Data and model archive summaries for estimating suspended-sediment concentrations in the Yellow River in Indiana, 2019-2025

This data release contains model-calibration datasets in comma-separated value (CSV) format and model archive summaries in portable document format (PDF) for regression models developed to estimate suspended-sediment concentrations (SSC) at three U.S. Geological Survey monitoring stations along the Yellow River in northwestern Indiana (05516665 Yellow River near Oak Grove, IN; 05517000 Yellow River at Knox, IN; 05517010 Yellow River near Brems, IN). SSC models were developed to estimate SSC at 15-minute intervals using continuous in situ turbidity data and discrete suspended sediment samples collected from April 2019 through August 2025. Ordinary least-squares regression analyses were conducted in R statistical software programming language (R Core Team, 2024) using turbidity and statistical transformations of turbidity as explanatory variables for estimating SSC and log-transformed SSC (log10(SSC)). Streamflow also was evaluated as a potential explanatory variable when available. Model selection was based on evaluation of residual plots, maximizing adjusted coefficient of determination (adjusted R²), and minimizing model residual standard error (root mean square error, RMSE), standard percentage error (MSPE), and Akaike information criterion (AIC). Model development followed U.S. Geological Survey (USGS) guidance described in Office of Surface Water Technical Memorandum 2016.07, Office of Water Quality Technical Memorandum 2016.10, and USGS Techniques and Methods, book 3, chap. C4 (Rasmussen and others, 2009; U.S. Geological Survey, 2016a and 2016b). All data were collected following USGS protocols described in the USGS National Field Manual for the Collection of Water-Quality Data (U.S. Geological Survey, variously dated), and Wagner and others, 2006. Data are stored in the USGS Water Data for the Nation (WDFN) database (U.S. Geological Survey, 2024).