This model archive summary documents the suspended-sediment concentration (SSC) model developed to estimate 15-minute SSC at Muddy Creek above Paonia Reservoir, U.S. Geological Survey (USGS) site number 385903107210800. The methods used follow USGS guidance as referenced in relevant Office of Surface Water Technical Memorandum (TM) 2016.07 and Office of Water Quality TM 2016.10, and USGS Techniques and Methods, book 3, chap. C5 (Landers and others, 2016). A total of 438 suspended-sediment samples were collected during the calibration period. Forty-one of these samples (22 equal-width-interval [EWI] samples and 19 single-point pump samples) were used in the model calibration dataset. These 41 samples were collected over the range of observed streamflow, Sediment Corrected Backscatter (SCB), and Sediment Attenuation Coefficient (SAC) conditions. Samples used in calibration were plotted on duration curve plots for streamflow from March 2005 to November 2016 (Colorado Division of Water Resources data from 2005 to 2014, and USGS data for 2015–16), and SAC for the period of record. The plots indicate that samples were collected for the observed range of conditions at the site. Suspended-sediment concentrations at this site were computed from a calibrated regression model between SSC and SAC. Streamflow, SCB, dummy variables, and seasonality were also examined as potential variables. An ordinary least squares linear regression model was developed using the ‘stats’ and ‘smwrStats’ packages in R (R Core Team, 2018). Streamflow, SCB, SAC, dummy variable, and seasonality were examined as potential explanatory variables for estimating SSC. A square root transformed SAC was selected as the explanatory variable.

From 2008 to 2012, the U.S. Geological Survey, in cooperation with the U.S. Army Corps of Engineers, Walla Walla District, developed an acoustic backscatter surrogate (model 1.0) for estimating real-time suspended-sediment concentration and loads at Clearwater River at Spalding, ID (USGS ID: 13342500) using a horizontally-mounted 3000kHz acoustic Doppler velocity meter (ADVM). This study is a continuation of the 2008 to 2012 acoustic backscatter surrogate study using samples collected since 2015 to validate the continued use of model 1.0. Sample data collected in 2015 to 2018 show a deviation from model 1.0 when backscatter data from the ADVM, after correction for acoustic losses, is greater than 65.7 decibels. The model archive summary documents acoustic backscatter surrogate model 2.0 for the real-time calculation of suspended-sediment concentration at Clearwater River at Spalding, ID (USGS ID: 13342500). Model 2.0 is in effect from March 24, 2015 to September 30, 2019 (end of the study). The methods used in the study follow U.S. Geological Survey protocols as referenced in relevant Office of Surface Water/Office of Water Quality Technical Memoranda and U.S. Geological Survey Techniques and Methods reports.

From 2008 to 2012, the U.S. Geological Survey, in cooperation with the U.S. Army Corps of Engineers, Walla Walla District, developed an acoustic backscatter surrogate (model 1.0) for estimating real-time suspended-sediment concentration and loads at Clearwater River at Spalding, ID (USGS ID: 13342500) using a horizontally-mounted 3000kHz acoustic Doppler velocity meter (ADVM). This study is a continuation of the 2008 to 2012 acoustic backscatter surrogate study using samples collected since 2015 to validate the continued use of model 1.0. Sample data collected in 2015 to 2018 show a deviation from model 1.0 when backscatter data from the ADVM, after correction for acoustic losses, is greater than 65.7 decibels. The model archive summary documents acoustic backscatter surrogate model 2.0 for the real-time calculation of suspended-sediment concentration at Clearwater River at Spalding, ID (USGS ID: 13342500). Model 2.0 is in effect from March 24, 2015 to September 30, 2019 (end of the study). The methods used in the study follow U.S. Geological Survey protocols as referenced in relevant Office of Surface Water/Office of Water Quality Technical Memoranda and U.S. Geological Survey Techniques and Methods reports.

