This data release supports an analysis of changes in dissolved organic carbon (DOC) concentration and in the slope of the concentration-discharge relation in the Sleepers River Research Watershed W9 near North Danville, Vermont 1991 to 2018. The data release includes pre-processed model inputs and model outputs. W9 is a 40.5-hectare forested sub-watershed of the Sleepers River Research Watershed that was originally established by the U.S. Department of Agriculture in 1959 and has been intensively studied by the USGS since 1991. The W9 headwater sub-watershed drains to Pope Brook and is gaged at USGS station ID 01135100. Pope Brook flows into Sleepers River and ultimately into the Connecticut River. Two models were used to analyze concentration-discharge (C-Q) relations: the Weighted Regressions on Time, Discharge and Season (WRTDS) model (Hirsch and others, 2010) and the Hydrograph Separation (HYSEP) model (Sloto and Crouse, 1996). The WRTDS model used discharge data downloaded from Shanley and others (2021) and DOC data downloaded from Matt and others (2021). The HYSEP model used discharge data downloaded from Shanley and others (2021). The WRTDS model was used to estimate daily concentrations and fluxes of DOC and the HYSEP model was used to estimate the fraction of daily discharge that could be classified as stormflow (event flow) as opposed to base flow. The time series data were analyzed for the entire period, water years 1992 (WY1992) to WY2018 where WY1992 is the period from October 1, 1991 to September 30, 1992. In addition to the WRTDS and HYSEP modeling conducted with the full range of data, two further analyses were conducted with subsets of the data. For one evaluation, the data were separated into two periods (with a breakpoint between WY2002 and WY2003) to detect and quantify changes in DOC concentration and slope of the C-Q relation over time. In a separate analysis, WRTDS was used to estimate DOC concentrations for the observed WY2007 daily discharge record when this record was inserted as WY1997 as a synthetic record. In this synthetic record the original water years 1997 through 2002 were incremented by one year. (Note: the input and output data files that contain "syn" in the filenames include this synthetic or modified daily discharge data and therefore would not match the observed records.) This data release is structured as a set of comma-separated values (CSV) files, including a site file (Site_Information.csv), 13 data files, and a data dictionary (Data_dictionary.csv), which provides information describing the content of each of the data files. The data dictionary contains the filenames of all 15 CSV files and their, headings, data types, descriptions, and units, and basic statistics where applicable. The site-information file contains the USGS 8-digit site identifier, site name, latitude, longitude, horizontal datum, land surface elevation, vertical datum, and drainage area for the Pope Brook Tributary (W9). The W9Info2.csv file is used by WRTDS to assign the appropriate information for the analysis for this watershed. There are four pairs of input files. There are two input files for each WRTDS run including a daily discharge file and a daily DOC concentration file, for example "W9Daily.csv" and "W9Sample.csv" for the full period 10/1/1991 to 9/30/2018 for discharge and concentration, respectively. The other input file pairs are for the date ranges given in the file names: "W9Daily1992.2002.csv" and "W9Sample1992.2002.2.7.csv"; "W9Daily2003.2018.csv" and "W9Sample2003.2018.2.7.csv"; "W9Daily1992.2003syn.csv and "W9Sample1992.2003syn.2.7.csv". There is one WRTDS output file for each of the four pairs of input files: "DailyOutput_2.7.csv" (full period); "DailyOutput1992.2002.2.7.csv" (early period); "DailyOutput2003.2013.2.7.csv" (recent period); "DailyOutput1992.2003syn.2.7.csv" (early period with identical hydrologic record spliced following WY1996). There is one output file from the HYSEP model:

This data release supports an analysis of changes in dissolved organic carbon (DOC) concentration and in the slope of the concentration-discharge relation in the Sleepers River Research Watershed W9 near North Danville, Vermont 1991 to 2018. The data release includes pre-processed model inputs and model outputs. W9 is a 40.5 hectares forested sub-watershed of the Sleepers River Research Watershed that was originally established by the U.S. Department of Agriculture in 1959 and has been intensively studied by the USGS since 1991. The W9 headwater sub-watershed drains to Pope Brook and is gaged at USGS station ID 01135100. Pope Brook flows into Sleepers River and ultimately into the Connecticut River. Two models were used to analyze concentration-discharge relations: the Weighted Regressions on Time, Discharge and Season (WRTDS) model (Hirsch and others, 2010) and the Hydrograph Separation model (HYSEP) (Sloto and Crouse, 1996). The WRTDS model used discharge data downloaded from Shanley and others (2021) and DOC data downloaded from Matt and others (2021). The HYSEP model used discharge data downloaded from Shanley and others (2021). The WRTDS model was used to estimate daily