Modeling the Effects of Reopening Abandoned Roads on Hydrology and Soil Loss in a Forest Watershed

Reopening abandoned forest roads altars the runoff and soil loss processes in a watershed. In this study, a forest watershed was chosen and abandoned roads were manually identified, aided by a high-resolution LiDAR topographic image and historic photographic images. GIS technologies were the used to describe topographic details and estimate runoff and erosion from a 10-m cell size non-road digital elevation model (DEM) for hillslopes. A 1-m cell size DEM was used to evaluate current effects of the road network, and a modified 1-m DEM to evaluate reopening closed roads with either an insloping or an outsloping design on surface hydrology and the stream network. The GeoWEPP interface to the Water Erosion Prediction Project (WEPP) Model was employed to predict road-related runoff and soil loss from hillslopes and upland channels using a 2m DEM. The WEPP:Road online interface was used to estimate runoff and erosion from current and reopened road segments. The results showed that watershed channel length may be reduced after road removal but increased if abandoned roads were reopened. Road reopening might eliminate some channels and generate new ones by intercepting upstream runoff and changing flow accumulation. Reopening roads by insloping will likely affect the channel network to a greater extent than by outsloping. Modeling results of typical sub-watersheds revealed that the average annual sediment delivery would be increased by 15.5% if roads are reopened using an inslope road profile, and reduced by nearly 20% if all roads are removal. The reduced soil losses from road removal are mainly from reduced channel erosion rather than from hillslope erosion. The sediment delivery ratio increased without the road’s effect in blocking overland sediment transport. WEPP-Road modeling results revealed that the average soil loss rates of reopened roads segments could be 7 times higher than the background rate. Soil loss risk was especially high for reopened inslope road segments close to streams or with steep gradients. Reopened roads that pass through silt loam soils are likely to generate more sediment than roads in sandy loam soils. Regression functions were developed to describe the spatial variation of soil loss risk due to road reopening. Critical distance between roads and channels can be determined to help make decisions in abandoned road reopening and watershed management.