Uncertainty in forest road hydrologic modeling and catchment scale assessment of forest road sediment yield.

摘要

The goal of this study was to advance methods for assessment of forest road hydrologic response and sediment yield at a catchment scale. This research looked at the effect of soil depth estimation on the Distributive Hydrology Soil Vegetation Model (DHSVM), assessed the uncertainty and accuracy of hydrologic modeling of forest roads by DHSVM, and evaluated the use of road runoff and sediment sampling for catchment scale road sediment estimates. The influence of soil depth estimation on DHSVM varied by spatial scale and hydrologic process modeled. Soil depth measurement improved DHSVM simulated streamflow and road ditchflow for the rising limb of the hydrograph with no improvements during baseflow. For site specific or small scale modeling a deterministic soil depth model fit to field measurements was best. For larger scale simulations of streamflow mean soil depth provided as good or better estimates.;Considerable uncertainty in estimates of road hydrologic response was observed from DHSVM. DHSVM over predicted individual road discharges. As the spatial scale and temporal scale was increased the uncertainty in DHSVM results decreased. This suggests that model structures chosen for DHSVM would be better determined with internal catchment data, at smaller scales. The GLUE assessment showed that change detection analysis with DHSVM will be limited to sites or scales of the catchment that behavioral model structures can be identified. From this research it was determined that only the catchment scale simulations and a few individual road locations could be used for change detection.;The storm runoff volumes and peak flows from road ditchflow had linear relationships with storm sediment load. These relationships had to be developed by classes of road locations and types in an intensively managed forest due to variability in road design, hydrologic response, and road use. Sediment from roads estimated from field measurements used with SEDMODL2 or WARSEM provided substantially lower estimates than without field measured adjustments. The use of road runoff for sediment estimation provided even lower catchment scale sediment results. DHSVM simulated road runoff for sediment estimation provided catchment scale results similar to the sediment yield estimated from observed road runoff.