National Hydrography Dataset Plus (NHDPlus) catchment spatial scale (DR_Table_3); d) an analysis of the percentage of censored median EoC concentration values by study basin, element, and fraction (DR_Table_4); e) decision tree analysis associated with the geospatial machine learning modeling approach, by study basin, element and fraction (DR_Table_5); f) discrete EoC concentration data merged with continuous discharge and in-situ sensor data at USGS real-time stations, by station ID, element and fraction (DR_Table_6); and g) an analysis of the total number of observations and the percentage of censored EoC data associated with the merged discrete EoC and continuous discharge and sensor data retrieved from USGS real-time stations, by station ID, element, and fraction (DR_Table_7). The current data release documents the results of these data analyses. The associated seven data tables presented herein are provided in machine-readable comma separated value (*.csv) format and are more fully described in the associated meta-data. REFERENCES Marvin-DiPasquale, M.C., Sullivan, S.L., Platt, L. R., Gorsky, A., Agee, J.L., McCleskey, B.R., Kakouros, E., Walton-Day, K., Runkel, R. L., Morriss, M. C., Wakefield, B. F., and Bergamaschi, B., 2022, Concentration Data for 12 Elements of Concern Used in the Development of Surrogate Models for Estimating Elemental Concentrations in Surface Water of Three Hydrologic Basins (Delaware River, Illinois River and Upper Colorado River): U.S. Geological Survey data release, https://doi.org/10.5066/P9L06M3G. USGS, 2023, Proxies Project, U.S. Geological Survey webpage, accessed 3/11/2025,

These data are comprised of measurements of elements (e.g., uranium, cobalt, nickel, copper, zinc, cadmium, lead, etc.), major anions (chloride, nitrite+nitrate as nitrogen, sulfate, etc.), dissolved organic carbon, and general water quality characteristics in Canyon Mine containment pond water samples collected in calendar year 2019.

These data are comprised of measurements of elements (e.g., uranium, cobalt, nickel, copper, zinc, cadmium, lead, etc.) in Canyon Mine containment pond vegetation samples collected in calendar year 2019.

These data are comprised of measurements of elements (e.g., uranium, cobalt, nickel, copper, zinc, cadmium, lead, etc.) in Canyon Mine containment pond vegetation samples collected in calendar year 2019.

Legacy mine waste from the Clark Fork River in Western Montana has contributed 100 million tons of tailings into the watershed between 1880 and 1982 (E.D. Andrews, Longitudinal dispersion of metals in the Clark Fork River, Montana, Lewis Publishers, 1987). Tailings deposited along the floodplain, streambanks and river channel continue to contribute metal contaminated material into the river in the form of metal-enriched particulate matter or seston, comprising a mixture of organic and inorganic materials (J.N. Moore and S.N. Luoma, Hazardous wastes from large-scale metal extraction: A case study. Environmental Science and Technology, v.24:1278-1285, 1990). Metal enriched seston poses a dietary exposure risk to filter-feeding macroinvertebrates that entrap and ingest suspended materials as a primary food source. Suspended particulate material and dissolved and total recoverable water samples were collected along a metal contamination gradient in 2017 and 2018 in the Clark Fork River and analyzed for metal concentrations to include the highly toxic metals arsenic, cadmium, and copper. Concentrations of seston arsenic (As), cadmium (Cd), and copper (Cu) ranged from 20.7–242 ug As/g, 2.7–16.2 ug Cd/g, and 129–1260 ug Cu/g. Dissolved (filtered) stream water concentrations ranged from 3.5–21.7 ug As/L, 0.1–0.38 ug Cd/L, and 1.5–12.5 ug Cu/L. Total recoverable (unfiltered) water concentrations ranged from 2.7–22.8 ug As/L, 0.1–0.35 ug Cd/L, and 2.0–14.9 ug Cu/L. Data presented here represent metal concentrations in water and seston from a mining-impacted river and provide insight to potential exposure of toxic metals to resident filter-feeding aquatic invertebrates.

Legacy mine waste from the Clark Fork River in Western Montana has contributed 100 million tons of tailings into the watershed between 1880 and 1982 (E.D. Andrews, Longitudinal dispersion of metals in the Clark Fork River, Montana, Lewis Publishers, 1987). Tailings deposited along the floodplain, streambanks and river channel continue to contribute metal contaminated material into the river in the form of metal-enriched particulate matter or seston, comprising a mixture of organic and inorganic materials (J.N. Moore and S.N. Luoma, Hazardous wastes from large-scale metal extraction: A case study. Environmental Science and Technology, v.24:1278-1285, 1990). Metal enriched seston poses a dietary exposure risk to filter-feeding macroinvertebrates that entrap and ingest suspended materials as a primary food source. Suspended particulate material and dissolved and total recoverable water samples were collected along a metal contamination gradient in 2017 and 2018 in the Clark Fork River and analyzed for metal concentrations to include the highly toxic metals arsenic, cadmium, and copper. Concentrations of seston arsenic (As), cadmium (Cd), and copper (Cu) ranged from 20.7–242 ug As/g, 2.7–16.2 ug Cd/g, and 129–1260 ug Cu/g. Dissolved (filtered) stream water concentrations ranged from 3.5–21.7 ug As/L, 0.1–0.38 ug Cd/L, and 1.5–12.5 ug Cu/L. Total recoverable (unfiltered) water concentrations ranged from 2.7–22.8 ug As/L, 0.1–0.35 ug Cd/L, and 2.0–14.9 ug Cu/L. Data presented here represent metal concentrations in water and seston from a mining-impacted river and provide insight to potential exposure of toxic metals to resident filter-feeding aquatic invertebrates.

