The dynamics of an experimental gravel bed meander with constant discharge and sediment supply
Abstract
As rivers meander, channel migration and cutoffs introduce continuous and episodic changes, respectively, in local boundary shear stress and bedload flux. These changes must affect the local and reach scale channel dynamics, but assessing their influence is limited by complications associated with varying discharge as well as challenging spatial and time scales. Here we explore the dynamics of a scaled-down gravel bed meandering river with constant discharge and sediment supply in a 6.1 m by 17 m long experimental flume at UC Berkeley's Richmond Field Station. The experiments are similar to Braudrick et al. (2009), but with constant rather than varying sediment supply. The flume was filled with a sorted sand with D50 of 0.85 mm, and had an initial 40 cm wide channel with a sinuosity of 1.1. Alfalfa sprouts provided bank and floodplain strength. The alfalfa was seeded by hand throughout the floodplain while a low flow provided irrigation during the 7-day alfalfa growth period. Sand (model gravel) and a lightweight plastic sediment (model sand) were fed independently from the upstream end of the flume at constant rates of 1.8 and 5 kg/hr, respectively. Despite the steady input conditions the experimental channel was quite dynamic as channel migration and bend morphology varied spatially and temporally. The sinuosity in the downstream 10 m of the flume (away from the inlet condition) increased from 1.1 to about 1.6 over the first 75 hours of the experiment, when 3 cutoffs in 29 hours decreased the sinuosity back to just over the initial value. Bank erosion was fastest when curvature was low at the beginning of the experiment and following cutoffs, and slowed once sinuosity increased. Once curvature increased the bends became asymmetric as bank erosion occurred almost exclusively at the bend apex. As the channel migrated, the local sinuosity increased, which decreasing the water surface slope and hence shear stress. The lower shear stress caused subsequent channel migration and also sediment transport to decrease. Consequently, the channel aggraded, forcing water onto the floodplain and further reducing the shear stress in the channel. While the channel was aggrading, most of the sediment flux out the bottom of the flume was the suspended model sand. Cutoffs occurred when the overbank flow was sufficient to alter floodplain strength either by eroding a path around alfalfa, or by limiting alfalfa growth in floodplain areas inundated during the low flow used to irrigate the alfalfa between the runs. Comparing the duration of these experiments to time in the field strongly depends on whether the timescale of interest is set by the flow or by sediment transport. Assuming a scaling factor between 0.01 and 0.02 and that flood flows occur approximately 8 days/year, this 120 hour experiments represent 4-6 years of field time using Froude similarity to scale time from the laboratory to the field, or 220-622 years assuming sediment transport similarity. This experiment showed decreased shear stress due to channel migration limited sediment transport, and that cutoffs were a function of both in-channel and floodplain processes.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFMEP51D..06B
- Keywords:
-
- 1820 HYDROLOGY / Floodplain dynamics;
- 1825 HYDROLOGY / Geomorphology: fluvial