H53D-01 13:40h
Long-term At-a-Station Channel Morphological and Hydraulic Changes in Response to Decreased Sediment Supplies
Information about past stream-channel conditions is often sought but difficult to obtain. This study analyzes at-a-station hydraulic geometry (AHG) using historic stream-flow gauging data to infer channel changes over the past century. Deviations in AHG relationships are examined in the American and Bear watersheds, Sierra Nevada tributaries to the lower Sacramento Basin, California. Many channels in this region are recovering from episodic hydraulic mining sedimentation in the late 1800s. Sediment reductions were often accelerated by construction of dams upstream, although some basins received large volumes of sediment from licensed hydraulic mining in the early 20th century. Records of width, depth, W/D, velocity, stage, cross-section area, and discharge were drawn from archival U.S.G.S. stream-flow measurement records made mostly at cableways. Ordinary least squares regression removed effects of discharge variations, and regression residuals were examined for long-term trends. Most channels experienced an early period of progressive incision accompanied by lowered stages, increased depths, and decreased top widths for a given discharge. Rapid incision began at most sites before gauge initiation and stabilized during the second half of the last century. This widespread degradation represents a tremendous flux of historical alluvium from within-channel storage, especially in the early 20th century. In some cases channel changes can be identified as responses to specific watershed events such as dam construction or 20th century hydraulic mining operations. Hydraulic variables such as Froude number, specific energy, and total hydraulic head computed from the stream-flow data can reveal systematic changes in flow environments. Specific energy at some sites increased progressively with flow depths even though total hydraulic head did not due to reduced elevation head. Evaluations of these variables allow inferences about three-dimensional changes in a channel reach. For example, Froude numbers at some sites fluctuated at a high frequency in response to variations in velocity and despite long-term depth increases. The scour and fill of a large pool can be inferred from this interaction along with changes in bed material availability following dam closure. Stream-flow records are rich in information about long-term changes to channel morphology, sediment storage, and flow environments from which much can be learned about former channel conditions to enhance historical reconstructions.
H53D-02 13:55h
Coupling sediment transport and channel morphology: must we?
There is broad agreement that stream channels adjust to the water and sediment supplied to them, indicating that sediment transport and channel morphology are coupled. There is contrary evidence. The hydraulic geometry provides broad empirical description of channel morphology using water discharge, but not sediment transport. Most stream channel designs make little, if any, reference to sediment transport and, once constructed, many appear to maintain their size and shape. The disruption of water and sediment supply by reservoirs produces downstream channel change, although the effect is often attributed to flow control and vegetation establishment, rather than a change in sediment supply. Is sediment transport relevant to channel morphology? There is some evidence. Stream channels subjected to high rates of sediment supply may not fit the broader trends of the hydraulic geometry. Some stream channel designs fail through excessive deposition. Some rivers below dams undergo considerable downcutting and armoring, indicating that they had once been adjusted to the sediment supply rate now interrupted by the reservoir. Evidently there is a threshold in the rate of sediment supply that determines whether sediment transport matters or not. Below the threshold, streams are shaped hydraulically, with a form determined by water discharge and channel materials, but not sediment transport rates. Above the threshold, channel morphology expresses an adjustment between water discharge and sediment transport rates. Given the range of streams whose morphology is described by the hydraulic geometry, and the limited adjustment observed in many built channels with a sediment-free design, the magnitude of the sediment supply threshold may be larger than typically thought. As models of channel morphology and sediment routing are refined and applied at increasing scales, and as countless miles of stream channel are rebuilt, a clearer definition of the sediment supply threshold will be useful.
H53D-03 14:10h
Bed Mobility and Channel Morphology Complexity in Urban and Rural Stream Channels
Stream morphology, which includes channel width, depth, grain size,grain arrangement, bank morphology, etc. varies over short distances along a stream. Many natural channels are quite complex and exhibit significant variability in channel characteristics. An effect of increased runoff in urbanized areas is to enhance stream bank and bed erosion, which may decrease channel morphological variability. The purpose of this study is to determine whether urban stream channel morphology is significantly less complex than non-urbanized streams in the same region. Stream reaches were chosen in 20 urbanized and 40 non-urbanized watersheds in Maryland, Wisconsin, and Washington. Watershed areas ranged from 2 to 30 km2. Channel morphological characteristics were measured at intervals spaced 1 channel width along the channel over reaches that 14-20 channel widths in length. Bankfull channel and grain size characteristics were used to determine local bed mobility at each site within each reach. The non-urbanized watershed data were used to determine: a) variations in channel morphology complexity with stream size (watershed area), b) variations in channel morphology complexity with bed mobility, and c) how to characterize channel morphological complexity over a reach. Channel morphological complexity showed an inverse correlation with bed mobility in both urban and non-urban channels. Channel morphology complexity did decrease with the amount of urbanization in the watershed if the critical dimensionless shear stress was significantly exceeded at bankfull stage.