concentrations and fluxes of DOC and the HYSEP model was used to estimate the fraction of daily discharge that could be classified as stormflow (event flow) as opposed to base flow. In a separate analysis, regression models were used to estimate DOC concentrations for an identical average daily discharge hydrologic record. The observed calendar year 2007 record was chosen for this analysis. The average daily discharge in 2007 was similar to the long-term average daily discharge, the timing of the spring runoff was closely aligned with the average, there was a characteristic period of low flow in late summer, and there were several typical large runoff events in early summer and fall. Daily discharge for calendar year 2007 was inserted into the original discharge record after calendar year 1997 (early period) or 2010 (recent period) using dates from 1998 or 2011. The original data from the second half of each of these periods (January 1,1998 through May 16, 2006) and (January 1, 2011 through December 31, 2018), respectively, were incremented by one year. Therefore, in this analysis the dates for the early period spanned 9/30/1991 to 5/16/2006 and the dates for the recent period spanned 5/17/2006 to 12/31/2019. Note: the input and output data files that contain 2007 in the filenames include this synthetic or modified daily discharge data and therefore would not match the observed records. This data release is structured as a set of comma-separated values (CSV) files, including a site file (Site_Information.csv), 17 data files, and a data dictionary (Data_Dictionary.csv), which provides information describing the content of each of the data files. The data dictionary contains the filename, data type, description, and units. The site-information file contains the USGS 8-digit site identifier, site name, latitude, longitude, horizontal datum, land surface elevation, vertical datum, and drainage area. There are five pairs of input files. There are two input files for each WRTDS run including a daily discharge file and a daily DOC concentration file, for example "W9Daily.csv" and "W9Sample.csv" for the full period 9/30/1991 to 12/31/2018 for discharge and concentration, respectively. The other input file pairs are "W9DailyEarly.csv" and "W9DailyRecent.csv" and "W9DailyEarly2007.csv" and "W9DailyRecent2007.csv." The last two files contain the identical hydrologic record spliced into the middle of each period. There is one WRTDS output file for each of the paired input and output files noted: "DailyOutput_W9.csv" (full period); "DailyOutput_W9Early.csv" (early period); "DailyOutput_W9Recent.csv" (recent period); "DailyOutput_W9Early20072.csv" (early period with identical hydrologic record spliced into the middle); "DailyOutput_W9Recent20073.csv" (recent period with

This data release supports an analysis of changes in dissolved organic carbon (DOC) concentration and in the slope of the concentration-discharge relation in the Sleepers River Research Watershed W9 near North Danville, Vermont 1991 to 2018. The data release includes pre-processed model inputs and model outputs. W9 is a 40.5 hectares forested sub-watershed of the Sleepers River Research Watershed that was originally established by the U.S. Department of Agriculture in 1959 and has been intensively studied by the USGS since 1991. The W9 headwater sub-watershed drains to Pope Brook and is gaged at USGS station ID 01135100. Pope Brook flows into Sleepers River and ultimately into the Connecticut River. Two models were used to analyze concentration-discharge relations: the Weighted Regressions on Time, Discharge and Season (WRTDS) model (Hirsch and others, 2010) and the Hydrograph Separation model (HYSEP) (Sloto and Crouse, 1996). The WRTDS model used discharge data downloaded from Shanley and others (2021) and DOC data downloaded from Matt and others (2021). The HYSEP model used discharge data downloaded from Shanley and others (2021). The WRTDS model was used to estimate daily concentrations and fluxes of DOC and the HYSEP model was used to estimate the fraction of daily discharge that could be classified as stormflow (event flow) as opposed to base flow. In a separate analysis, regression models were used to estimate DOC concentrations for an identical average daily discharge hydrologic record. The observed calendar year 2007 record was chosen for this analysis. The average daily discharge in 2007 was similar to the long-term average daily discharge, the timing of the spring runoff was closely aligned with the average, there was a characteristic period of low flow in late summer, and there were several typical large runoff events in early summer and fall. Daily discharge for calendar year 2007 was inserted into the original discharge record after calendar year 1997 (early period) or 2010 (recent period) using dates from 1998 or 2011. The original data from the second half of each of these periods (January 1,1998 through May 16, 2006) and (January 1, 2011 through December 31, 2018), respectively, were incremented by one year. Therefore, in this analysis the dates for the early period spanned 9/30/1991 to 5/16/2006 and the dates for the recent period