Forests and Water: A State of the Art Review for Colorado

Home Browse Search

ation activities such as off - highway vehicles, mountain bikes, or trail use by foot or animal can compact the soil surface and increase soil erosion rates (Leung and Marion, 1996). Erosion rates usually decrease after con- struction as the disturbed sites develop an armored sur- face or are revegetated. Erosion rates can decline by 90% or more as roads age (Ketcheson and Megahan, 1996). Road surfaces, cutslopes, and inside ditches can continue to produce large amounts of sediment as long as traffic or road maintenance operations prevent revegetation or surface stabilization. Management practices such as the application of gravel can reduce road surface erosion rates by 80 percent or more (Burroughs and King, 1989). In the absence of any treatment, abandoned roads and trails can be slow to revegetate and continue to produce substantial amounts of sediment. Any assessment of road erosion rates must also con- sider the effects of roads and trails on runoff, as this will greatly affect the rate of erosion and the subsequent delivery of sediment to the stream channel. Roads, skid trails, and other travelways have lower infiltration rates and generate more runoff than adjacent undisturbed wa- tershed areas. The increased runoff and the construction of cut and fill slopes can generate mass movements, and in some environments this is a greater concern than road - induced surface erosion. Road erosion rates have been measured in a variety of environments (Elliot, 2000). Table 3.1 lists published erosion rates from the road tread, cutslope, and fill slopes in the original units and in kilograms per square meter per year in order to facilitate comparisons. This table shows that road tread erosion rates can exceed 100 tons ac' yr' (22 kg M-2 yr'), while more "typical' values are in the range of 1 -10 tons ac' yr' (0.2 -2 kg m-2 yr') (Table 3.1). The point is that there is a wide range of values, and some of the data from other countries, such as Australia, may be more similar to some parts of Colorado than data from areas closer to Colorado, such as Oregon or Washington. The other key issue is how much of the sediment gener- ated from roads gets delivered to wetlands, lakes, or the stream network. Relatively few studies have measured the delivery of road - derived sediment to the drainage network. Roads located close to streams may have a greater impact on water quality as the shorter travel dis- tance increases the likelihood that sediment -laden waters will reach streams or other water bodies. Several stud- ies have shown that road ditches and concentrated road drainage increase drainage density (defined as the total length of streams per unit area) (Wemple et al., 1992; Montgomery, 1995). This increase in drainage density is presumed to increase the size of peak flows and the proportion of road - derived sediment that is delivered to the stream network (Wemple et al., 1992; Montgomery, 1995). As in the case of forest management, vegetative filter strips can reduce the delivery of road runoff and erosion to surface waters (Campbell, 1984). Sediment yields from weir pond accumulations in the Fraser Experimental Forest indicate that roads and tim- ber harvest increased sediment yields by only 0.014 tons ac' yr' (Stottlemeyer, 1987). The increase in sediment production was attributed primarily to the increase in flow and corresponding increase in sediment transport capacity than erosion from the roads and harvest units (Troendle and King, 1987; Troendle and Olsen, 1994). Roads and timber harvest on the Fool Creek experimen- tal watershed increased average sediment yields by only 0.005 tons ac' yr' (Leaf, 1974). Approximately 10 per- cent of the increase in sediment yield was attributed to flow changes and 90 percent of the sediment was related to roads. The observed increase in sediment yields de- clined sharply in the first 10 years after harvest. 24 An increase in erosion rates and sediment yields was difficult to detect on the larger -scale Coon Creek study. This was due to the limited amount of pre- treatment data and the absence of any sediment measurements except instream transport rates adjacent to the downstream gauging stations (Troendle et al., 2001). The data did not show any significant difference in suspended sedi- ment concentrations between the treated and the control watersheds, or in the treated catchment before and after the management activities. Bedload transport rates in the treated watershed were significantly greater than for the control watershed, but an assessment of change relative to the pre- treatment condition was precluded by the ab- sence of pre- treatment data. The greater bedload trans- port rates in the treated watershed are consistent with the observed increase in discharge, implying that the increase in sediment production from roads and harvest activities units were not as significant at the watershed scale as the increase in transport capacity (Troendle et al., 2001). As noted earlier, road maintenance and high traffic vol- umes can increase erosion rates from unpaved roads. Road maintenance can include cleaning ditches and trimming the cut bank as well as regrading the road sur- face. Maintenance can generate more erosion than road usage (Luce and Black, 1999). Older roads often are of greatest concern because they tend to concentrate rather than disperse runoff, and the runoff is often directed into