H53D-04 14:25h
Experimental Study of Sinuous Channel Evolution Associated with Depositional Turbidity Currents
We present results from a laboratory experiment documenting the evolution of a sinuous channel form via sedimentation from a series of 24 turbidity currents. An original channel was built using a sine-generated curve to describe its planform. Wavelength and amplitude of this curve were 2.27 m and 0.39 m, respectively, producing a channel with 3 bends and a sinuosity of 1.23. Initial thickness for each current was 0.10 m, the depth of the original channel form, and the mean streamwise velocity was 0.07 m/s. Sediment suspended in each current consisted of silica with particle diameters between 5-120 Ym and a median grain size of 30 Ym. Sedimentation patterns differed from those common to terrestrial channels. In particular, sedimentation in the channel was greatest on the outer banks of bends. This lead to a reduction in sinuosity as the channel aggraded vertically, a pattern commonly observed in acoustical surveys of submarine channels on continental slopes worldwide. The primary reason for this style of channel evolution was the small excess density of the laboratory currents, 3.8%, relative to the ambient fluid, water. This small density difference lead to a fraction of the confined flow running up and out the channel at the outer bank of each bend and to an exaggerated super-elevation for the remaining channelized current. Sedimentation within the channel was 3-4 times greater along the outer bank than the inner bank of bends, altering cross-sectional geometry with each flow. Deposition on the channel bottom reduced channel relief and increased the fraction of each current exiting the channel at the outer banks of bends. At all bends the coarsest deposit was found high on the outer wall of the channel at the levee crest. Grain size decreased systematically moving down from this position into the channel thalweg and then onto the inner channel bank. Deposition rates on the outer levee crest were 30-70% of deposition rates on the channel bottom at the first bend. The ability of turbidity currents to partially exit the channel at bends resulted in focused levee growth, reducing the rate at which channel relief decreases through time. This feedback provides an explanation for the observed persistence of strongly aggrading sinuous channels in the submarine environment.
H53D-05 14:40h
Experimental investigation on the formation of alluvial cut terraces
In order to better understand the formation of alluvial terraces, experiments were conducted on a micro scale experiment. 22 braid plains (with varying control parameters such as input water, bedload and initial slope), were reproduced and their evolution in time monitored. In each case boundary conditions remained constant during each experiment and were set so the system is initially in a disequilibrium state that favors erosion of the braid plain. These experiments reveal that, as the system relaxes towards equilibrium, incision and terrace abandonment occurs in a complex manner. In several cases the rivers homogeneously and continuously cuts into its bed leaving only one paired terrace. In other cases the systems apparently destabilizes and successions of asynchronous phases of widening and incisions are observed. Series of terraces both paired and unpaired are abandoned. Analysis of mass transport and of the topographic evolution of the network enables to define two relevant parameters: the space occupation ratio corresponds to the hypsometric integral of the braid plain, whereas the effective stream power index defines the ratio of excess stream power to bedload input. A threshold of effective stream power is then shown to exist below which continuous incision occurs and above which autogenic cycles of terrace formation develop.
H53D-06 14:55h
Laboratory Modeling of Self-Formed Leveed Channels From Sediment-Laden Flows Entering Still Water
Self-formed leveed channels constructed by deposition of suspended sediment from sediment-laden flows entering still water are common features in nature. Such channels drive delta progradation, develop at tidal inlets and occur where mainstem river flows empty into oxbows and blocked valley lakes. Presently there is no theory for the formation of such channels. This lack of theory is partly due to a lack of field or laboratory studies that provide insight about the mechanism controlling these self-formed, propagating channels. The creation of such features in the laboratory, have proved illusive to date. Our ongoing experiments aimed at modeling the formation of floodplain tie channels provide insight into the necessary conditions for levee formation and channel growth. Under conditions of steady water discharge, constant sediment feed rate, unimodal sediment distribution and invariant basin stage we are able to create subaqueous lateral bars (submerged levees) along the margins of a sediment laden jet. Our results highlight the sensitivity of channel formation to issues of scaling and experimental design. In the laboratory, levee formation has only been possible with the use of plastic particles (specific gravity ~1.5); complete bed alluviation and dune formation results from the use of particles with specific gravities of ~ 2.65 across a range grain diameters and shapes. We hypothesize this effect is related to high entrainment thresholds relative to suspension thresholds of small (< 100 mm) natural particles under conditions of reduced turbulence in laboratory scaled flows. Additionally, both the width to depth ratio and the form of the outlet channel introducing the sediment laden flow into the experimental basin exert a strong control on sedimentation pattern and levee growth. Continuing experiments are focused on generating emergent channel levees and a basin ward propagation of the channel by adjusting the form of the feed channel, varying basin stage, and the use of unsteady discharge.
