|
|
- Info
Forest road erosion, sediment transport, and model validation in the southern Appalachians
| Author: |
Riedel, M.S.; Vose, J.M. |
| Date: |
2002 |
| Periodical: |
Presented at: Second Federal Interagency Hydrologic Modeling Conference, July 28-August 1, 2002, Riviera Hotel, Las Vegas, Nevada. [Place of Publication Unknown]: [Publisher unknown]. 12 p. On file with: U.S. Department of Agriculture, Forest Service, Southern Research Station, Coweeta Hydrologic Laboratory. |
| Link: |
http://www.srs.fs.usda.gov/pubs/ja/ja_riedel002.pdf
|
| Abstract: |
The Conasauga River Watershed, located in northern Georgia and southern Tennessee, has one of the most diverse aquatic ecosystems in this region and is currently being considered for designation as a wild and scenic river. The Conasauga River also serves as a major source of drinking water for numerous large cities. Due to the close proximity with the cities of Knoxville, Atlanta, and Chattanooga, intensive public usage, and the high quality of this aquatic resource, the United States Department of Agriculture (USDA) Forest Service has designated the Conasauga River as one of the twelve large-scale watershed restoration projects in the nation. This is warranted as the Conasauga River is experiencing excessive sedimentation from the erosion of private agricultural lands, streambanks, and forest roads. We are working with an erosion model, the Sediment Tool, to facilitate decision-making in the restoration of forest roads. The sediment tool, and its parent model the Watershed Characterization System (WCS), were developed by the US Environmental Protection Agency (EPA). The Sediment Tool is a spatially explicit, GIS based, finite element, lumped parameter model which generates estimates of soil erosion, sediment routing and sediment yield. We applied WCS along segments of thirteen mountain roads in the Conasauga Watershed. The segments provide replication of road types under a variety of usage levels, road base materials and slopes. We sampled overland flow from each segment for total suspended solids (TSS) and surveyed all pertinent road characteristics. While we were able to qualitatively calibrate the model, predicted sediment yields were typically much greater than observed data. Model results improved with digital elevation model (DEM) and computational grid resolution. Error analysis indicated that model sensitivity is limited by the governing equations within the model and the resolution of the input data. The model currently employs the universal soil loss equation (USLE) to estimate soil erosion and empirical sediment yield equations to transport sediment. These empirical equations were not developed for application on aggregate road surfaces. DEM resolution will also present problems in routing the sediment to streams. Streams in the study areas are only one to three meters wide. Floodplains adjacent to these streams are typically four or five meters wide and frequently trap sediment-laden runoff before it reaches the streams. Current efforts to improve upon the model include an adaptation of the process based Water Erosion Prediction Project (WEPP) model and attainment of finer resolution DEM data that will more accurately represent the road surfaces. |
|
