Relation between shear stress, surface-layer armoring, and sediment transport in a mountain stream

This paper examines the relationship between surface and subsurface sediment grain size and shear stress in three sites in a subalpine stream of Colorado. The spatial variability of sediment transport at the reach scale is controlled both by the available shear stress, τ and the grain size distribution of the channel bed. Shear stress is highly variable at the reach scale; at flows near bankfull it can vary from ~2 times the reach average shear stress to less than 0.5% of this average value. It is expected that as discharge increases more grain sizes and regions of the bed become mobile. Understanding the interactions between τ and sediment size (both of the surface and the subsurface) has important implications not only for river morphology and channel stability, but also for benthic biota. Shear stress was calculated using the Multi- Dimensional Surface Water Modeling System developed by the US Geological Survey which calculates shear stress, velocity, and Shield stress, after solving depth-average versions of the Navier-Stokes equations for turbulent flow. The input data (detailed channel geometry, water surface elevation, and grain size distribution) were obtained from field surveys conducted during the summers of 2004 and 2005 in three relatively undisturbed alluvial reaches of the Williams Fork River, Colorado. The channel width at the 3 reaches varies from 10 to 22m, slopes vary from 0.004 to 0.015, and the median grain size of the bed surface (D50) varies from 40 to 85mm, whereas of the subsurface (D50s) varies between 10 and 23mm. Spatial distributions of τ were modeled in each of the study reaches for a series of flows ranging from 35 to 90% of bankfull flow. The relationship between τ and the degree of armoring (~D50s / D50) will be explored and contrasted with a tracer study conducted in each of the three sites. Maps of the channel bed indicating sediment transport intensity at every flow will be constructed. In addition, we will contrast the local shear stress obtained from the flow model with the reach average shear stress.