spanned 5/17/2006 to 12/31/2019. Note: the input and output data files that contain 2007 in the filenames include this synthetic or modified daily discharge data and therefore would not match the observed records. This data release is structured as a set of comma-separated values (CSV) files, including a site file (Site_Information.csv), 17 data files, and a data dictionary (Data_Dictionary.csv), which provides information describing the content of each of the data files. The data dictionary contains the filename, data type, description, and units. The site-information file contains the USGS 8-digit site identifier, site name, latitude, longitude, horizontal datum, land surface elevation, vertical datum, and drainage area. There are five pairs of input files. There are two input files for each WRTDS run including a daily discharge file and a daily DOC concentration file, for example "W9Daily.csv" and "W9Sample.csv" for the full period 9/30/1991 to 12/31/2018 for discharge and concentration, respectively. The other input file pairs are "W9DailyEarly.csv" and "W9DailyRecent.csv" and "W9DailyEarly2007.csv" and "W9DailyRecent2007.csv." The last two files contain the identical hydrologic record spliced into the middle of each period. There is one WRTDS output file for each of the paired input and output files noted: "DailyOutput_W9.csv" (full period); "DailyOutput_W9Early.csv" (early period); "DailyOutput_W9Recent.csv" (recent period); "DailyOutput_W9Early20072.csv" (early period with identical hydrologic record spliced into the middle); "DailyOutput_W9Recent20073.csv" (recent period with

This data release supports an analysis of changes in dissolved organic carbon (DOC) concentration and in the slope of the concentration-discharge relation in the Sleepers River Research Watershed W9 near North Danville, Vermont 1991 to 2018. The data release includes pre-processed model inputs and model outputs. W9 is a 40.5 hectares forested sub-watershed of the Sleepers River Research Watershed that was originally established by the U.S. Department of Agriculture in 1959 and has been intensively studied by the USGS since 1991. The W9 headwater sub-watershed drains to Pope Brook and is gaged at USGS station ID 01135100. Pope Brook flows into Sleepers River and ultimately into the Connecticut River. Two models were used to analyze concentration-discharge relations: the Weighted Regressions on Time, Discharge and Season (WRTDS) model (Hirsch and others, 2010) and the Hydrograph Separation model (HYSEP) (Sloto and Crouse, 1996). The WRTDS model used discharge data downloaded from Shanley and others (2021) and DOC data downloaded from Matt and others (2021). The HYSEP model used discharge data downloaded from Shanley and others (2021). The WRTDS model was used to estimate daily concentrations and fluxes of DOC and the HYSEP model was used to estimate the fraction of daily discharge that could be classified as stormflow (event flow) as opposed to base flow. In a separate analysis, regression models were used to estimate DOC concentrations for an identical average daily discharge hydrologic record. The observed calendar year 2007 record was chosen for this analysis. The average daily discharge in 2007 was similar to the long-term average daily discharge, the timing of the spring runoff was closely aligned with the average, there was a characteristic period of low flow in late summer, and there were several typical large runoff events in early summer and fall. Daily discharge for calendar year 2007 was inserted into the original discharge record after calendar year 1997 (early period) or 2010 (recent period) using dates from 1998 or 2011. The original data from the second half of each of these periods (January 1,1998 through May 16, 2006) and (January 1, 2011 through December 31, 2018), respectively, were incremented by one year. Therefore, in this analysis the dates for the early period spanned 9/30/1991 to 5/16/2006 and the dates for the recent period spanned 5/17/2006 to 12/31/2019. Note: the input and output data files that contain 2007 in the filenames include this synthetic or modified daily discharge data and therefore would not match the observed records. This data release is structured as a set of comma-separated values (CSV) files, including a site file (Site_Information.csv), 17 data files, and a data dictionary (Data_Dictionary.csv), which provides information describing the content of each of the data files. The data dictionary contains the filename, data type, description, and units. The site-information file contains the USGS 8-digit site identifier, site name, latitude, longitude, horizontal datum, land surface elevation, vertical datum, and drainage area. There are five pairs of input files. There are two input files for each WRTDS run including a daily discharge file and a daily DOC concentration file, for example "W9Daily.csv" and "W9Sample.csv" for the full period 9/30/1991 to 12/31/2018 for discharge and concentration, respectively. The other input file pairs are "W9DailyEarly.csv" and "W9DailyRecent.csv" and "W9DailyEarly2007.csv" and "W9DailyRecent2007.csv." The last two files contain the identical hydrologic record spliced into the middle of each period. There is one WRTDS output file for each of the paired input and output files noted: "DailyOutput_W9.csv" (full period); "DailyOutput_W9Early.csv" (early period); "DailyOutput_W9Recent.csv" (recent period); "DailyOutput_W9Early20072.csv" (early period with identical hydrologic record spliced into the middle); "DailyOutput_W9Recent20073.csv" (recent period with

