Skip to main content
U.S. flag

An official website of the United States government

Assessing Water Quality from Highway Runoff at Selected Sites in North Carolina with the Stochastic Empirical Loading and Dilution Model (SELDM)

September 25, 2019

In 2013, the U.S. Geological Survey (USGS) in partnership with the U.S. Federal Highway Administration (FHWA) published a new national stormwater quality model called the Stochastic Empirical Loading Dilution Model (SELDM). The model is optimized for roadway projects but in theory can be applied to a broad range of development types. SELDM is a statistically-based empirical model pre-populated with much of the data required to run the application successfully. The model uses Monte Carlo methods as opposed to deterministic methods to generate a wide range of possible discharge and precipitation inputs from the upstream basin and highway site. SELDM is particularly useful for stormwater managers in its ability to provide the statistical probability of a water-quality standard exceedance downstream of a stormwater discharge location over the course of a long period of record. SELDM supports a variety of best management practices (BMPs), which the user can apply to the stormwater discharge and evaluate the subsequent instream water-quality benefit. This functionality makes the model well suited for supporting BMP-specific cost/benefit analyses. In 2015, the North Carolina Department of Transportation (NCDOT) initiated a partnership with USGS to enhance the national SELDM model with additional NC-specific data to improve the model's predictive performance across the State. Specific USGS data incorporated into the NC-enhanced SELDM model included selected North Carolina streamflow data as well as water-quality transport curves for selected constituents. Streamflow statistics (based on data through the 2015 water year) were computed for 266 selected continuous-record streamgages and updated in the StreamStatsDB, which is accessible to users from the USGS StreamStats application for North Carolina. Instantaneous streamflow data available at 30 selected continuous-record streamgages across North Carolina with drainage areas ranging from 4.12 to 63.3 square miles were used to develop site-specific recession ratio statistics at these sites. Using data through the 2016 water year, water-quality transport curves were developed for 27 streamgages for the following constituents: suspended sediment concentration, total nitrogen, total phosphorus, turbidity, copper, lead, and zinc. The NCDOT identified NC highway-runoff water-quality and quantity data available from non-USGS sources (compiled by a private NCDOT contractor). These data were reviewed by USGS and -- where deemed acceptable --were uploaded into the FHWA Highway-Runoff Database, the data warehouse and preprocessor for SELDM (Granato and Cazenas, 2009; Smith and Granato, 2010). Based on the analysis techniques documented by Granato (2014) in a national BMP study and using available water-quality sample data from selected highway-runoff and BMP site pairs, performance data from the NC highway-runoff research reports were analyzed and incorporated into the NC SELDM model for three BMP types. As part of the NC SELDM study, simulations using the NC SELDM model were completed to demonstrate example impacts and evaluations of stormwater runoff on downstream water quality at six hypothetical upstream basin and highway sites across the State.

Publication Year 2019
Title Assessing Water Quality from Highway Runoff at Selected Sites in North Carolina with the Stochastic Empirical Loading and Dilution Model (SELDM)
DOI 10.5066/F7V69HV8
Authors John C Weaver, Gregory E Granato, Sharon A Fitzgerald
Product Type Data Release
Record Source USGS Digital Object Identifier Catalog
USGS Organization South Atlantic Water Science Center