This report describes work performed to quantify the erodibility of surface soils in the Yolo Bypass (Bypass) near Sacramento, California, for use in the California Department of Water Resources (DWR) Yolo Bypass D-MCM mercury model. The Bypass, when not serving as a floodway, is heavily utilized for agriculture. During flood events, surface water flows over the soil, resulting in the application of a shear stress to the soil. The shear stress is a function of flow speed and is often assumed to vary as the square of flow speed. Once the shear stress reaches a critical value, erosion commences, and the erosion rate typically increases with applied shear stress. The goal of the work described here was to quantify this process and how it varies throughout the major land uses found in the Yolo Bypass.
Each of the major land uses found in the Bypass was targeted for sediment coring and two side-by-side cores, 10 centimeters in diameter, were extracted at each site for testing in a Gust erosion chamber. This device consists of a cylinder with a piston and cap installed to contain a sediment sample and overlying water. In most instances, coring was done with the cylinder, the piston and cap were installed, and testing commenced immediately. The cap at the top of the cylinder contains vanes to induce rotation of the flow and is driven by an electric motor, simulating the bed shear stress experienced by the soil in a flood event. Ambient water is introduced to the cylinder, passes through the device, and carries eroded sediment out of the chamber. The exiting water is tested for turbidity, and water samples obtained to relate turbidity to suspended sediment concentration are used to compute erosion rates for each of the applied shear stresses.
The result for each sediment core is (1) definition of the critical shear stress required to initiate sediment erosion and (2) estimation of coefficients required to relate erosion rate to applied shear stress once this critical shear-stress threshold has been exceeded. These quantities were computed for each of the sites sampled. In total, 10 locations were sampled, representing 10 land uses ranging from wild and white rice fields to the flooded Liberty Island and the Toe Drain that receives runoff from much of the cultivated land (table 1).
The Gust chamber test causes the erosion of a very small layer of sediment, typically less than a millimeter thick. The strength of the soil within this layer increases with depth, typically, and this soil strength versus depth is measured in the testing process.
Results for each land use type tested are presented as the initial critical shear stress at which erosion began and the rate at which erosion increases as shear stress increases (table 2). Of the land use types sampled, irrigated pasture displayed the lowest critical shear stress, meaning that it required the smallest flow speed to initiate erosion. But in this case, the rate of increase of the subsequent erosion, given higher flow speeds, was small. The wild rice field samples exhibited a higher critical shear stress but also exhibited a much higher erosion rate once the critical shear stress was exceeded. The erosion rate for wild rice was about three times greater than that for white rice. Bear in mind that these results are based on only two cores tested per site, and variability between fields with the same crop could be significant. Approved digital data can be viewed and downloaded from ScienceBase, at https://doi.org/10.5066/F7BV7DQC. These results are being used to calculate erosion rates in the DWR Yolo Bypass D-MCM mercury model.
The Toe Drain was very difficult to sample, owing to hard, consolidated sediments on the channel bed. On the first visit, two cores were obtained successfully, and testing revealed very different results. A second visit was made, but it was not possible to obtain cores suitable for testing with the coring equipment used. The available results suggest that Toe Drain soil is highly erodible (low critical shear stress and high erosion rate once initiated) despite being difficult to sample. As a collector of runoff, it also has the potential to accumulate soils eroded from adjacent areas, subsequent to storm events, as flows subside. This deposited material will typically be more erodible than the material that it lands on. The deposition and resuspension of material was not simulated in the testing described here because the applied shear stress increases monotonically during testing.
The spatial distribution of mean grain size, loss on ignition, and percent fines of Yolo Bypass soils are also presented. Sediment sampling for this effort was performed by DWR; the U.S. Geological Survey (USGS) performed the sample analysis. These data should thus be considered provisional, but the remainder of the data presented here, and this report, have been through the formal U.S. Geological Survey review process.
A separate effort has been made by others to develop numerical model results defining the spatially varying, time-dependent hydrodynamics in the Yolo Bypass. These model results are being used to quantify shear stress on the soil surface, which together with the Gust chamber results shown here, are used for the DWR Yolo Bypass D-MCM mercury transport model to compute erosion rates for each time step.