This data release provides data in support of an assessment of changes in slope of the concentration discharge relation for total organic carbon in eight major rivers at gaging stations in Connecticut from 1972 to 2019. The rivers include the Connecticut at Thompsonville (01184000), Housatonic at Stevenson (01205500), Quinebaug at Jewett City (01127000), Farmington at Tariffville (01189995), Shetucket at Willamantic (01122500), Naugatuck at Beacon Falls (011208500), Quinnipiac at Wallingford (01196500), and Salmon at East Hampton (01193500). The assessment is described in the article “An increase in the slope of the concentration discharge relation for total organic carbon in major rivers in New England, 1973 to 2019” (Huntington and Wieczorek, in review). Two models were used in this assessment; the Weighted Regressions on Time, Discharge and Season (WRTDS) model (Hirsch and others, 2010), and the Hydrograph Separation (HYSEP model) (Sloto and Crouse, 1996). Both models use data from the USGS National Water Information System (NWIS) database (USGS, 2020). The WRTDS model was used to estimate daily concentrations of various water quality constituents and the HYSEP model was used to estimate the fraction of daily discharge that could be classified as storm flow (event flow) as opposed to base flow. In a separate analysis, for the Connecticut, Farmington, and Quinnipiac Rivers, independent regression models for the early and recent periods were used to compare predicted TOC concentrations given a common daily discharge record for one year. A synthetic record was developed from the average daily discharges for these rivers for the period 10/1/1992 to 9/30/1998. To run the appropriate WRTDS model with this discharge record the dates for this synthetic record were set to 10/1/1983 to 9/30/1984 when run with the early period model and 10/1/2005 to 9/30/2006 when run with the recent period model. Using these synthetic discharge periods centered the synthetic year in the middle of each period thus controlling for any temporal trends in the C-Q relations during the periods. On the few days of the year during those periods when samples were collected for TOC analyses, the observed discharge in the actual record was substituted for the synthetic discharge so the original concentration discharge data pair would be used in the regression model. This data release is structured as a set of comma-separated values (CSV) files, including a site file, 44 data files, and a data dictionary (Data_Dictionary.csv), which provides information describing the content of each of the data files. The data dictionary contains the filename, data type, description, and units. The site-information file contains the USGS 8-digit site ID, site name, latitude, longitude, horizontal datum, land surface elevation, vertical datum, and drainage area. The data files contain input data for one set of model runs and output data for all model runs of the WRTDS model as follows: • There are six input files, one for the early period (1973 to 1995) and one for the recent period (1996 to 2019) for three rivers. These input files contain two columns (date and daily discharge). The file naming convention begins with WRTDS_INPUT_DAILY_Discharge_ and is followed by the date range and the USGS site ID for the specific river. For example, “WRTDS_INPUT_DAILY_Discharge_1996-2019_01196500.csv” is for the recent period for the Quinnipiac River at Wallingford • There are eight WRTDS model output files (one for each of the eight rivers) for the full period (1973 to 2019) that contain date, daily discharge, and estimated concentrations of total organic carbon, total nitrogen, total phosphorus, and sulfate. The file naming convention begins with WRTDS_Output_Daily_1973-2019_ and is followed by the USGS site ID for the specific river. For example, “WRTDS_Output_Daily_1973-2019_01184000.csv” is for the full period for the Connecticut River at Thompsonville. • There are eight WRTDS model

This data release provides data in support of an assessment of changes in slope of the concentration discharge relation for total organic carbon in eight major rivers at gaging stations in Connecticut from 1972 to 2019. The rivers include the Connecticut at Thompsonville (01184000), Housatonic at Stevenson (01205500), Quinebaug at Jewett City (01127000), Farmington at Tariffville (01189995), Shetucket at Willamantic (01122500), Naugatuck at Beacon Falls (011208500), Quinnipiac at Wallingford (01196500), and Salmon at East Hampton (01193500). The assessment is described in the article “An increase in the slope of the concentration discharge relation for total organic carbon in major rivers in New England, 1973 to 2019” (Huntington and Wieczorek, in review). Two models