H53D-07 15:10h
Experimental Investigation of Gravity-Driven Particle Flows in a Turbulent Stream
Bed load transport is a longstanding problem in two-phase flows. Notably, the physical processes ruling coarse-particle/fluid systems are poorly known, despite their implications in a number of industrial and geophysical flows. We present an experimental study of the motion of coarse spherical glass beads entrained by a shallow turbulent water flow down a steep channel with a mobile bed. Typically, the particle diameter is 6 mm, the Froude number is slightly over unity, the channel inclination is 7.5 to 15%, the ratio of water flow depth to particle diameter is 1 to 4, and the observing time is one minute. The particle flow is fairly two-dimensional, the channel width being slightly larger than the particle diameter. The water flow rate and the solid discharge are kept constant at the upstream entrance. These are adjusted to obtain bed load equilibrium, that is, neither bed erosion nor deposition over sufficiently long time intervals. Flows are filmed from the side by a high-speed camera. Using an image processing software makes it possible to determine the flow characteristics such as particle positions and velocities, trajectories, change of the state of motion (rest, rolling or saltating motion) and flow depth. 1) The first striking result of our experiments is that, over short time periods, bed load transport appears as a very intermittent process: for our experimental conditions, the solid-discharge fluctuations are as large as the mean value imposed at the upstream entrance. To some degree these large fluctuations result from the finite size of our observation window. However we identified as well intrinsic fluctuations which are due (i) to the exchanges of particles between the moving solid phase and the stationary bed and (ii) to the collective entrainment of particles (see [1]). 2) Experimental runs with different solid discharges, water discharges and channel slopes allow us to study the influence of those control parameters on the bed load transport phenomena. One main result concerns the quantification of the contributions of the rolling and the saltating particles to the solid discharge: In our experiments the contribution of the rolling particles rises considerably with the channel slope. Reference: [1] T. B{\"o}hm, C. Ancey, P. Frey, J.L. Reboud, and C. Ducottet. Fluctuations of the solid discharge of gravity-driven particle flows in a turbulent stream. Physical Review E, 69:061307, 2004.
http://scitation.aip.org/getabs/servlet/GetabsServlet&prog=normal&id=PLEEE8000069000006061307000001&
H53D-08 15:25h
Sand in the cobbles: Laboratory measurements of fine-sediment transport over a coarse and immobile bed
The transported load in most fluvial systems includes a significant component of fine-grained sediment. Even in gravel- and cobble-bedded rivers, much of the sediment load may consist of sand and finer material that is transported on the bed and in suspension. Thus, the morphology of rivers with coarse, generally immobile beds is largely determined by deposition of fine sediment on the bed and banks. The theoretical and empirical framework for predicting the transport and routing of fine sediment through such rivers is incomplete, making predictions of morphologic change difficult. Existing models for suspended sediment transport focus primarily on sand-covered beds. For a sand bed among coarse immobile grains, adjustments for the effects of fractional sand coverage on the bed and drag exerted by large bed grains are available, but have not been specifically tested. We conducted two sets of laboratory flume experiments with fine-grained suspended transport over large immobile bed grains. The experiments were scaled such that (1) immobile bed particles were much larger than the sediment in transport, but less than 10% of flow depth, (2) transported sediment was in the same size-range as occurs in field settings, and (3) bed shear stresses scaled by the grain size of the transported sediment were also similar to those that occur in the field. The first set of experiments was conducted under conditions of equilibrium transport to evaluate the effect of the near-bed boundary condition on the suspended sediment transport field. In the second set of experiments, non-uniform transport conditions were imposed to examine the migration of sand pulses through a coarse-bedded channel. We measured near-bed sediment concentrations and monitored interstitial sand storage among hemispherical roughness elements for a range of flow and transport rates. Runs with less sand coverage on the bed had higher near-bed sand concentrations compared to runs with greater sand coverage. As sand bed elevation drops among the roughness elements, turbulent wakes shed by the large grains appear to enhance grain entrainment more than the corresponding decrease in bed area covered by sand decreases sand entrainment. These higher concentrations are maintained until the bed is depleted of fine sediment. We also observed that partial filling of interstitial spaces occurred over a narrow range of flow and transport rates, suggesting that a sharp threshold may exist between a bed with no interstitial sand storage and a bed that is completely covered by sand. We are testing these transport thresholds with a morphodynamic model applied to the sand migration observed in the nonuniform transport experiments.