AmericanChemicalSociety.com

Sediment supply is widely acknowledged as a fundamental control on channel morphology, but the mechanisms by which sediment supply affect channel planform and dynamics are understood only qualitatively. Numerical models do not explicitly account for sediment supply effects on bed topography and planform morphology. Because changes to sediment supply are often associated with changes in both climate and land use, understanding the history and future of river basins requires inclusion of sediment supply. We hypothesize that changes in topographic steering link sediment supply and bar morphology. This hypothesis posits that increased sediment supply would enhance topographic steering over the bar, causing the bar to advance laterally into the pool. Contrarily, decreased supply would diminish topographic steering, causing bars to shrink laterally and pools to expand. We are testing this hypothesis using flume experiments and will use those results to test and modify a quasi 3-D hydrodynamic model (MD_SWMS) to include the effects of sediment supply. Prior to receiving this grant we succeeded in creating a meandering channel in a laboratory flume using alfalfa sprouts to provide strength and both coarse and fine sediment. We therefore have a unique opportunity to examine how changes to sediment supply affect self-formed single thread channels. Over the course of this project, we plan to vary the sediment feed for three conditions: straight with alternate bars, sinuous but not migrating, and a migrating sinuous channel.  Over the first year of this grant we have completed the supply depletion experiments in the straight flume, made modifications to our meandering flume, and begun the experiments in our meandering basin.

We first examined this hypothesis in a 28-m long, 0.86-m wide flume with a constant discharge of 5.4 l/s and sand with a median grain size of 0.8 mm. The conditions were sufficient to support alternate bars downstream of the upper 8-10 m. Cutting off the feed initially resulted in an increase in flux at the bottom of the flume as fines were released from upstream. This resulted in shorter bar wavelengths, increased celerity, and no change in bar height. The bars did swell during this period encroaching into the pools and causing the flow over the top of the bar to deepen. Preliminary modeling with MD_SWMS, a quasi-3-D hydraulic model indicates that the area over which topographic steering increased as the bars grew.

Once the initial pulse of fine sediment moved through the flume, the sediment flux began to gradually decrease. With the decreased sediment flux, bar migration decreased and bar wavelength increased until migration stopped. As the sediment flux decreased the bar height decreased as the pools filled in. These shallower pools then eroded the edges of the bar resulting in a wider pool and narrow bar, as our theory would predict. MD_SWMS modeling indicates that the maximum magnitude of the topographic steering was similar to the equilibrium condition, but that the area over which the topographic steering occurred decreased substantially. Overall, bar wavelength was the most responsive variable to the change in sediment supply, with shorter wavelengths associated with higher flux, but the bar height was relatively constant until the bars stopped migrating and became emergent. We will conduct experiments where we increase the feed in the straight flume with alternate bars in early 2010 when the weather is too cold to run the meandering flume.   

We are currently conducting experiments in the 6.2 X 17 m meandering flume where we have increased the bank strength relative to our previous experiments on meandering. We are currently running the flume to reach an equilibrium morphology prior to increasing the feed. We have yet to analyze the data from these ongoing experiments, but qualitatively, the bars seem wider relative to the pool width than previous experiments at lower supply and stress. We have also observed that the higher stress and bank strength creates a more sinuous channel. We expect that these experiments will be completed by this winter or early spring.

In October 2009, we will be conducting an experiment at the St. Anthony Falls Outdoor Stream Lab (OSL) at the University of Minnesota. We will be conducting this experiment in conjunction with researchers from Johns Hopkins University and Utah State University. The OSL is 3 m wide and approximately 0.5 m deep, and is sinuous but does not migrate. The OSL has the advantage of being sufficiently deep to make near-bed velocity and turbulence measurements. We will start with the sediment feed set to near-zero, and increase the sediment feed in two steps. We will measure changes to the bar morphology, water surface, and paired velocity and sediment transport. Because it does not migrate and we can take detailed stress and velocity measurements, the OSL has several advantages relative to our smaller basin at the Richmond Field Station, and the data can be more easily used to test and modify MD_SWMS.

In conjunction with researchers from Johns Hopkins University and Utah State University, we are working with the USGS to adjust their model to incorporate the effects of sediment supply. To date we have used MD_SWMS to model the hydrodynamics in the alternate bar experiment, but are waiting to make some changes to the model to include morphodynamics.

This project has supported the Ph.D. of Christian Braudrick (expected completion 2010), enabling him to expand on previous work on meandering rivers to include the specific effects of sediment supply on channel form. Two undergraduates, Cory Hiraga and Sumner Collins have had there first exposure to research as laboratory assistants.