In 2019, the U.S. Geological Survey (USGS), in cooperation with the National Park Service, initiated a study using surrogate technology to predict real-time metallic-contaminant concentrations (MCCs) in the Clark Fork at two USGS streamgages that bracket Grant-Kohrs Ranch National Historic Site (GRKO) near Deer Lodge, Montana. Clark Fork at Deer Lodge(streamgage 12324200), Mont., about one mile upstream from GRKO, and Clark Fork above Little Blackfoot River near Garrison (streamgage 12324400), Mont., about 12 miles downstream from GRKO property were instrumented with turbidity and acoustic sensors for monitoring the Clark Fork during National Park Service Superfund remediation activities. Time-series data from backscatter signals from fixed-point turbidity and acoustic sensors were correlated with discrete MCC samples collected from the Clark Fork and were used as surrogates for estimating real-time cadmium, copper, iron, lead, manganese, zinc, and the metalloid trace element arsenic. A stepwise regression approach was used to develop statistical models to predict MCCs based on instantaneous values of turbidity and acoustic backscatter. Simple linear regression models using turbidity as the sole explanatory variable produced the best models with R-squared values exceeding 0.90 in 9 of 12 models. Nash-Sutcliffe Efficiency values were used to evaluate the effectiveness of predictive models to approximate measured MCCs, and model biases were calculated as an additional check on model accuracy. The R-LOADEST statistical package was used to compute annual and daily metallic-contaminant loads along with 95-percent prediction intervals. R-LOADEST loads were compared to time-series computed loads to evaluate the applicability of time-series data for calculating daily and annual metallic-contaminant loads. Results from annual load estimates indicated an increase in loads for all metallic contaminants between the two monitoring sites. These results provided real-time information to National Park Service management for evaluating variation in water quality during Superfund remediation, comparing MCC values relative to aquatic life standards, and will help quantify benefits from Superfund remediation activities.

In 2019, the U.S. Geological Survey (USGS), in cooperation with the National Park Service, initiated a study using surrogate technology to predict real-time metallic-contaminant concentrations (MCCs) in the Clark Fork at two USGS streamgages that bracket Grant-Kohrs Ranch National Historic Site (GRKO) near Deer Lodge, Montana. Clark Fork at Deer Lodge(streamgage 12324200), Mont., about one mile upstream from GRKO, and Clark Fork above Little Blackfoot River near Garrison (streamgage 12324400), Mont., about 12 miles downstream from GRKO property were instrumented with turbidity and acoustic sensors for monitoring the Clark Fork during National Park Service Superfund remediation activities. Time-series data from backscatter signals from fixed-point turbidity and acoustic sensors were correlated with discrete MCC samples collected from the Clark Fork and were used as surrogates for estimating real-time cadmium, copper, iron, lead, manganese, zinc, and the metalloid trace element arsenic. A stepwise regression approach was used to develop statistical models to predict MCCs based on instantaneous values of turbidity and acoustic backscatter. Simple linear regression models using turbidity as the sole explanatory variable produced the best models with R-squared values exceeding 0.90 in 9 of 12 models. Nash-Sutcliffe Efficiency values were used to evaluate the effectiveness of predictive models to approximate measured MCCs, and model biases were calculated as an additional check on model accuracy. The R-LOADEST statistical package was used to compute annual and daily metallic-contaminant loads along with 95-percent prediction intervals. R-LOADEST loads were compared to time-series computed loads to evaluate the applicability of time-series data for calculating daily and annual metallic-contaminant loads. Results from annual load estimates indicated an increase in loads for all metallic contaminants between the two monitoring sites. These results provided real-time information to National Park Service management for evaluating variation in water quality during Superfund remediation, comparing MCC values relative to aquatic life standards, and will help quantify benefits from Superfund remediation activities.

These data are comprised of measurements of elements (e.g., uranium, cobalt, nickel, copper, zinc, cadmium, lead, etc.) in Canyon Mine containment pond invertebrate samples collected in calendar year 2019.

These data are comprised of measurements of elements (e.g., uranium, cobalt, nickel, copper, zinc, cadmium, lead, etc.) in Canyon Mine containment pond invertebrate samples collected in calendar year 2019.