Journal of Hydraulic Engineering

A plan was made to remove Matilija Dam on the Ventura River. With dam removal, the delta in the reservoir and the downstream channel were expected to undergo major changes in morphology. The FLUVIAL-12 model was employed to simulate reservoir and river channel responses after dam removal. As a first step, the model was calibrated using the Ventura River data to establish its validity. In calibration, the model was used to simulate the fluvial processes starting from the time of dam completion. The simulated sediment deposition above the Matilija Dam matches closely with the deposits measured by the U.S. Bureau of Reclamation. A large amount of sediment was stored in the reservoir; some of the stored sediment would be transported downstream after dam removal. An important consequence of dam removal is the major increase of sediment release to the river channel downstream. The sediment supply to the downstream reach is not only from the stored sediment in the reservoir but also from natural sediment inflow from the upstream watershed. Therefore, sediment supply to the downstream reach will exceed the natural sediment flow before the dam presence. This situation tends to overload the downstream reach with sediment, resulting in excessive deposition. The amount of sediment release from the area above the removed dam is closely related to the changes in reservoir morphology. It is necessary to model changes in the channel bed profile and channel width during erosion in order to determine the amount of sediment removal. The amount of sediment release may not be simulated using an erodible-bed model but it may be determined using an erodible boundary model.

Tidal current and wave forces are the primary factors causing agitation, suspension, and transport of littoral materials in a bay. This paper presents a study of the circulation patterns of the water and sediment fluxes in a real bay during ordinary natural weather conditions. A coupled hydrodynamic and wave model is developed to support the research. A two-dimensional parallel advanced circulation hydrodynamic model with an extended transport module is used to calculate the water circulation within the bay, while the simulating wave nearshore model is implemented to provide the wave parameter data. The month of March 2003 is selected to represent the test case for the sediment transport model in Matagorda Bay, Tex. The presence of wave action significantly, increases the amount of suspended sediment, and consequently, more deposited sediments occur around the ship channel, where shallow shoals are located. Comparison of the simulated deposited sediments in the Matagorda ship channel with the two selected points of historical dredging data of the bay gives a maximum difference of 22%. This result suggests the need for an extensive comparison study.

The paper presents the application of a two-dimensional depth-averaged numerical model to simulate the lateral migration processes of a meandering reach in the West Jordan River in the state of Utah. A new bank erosion model was developed and then integrated with a depth-averaged two-dimensional hydrodynamic model. The rate of bank erosion is determined by bed degradation, lateral erosion, and bank failure. Because bank material in the West Jordan River is stratified with layers of cohesive and noncohesive materials, a specific bank erosion model was developed to consider stratified layers in the bank surface. This bank erosion model distinguishes itself from other models by relating bank erosion rate with not only flow but also sediment transport near the bank. Additionally, bank height, slope, and thickness of two layers in the bank surface were considered when calculating the rate of bank erosion. The developed model was then applied to simulate the processes of meandering migration in the study reach from 1981 to 1992. Historical real-time hydrographic data, as well as field survey data of channel geometry and bed and bank materials, were used as the input data. Simulated cross-sectional geometries after this 12-year period agreed with field measurements, and the $R^2$ value for predicting thalweg elevation and bank shift are 0.881 and 0.706, respectively.

This paper presents a comprehensive analysis of suspended sediment transport in open channels under various flow conditions through a kinetic-model-based simulation. The kinetic model, accounting for both sediment-turbulence and sediment-sediment interactions, successfully represents experimentally observed diffusion and transport characteristics of suspended sediments with different densities and sizes. Without tuning any model coefficients, the nonmonotonic concentration distribution and the noticeable lag velocity with a negative value close to the wall are reasonably reproduced. Examination of flow conditions typical of suspension dominated rivers shows that the conventional method may overestimate or underestimate unit suspended-sediment discharge, depending on the Rouse number, sediment size, as well as shear velocity. The error may be less than 20% for $d_p<0.5\mathrm{mm}$ and might exceed 60% for $d_p>1.0\mathrm{mm}$ under typical flow conditions where shear velocity ranges from $1.0\text{to}12.5\mathrm{cm}∕\mathrm{s}$ and flow depth ranges from $1.0\text{to}5.0\mathrm{m}$ .

In the saltation regime where bed-shear stress is low, bed load moves by sliding, rolling, and saltating along the bed, while in the sheet-flow regime where bed-shear stress is high, it travels by a combination of saltation and sheet flow. In this paper a theoretical model is developed for predicting the onset of the sheet-flow regime as shear stress increases. This model is based on a new variable $P_b$ representing the proportion of grains on the bed that are entrained as bed load. The model yields the equation $P_b=2.56\theta G^3$ in which $G=1-{\theta }_c∕\theta$ , $\theta =\text{dimensionless}$ bed-shear stress; and ${\theta }_c=\text{critical}$ value of $\theta$ at which grains begin to move. The equation shows that ${\theta }_t$ , which is the value of $\theta$ at the onset of the sheet-flow regime and is assumed to occur when $P_b=1$ , is around 0.5 with the exact value controlled by ${\theta }_c$ . For example, when ${\theta }_c=0.045$ , ${\theta }_t=0.52$ . The theoretical model is verified by performing a nonlinear regression analysis on data from 285 flume experiments. Additional flume experiments with a high-speed video (HSV) system result in consistent values of $\theta$ for the onset of the sheet-flow regime, which support the theoretical model. The HSV images further reveal that: (1) the sheet-flow regime is characterized by granular sheets or laminations; and (2) a zone of mixed saltation and rolling grains exists not only in the saltation regime but also in the sheet-flow regime.

Stochastic bed load transport formulas for nonuniform sediment exist, but most of them do not account for the composition of surface material to predict fractional transport rate. This study transformed a surface-based bed-load transport predictor to a stochastic one by approximating the fluctuation of bed-shear stress with a standard log-normal distribution. The deterministic predictor underpredicts fractional transport rate at low values of bed-shear stress and Reynolds number. The modified stochastic predictor predicts fractional transport rate more accurately and converges to the deterministic one at high shear stresses.