were used in this assessment; the Weighted Regressions on Time, Discharge and Season (WRTDS) model (Hirsch and others, 2010), and the Hydrograph Separation (HYSEP model) (Sloto and Crouse, 1996). Both models use data from the USGS National Water Information System (NWIS) database (USGS, 2020). The WRTDS model was used to estimate daily concentrations of various water quality constituents and the HYSEP model was used to estimate the fraction of daily discharge that could be classified as storm flow (event flow) as opposed to base flow. In a separate analysis, for the Connecticut, Farmington, and Quinnipiac Rivers, independent regression models for the early and recent periods were used to compare predicted TOC concentrations given a common daily discharge record for one year. A synthetic record was developed from the average daily discharges for these rivers for the period 10/1/1992 to 9/30/1998. To run the appropriate WRTDS model with this discharge record the dates for this synthetic record were set to 10/1/1983 to 9/30/1984 when run with the early period model and 10/1/2005 to 9/30/2006 when run with the recent period model. Using these synthetic discharge periods centered the synthetic year in the middle of each period thus controlling for any temporal trends in the C-Q relations during the periods. On the few days of the year during those periods when samples were collected for TOC analyses, the observed discharge in the actual record was substituted for the synthetic discharge so the original concentration discharge data pair would be used in the regression model. This data release is structured as a set of comma-separated values (CSV) files, including a site file, 44 data files, and a data dictionary (Data_Dictionary.csv), which provides information describing the content of each of the data files. The data dictionary contains the filename, data type, description, and units. The site-information file contains the USGS 8-digit site ID, site name, latitude, longitude, horizontal datum, land surface elevation, vertical datum, and drainage area. The data files contain input data for one set of model runs and output data for all model runs of the WRTDS model as follows: • There are six input files, one for the early period (1973 to 1995) and one for the recent period (1996 to 2019) for three rivers. These input files contain two columns (date and daily discharge). The file naming convention begins with WRTDS_INPUT_DAILY_Discharge_ and is followed by the date range and the USGS site ID for the specific river. For example, “WRTDS_INPUT_DAILY_Discharge_1996-2019_01196500.csv” is for the recent period for the Quinnipiac River at Wallingford • There are eight WRTDS model output files (one for each of the eight rivers) for the full period (1973 to 2019) that contain date, daily discharge, and estimated concentrations of total organic carbon, total nitrogen, total phosphorus, and sulfate. The file naming convention begins with WRTDS_Output_Daily_1973-2019_ and is followed by the USGS site ID for the specific river. For example, “WRTDS_Output_Daily_1973-2019_01184000.csv” is for the full period for the Connecticut River at Thompsonville. • There are eight WRTDS model

This data release supports an analysis of changes in dissolved organic carbon (DOC) and nitrate concentrations in Buck Creek watershed near Inlet, New York 2001 to 2021. The Buck Creek watershed is a 310-hectare forested watershed that is recovering from acidic deposition within the Adirondack region. The data release includes pre-processed model inputs and model outputs for the Weighted Regressions on Time, Discharge and Season (WRTDS) model (Hirsch and others, 2010) to estimate daily flow normalized concentrations of DOC and nitrate during a 20-year period of analysis. WRTDS uses daily discharge and concentration observations implemented through the Exploration and Graphics for River Trends R package (EGRET) to predict solute concentration using decimal time and discharge as explanatory variables (Hirsch and De Cicco, 2015; Hirsch and others, 2010). Discharge and concentration data are available from the U.S. Geological Survey National Water Information System (NWIS) database (U.S. Geological Survey, 2016). The time series data were analyzed for the entire period, water years 2001 (WY2001) to WY2021 where WY2001 is the period from October 1, 2000 to September 30, 2001. This data release contains 5 comma-separated values (CSV) files, one R script, and one XML metadata file. There are four input files (“Daily.csv”, “INFO.csv”, “Sample_doc.csv”, and “Sample_nitrate.csv”) that contain site information, daily mean discharge, and mean daily DOC or nitrate concentrations. The R script (“Buck Creek WRTDS R script.R”) uses the four input datasets and functions from the EGRET R package to generate estimations of flow normalized concentrations. The output file (“WRTDS_results.csv”) contains model output at daily time steps for each sub-watershed and for each solute. Files are automatically associated with the R script when opened in RStudio using the provided R project file ("Files.Rproj"). All input, output, and R files are in the "Files.zip" folder.

