Predicting Forest Road Surface Erosion and Storm Runoff from High-Elevation Sites

Abstract

Abstract. Forest roads are a concern in management because they represent areas of elevated risks associated with soil erosion and storm runoff connectivity to stream systems. Storm runoff emanating from forest roads and their connectivity to downslope resources can be influenced by a myriad of factors, including storm characteristics, management practices, and the interaction of management practices and successive storm events. Mitigating sediment export and ensuring that storm runoff has negligible impacts on downstream resources requires a more complete understanding of forest road erosion and sediment delivery dynamics. Progress in the area of road and stream connectivity issues hinges on reliable prediction tools to inform broader-scale modeling of watershed-scale effects of forest roads and management practices. In this study, the Water Erosion Prediction Project (WEPP) model was evaluated based on the results from 156 runoff-generating storm events during a continuous five-year study of nine high-elevation road sections in the Appalachian Mountains. The model adequately predicted sediment yield from the road sections with an overall Nash-Sutcliffe model efficiency (E) of 0.76, Willmott refined index of agreement (dr) of 0.56, percent error of 5%, and average storm difference (ASD) of 1.2 kg. In contrast, WEPP predictions of storm runoff were not as good, and the poor agreement was attributed to an inability to determine the source area for runoff from some of the larger runoff events. In general, the WEPP model for these high-elevation sites adequately described the sediment yield for the road sections. Keywords: Forest roads, Long-term simulation, Runoff, Sediment, Water Erosion Modeling, WEPP.