Postprint version. Published in Canadian Journal of Forest Research, Volume 41, Issue 10, October 1, 2011, pages 1970-1980.
Copyright © 2010 NRC Research Press.
NOTE: At the time of publication, the author Christopher G. Surfleet was not yet affiliated with Cal Poly.
The definitive version is available at http://dx.doi.org/10.1139/x11-104.
In this study, we demonstrate that watershed-scale estimates of road sediment production are improved if field measurements of road runoff and sediment production are used in the analysis. We used several techniques to spatially extrapolate measurements of road runoff and sampled sediment: comprehensive road runoff measurements, runoff estimates derived from the Distributed Hydrology Soil Vegetation Model (DHSVM), and adjustment of the road erosion models WARSEM and SEDMODL2.The sediment yield for the Oak Creek, Oregon, road network based on measured road runoff and sediment was 6.5 tons/year. When DHSVM was used to simulate road runoff, the estimated sediment from roads was similar, 6.9 tons/years. The road sediment production estimated by SEDMODL2 and WARSEM, adjusted with field-measured road runoff and sediment, was 28% and 34% less, respectively, than using the models with the default parameters. When applied to a road network in commercial forest land with frequent road use, the sediment yield estimated by SEDMODL2 and WARSEM without adjustment from field measurements was 480% and 610% higher, respectively, than with adjustments. We found that measuring runoff and sediment from one large storm event (≥1 year recurrence) provided a statistically significant relationship with the annual sediment yield.