A series of linear regression models were developed and calibrated for two Lower Minnesota River sites. The linear regression models were either calibrated using acoustic or streamflow data to estimate suspended-sediment or sand concentration data. Data were collected during calendar years 2012 through 2019. The estimates of suspended-sediment and concentrations from the linear regression were used to calculate loads. The calibrated models were used to improve understanding of sediment and sand transport processes and increase accuracy of estimating sediment and sand concentrations and loads for the Lower Minnesota River, as part of the associated report, U.S. Geological Survey Open File Report 2021–1005 (https://doi.org/10.3133/ofr20211005).

A series of linear regression models were developed and calibrated for two Lower Minnesota River sites. The linear regression models were either calibrated using acoustic or streamflow data to estimate suspended-sediment or sand concentration data. Data were collected during calendar years 2012 through 2019. The estimates of suspended-sediment and concentrations from the linear regression were used to calculate loads. The calibrated models were used to improve understanding of sediment and sand transport processes and increase accuracy of estimating sediment and sand concentrations and loads for the Lower Minnesota River, as part of the associated report, U.S. Geological Survey Open File Report 2021–1005 (https://doi.org/10.3133/ofr20211005).

Datasets of suspended sediment concentration and percent fines, sampling information, and daily streamflow data were compiled and harmonized for 16 sites to better understand sediment transport and delivery in the Lower Mississippi and Atchafalaya Rivers. The compiled data were harmonized by removing unnecessary columns, screening data for laboratory or sampling issues, creating consistent entries for character columns, and dropping irrelevant data, among other steps. Fourteen of the sites are in the Lower Mississippi-Atchafalaya River Basin with two additional sites on the Middle Mississippi and Ohio Rivers. Suspended sediment concentration (total, all size fractions) and percent fines for multiple size fractions were retrieved from the U.S. Geological Survey (USGS) National Water Information System (NWIS) database. These data were matched to related sampling information, such as sampler type and sampling method, also retrieved from NWIS. Continuous daily streamflow was compiled (or estimated where missing) for all sites and these data were from NWIS and the U.S. Army Corps of Engineers (USACE). Daily streamflow records extend as far back as possible and contain no gaps, whereas suspended sediment data and sampling information were measured and reported periodically and may contain multiyear gaps depending on the site. Note, siteIndex is used as the main sediment site identifier since sediment records from more than one USGS site are combined for at least one siteIndex. Additionally, gageIndex is used as the main streamgage identifier since streamflow records from multiple streamgages are sometimes combined for a single gageIndex and a single gageIndex may be used for more than one siteIndex. See the siteTable.csv for linkages between siteIndex, gageIndex, and USGS and USACE site/streamgage numbers.

Datasets of suspended sediment concentration and percent fines, sampling information, and daily streamflow data were compiled and harmonized for 16 sites to better understand sediment transport and delivery in the Lower Mississippi and Atchafalaya Rivers. The compiled data were harmonized by removing unnecessary columns, screening data for laboratory or sampling issues, creating consistent entries for character columns, and dropping irrelevant data, among other steps. Fourteen of the sites are in the Lower Mississippi-Atchafalaya River Basin with two additional sites on the Middle Mississippi and Ohio Rivers. Suspended sediment concentration (total, all size fractions) and percent fines for multiple size fractions were retrieved from the U.S. Geological Survey (USGS) National Water Information System (NWIS) database. These data were matched to related sampling information, such as sampler type and sampling method, also retrieved from NWIS. Continuous daily streamflow was compiled (or estimated where missing) for all sites and these data were from NWIS and the U.S. Army Corps of Engineers (USACE). Daily streamflow records extend as far back as possible and contain no gaps, whereas suspended sediment data and sampling information were measured and reported periodically and may contain multiyear gaps depending on the site. Note, siteIndex is used as the main sediment site identifier since sediment records from more than one USGS site are combined for at least one siteIndex. Additionally, gageIndex is used as the main streamgage identifier since streamflow records from multiple streamgages are sometimes combined for a single gageIndex and a single gageIndex may be used for more than one siteIndex. See the siteTable.csv for linkages between siteIndex, gageIndex, and USGS and USACE site/streamgage numbers.