This data release supports an analysis of changes in dissolved organic carbon (DOC) and nitrate concentrations in Buck Creek watershed near Inlet, New York 2001 to 2021. The Buck Creek watershed is a 310-hectare forested watershed that is recovering from acidic deposition within the Adirondack region. The data release includes pre-processed model inputs and model outputs for the Weighted Regressions on Time, Discharge and Season (WRTDS) model (Hirsch and others, 2010) to estimate daily flow normalized concentrations of DOC and nitrate during a 20-year period of analysis. WRTDS uses daily discharge and concentration observations implemented through the Exploration and Graphics for River Trends R package (EGRET) to predict solute concentration using decimal time and discharge as explanatory variables (Hirsch and De Cicco, 2015; Hirsch and others, 2010). Discharge and concentration data are available from the U.S. Geological Survey National Water Information System (NWIS) database (U.S. Geological Survey, 2016). The time series data were analyzed for the entire period, water years 2001 (WY2001) to WY2021 where WY2001 is the period from October 1, 2000 to September 30, 2001. This data release contains 5 comma-separated values (CSV) files, one R script, and one XML metadata file. There are four input files (“Daily.csv”, “INFO.csv”, “Sample_doc.csv”, and “Sample_nitrate.csv”) that contain site information, daily mean discharge, and mean daily DOC or nitrate concentrations. The R script (“Buck Creek WRTDS R script.R”) uses the four input datasets and functions from the EGRET R package to generate estimations of flow normalized concentrations. The output file (“WRTDS_results.csv”) contains model output at daily time steps for each sub-watershed and for each solute. Files are automatically associated with the R script when opened in RStudio using the provided R project file ("Files.Rproj"). All input, output, and R files are in the "Files.zip" folder.

In 1991, the U.S. Geological Survey (USGS) began a study of more than 50 major river basins across the Nation as part of the National Water-Quality Assessment (NAWQA) project of the National Water-Quality Program. One of the major goals of the NAWQA project is to determine how water-quality conditions change over time. To support that goal, long-term consistent and comparable monitoring has been conducted on streams and rivers throughout the Nation. Outside of the NAWQA project, the USGS and other Federal, State, and local agencies also have collected long-term water-quality data to support their own assessments of changing water-quality conditions. Data from these multiple sources have been combined to support one of the most comprehensive assessment conducted to date of water-quality trends in the United States. Ultimately, these data will provide insight into how natural features and human activities have contributed to water-quality changes over time in Nation’s streams and rivers. This USGS data release contains all of the input and output files necessary to reproduce the results from the Weighted Regressions on Time, Discharge, and Season (WRTDS) models described in the associated U.S. Geological Survey Scientific Investigations Report. Data preparation for input to the models is also fully described in the above-mentioned report.

In 1991, the U.S. Geological Survey (USGS) began a study of more than 50 major river basins across the Nation as part of the National Water-Quality Assessment (NAWQA) project of the National Water-Quality Program. One of the major goals of the NAWQA project is to determine how water-quality conditions change over time. To support that goal, long-term consistent and comparable monitoring has been conducted on streams and rivers throughout the Nation. Outside of the NAWQA project, the USGS and other Federal, State, and local agencies also have collected long-term water-quality data to support their own assessments of changing water-quality conditions. Data from these multiple sources have been combined to support one of the most comprehensive assessment conducted to date of water-quality trends in the United States. Ultimately, these data will provide insight into how natural features and human activities have contributed to water-quality changes over time in Nation’s streams and rivers. This USGS data release contains all of the input and output files necessary to reproduce the results from the Weighted Regressions on Time, Discharge, and Season (WRTDS) models described in the associated U.S. Geological Survey Scientific Investigations Report. Data preparation for input to the models is also fully described in the above-mentioned report.