Multivariate models of watershed suspended sediment loads for the eastern United States

Abstract

Estimates of mean annual watershed sediment load, derived from suspended sediment concentration and streamflow data, are often not available at locations of interest. The purpose of this study was to develop multivariate regression models of mean annual suspended sediment loads useful for most river locations in the eastern United States. The resulting models may be used at ungaged river locations to predict mean annual river sediment loads as a function of general basin characteristics. The analysis is based on long-term mean sediment load estimates and explanatory variables from a combined dataset of 1,201 USGS stations obtained from a SPAtially Referenced Regression On Watershed attributes (SPARROW) study and the Geospatial Attributes of Gages for Evaluating Streamflow (GAGES) database. The resulting regional regression models, summarized for major US water resources regions 1 through 8, and estimated in logarithmic space, exhibited prediction R2 values ranging from 76.9% to 92.7%. The results indicate that mean annual sediment loads in the eastern United States are generally influenced by a combination of basin area, land use patterns, seasonal precipitation, soil composition, hydrologic modification, and to a lesser extent, topography. © 2010 ASCE.