This model archive summary documents the suspended-sediment concentration (SSC) model developed to estimate 15-minute SSC at North Fork Gunnison River below Raven Gulch, U.S. Geological Survey (USGS) site number 385553107243301. The methods used follow USGS guidance as referenced in relevant Office of Surface Water/Office of Water Quality Technical Memoranda and USGS Techniques and Methods, book 3, chap. C5 (Landers and others, 2016), and USGS Techniques and Methods, book 3, chap. C4 (Rasmussen and others, 2009). A total of 456 suspended-sediment samples were collected during the calibration period (45 cross-section and 411 single-station automatic pump samples). Thirty-nine samples (18 pump samples and 21 equal-width-interval (EWI) samples) with associated streamflow and turbidity were used in the model calibration dataset. These 39 samples were collected over the range of observed streamflow and turbidity conditions. Samples used in calibration were plotted on duration curve plots for streamflow from April 24, 2015 to October 7, 2017 and turbidity from April 24, 2015 to October 7, 2017. The plots indicate that samples were collected for the observed range of conditions at the site. Suspended-sediment concentrations at this site were computed from a calibrated regression model between SSC and turbidity. Streamflow, 2 frequencies of sediment corrected backscatter (SCB) (1.5 and 3.0 megahertz (MHz)), and 2 frequencies of sediment attenuation coefficient (SAC) (1.5 and 3.0 MHz) were also examined as potential variables but did not significantly improve the model. An ordinary least squares linear regression model was developed using the ‘stats’ and ‘smwrStats’ packages in R (R Core Team, 2018). Streamflow, SCB, SAC, and turbidity were examined as potential explanatory variables for estimating SSC. A natural log transformed turbidity was selected as the best explanatory variable.

This model archive summary documents the suspended-sediment concentration (SSC) model developed to estimate 15-minute SSC at North Fork Gunnison River below Raven Gulch, U.S. Geological Survey (USGS) site number 385553107243301. The methods used follow USGS guidance as referenced in relevant Office of Surface Water/Office of Water Quality Technical Memoranda and USGS Techniques and Methods, book 3, chap. C5 (Landers and others, 2016), and USGS Techniques and Methods, book 3, chap. C4 (Rasmussen and others, 2009). A total of 456 suspended-sediment samples were collected during the calibration period (45 cross-section and 411 single-station automatic pump samples). Thirty-nine samples (18 pump samples and 21 equal-width-interval (EWI) samples) with associated streamflow and turbidity were used in the model calibration dataset. These 39 samples were collected over the range of observed streamflow and turbidity conditions. Samples used in calibration were plotted on duration curve plots for streamflow from April 24, 2015 to October 7, 2017 and turbidity from April 24, 2015 to October 7, 2017. The plots indicate that samples were collected for the observed range of conditions at the site. Suspended-sediment concentrations at this site were computed from a calibrated regression model between SSC and turbidity. Streamflow, 2 frequencies of sediment corrected backscatter (SCB) (1.5 and 3.0 megahertz (MHz)), and 2 frequencies of sediment attenuation coefficient (SAC) (1.5 and 3.0 MHz) were also examined as potential variables but did not significantly improve the model. An ordinary least squares linear regression model was developed using the ‘stats’ and ‘smwrStats’ packages in R (R Core Team, 2018). Streamflow, SCB, SAC, and turbidity were examined as potential explanatory variables for estimating SSC. A natural log transformed turbidity was selected as the best explanatory variable.

This model archive summary (MAS) and associated model calibration data describes a surrogate model used to compute suspended sediment concentrations at USGS streamgage 11516530 for water years 2019-2023. The methods used follow USGS guidance as referenced in Office of Surface Water (OSW)/Office of Water Quality (OWQ) technical memoranda 2016.07 (https://water.usgs.gov/osw/pubs/memo.summaries.html) and USGS techniques and methods, book 3-chapter C4 (doi.org/10.3133/tm3C4).