Yavuz Ozeren

Analysis of Vegetation Effect on Waves Using a Vertical 2D RANS Model

ABSTRACT Wu, W.; Zhang, M.; Ozeren, Y., and Wren, D., 2013. Analysis of vegetation effect on waves using a vertical 2D RANS model. A vertical two-dimensional (2D) model has been applied in the simulation of wave propagation through vegetated waterbodies. The model is based on an existing model, SOLA-VOF, that solves the Reynolds-averaged Navier–Stokes (RANS) equations with the finite difference method on a staggered rectangular grid and uses the volume of fluid method to capture the free surface. The model is enhanced in this study by adding the drag and inertia forces in the momentum equations to account for the vegetation effects, implementing the subgrid-scale model for turbulence closure, and incorporating wave-maker, sponge layer, and bottom friction in boundary conditions. The model was first validated using measurement data collected from the literature and then applied to simulate wave propagation in flumes covered by rigid and flexible model and live vegetation. The considered live vegetation species are Spartina alterniflora (smooth cord grass) and Juncus roemerianus (needlegrass rush), which are commonly distributed on coastlines. The model is able to reproduce wave attenuation through the vegetation zone observed in the experiments. The drag coefficients are calibrated in the vertical 2D RANS model and the analytical model based on the wave energy equation and linear wave theory, and the calibrated values in the two models are reasonably close.

Coupling a Two-Dimensional Model with a Deterministic Bank Stability Model

Stream bank erosion can be an important form of channel adjustment in unstable alluvial environments and hence should be accounted for in geomorphic studies, river restoration, dam removal, and channel maintenance projects. Recently, one-dimensional and two-dimensional simulation models have become useful tools for predicting channel responses; but most either ignore bank failure mechanisms or implement only simple ad hoc methods. In this study, a twodimensional model (SRH-2D) is coupled with a deterministic bank stability and toe erosion model (BSTEM) to predict channel adjustment and planform development. Herein, the proposed coupling approach is described, along with numerical aspects of the procedures. For test and verification purposes, the coupled model is used to predict bank retreat of Goodwin Creek in Mississippi. A comparison of the model results with the measured data is presented and discussed.

Representation of Linear Terrain Features in a 2D Flood Model with Regular Cartesian Mesh

Highly transient floods resulting from the failure of dams and levees can lead to loss of life and property damage. Two-dimensional (2D) numerical simulations that use modern shock-capturing schemes are particularly suited to simulate these mixedregime floods for flood mapping, consequence analysis and emergency management planning. The Digital Elevation Models (DEM) are often used as a regular computational mesh. Unfortunately, linear terrain features, such as road and railroad embankments and dikes, which may influence flood patterns, are not adequately captured in DEMs. This study describes a two-sided cut-cell boundary method for representing linear terrain features on a regular Cartesian mesh. The proposed method is briefly described and some test simulations are presented.

Boat-Generated Wave and Turbidity Measurements: Connecticut River

Waves generated by high-speed recreational boats can play a significant role in bank erosion and failure along riverbanks. Frequent passes of these boats in confined channels increase suspended-sediment concentration and turbidity at relatively shallow water depths, and increase the erosion rate by wave breaking and long-shore currents. In the current study, boat-generated waves and turbidity were measured in the Connecticut River along a reach between Vernon Dam, Vernon, VT and French King Bridge, Miller Falls, MA. Two-meter-long self-powered, self-logging wave staffs were used to measure the water level and waves at three sites. Based on these uninterrupted measurements, boat-generated wave-characteristics were obtained for a period of four months between May 2015 and September 2015. As part of this study, turbidity and wave measurements at these three sites were carried out during a series of in situ experiments with a 5.7 m long tri-hull motorboat. The experiments included a total of 36 controlled passes parallel to the shoreline at speeds ranging from 8 km/h up to 55 km/h. Boat speed and path was measured with a global positioning system (GPS) logger and the turbidity level near the bank was measured using two optical backscatter sensors. Relations between the wave height, wave period and turbidity level were investigated. This paper presents the results of these additional measurements and discusses relations between turbidity levels and boat-generated waves.

Generation of Boat Traffic Data: Techniques for Temporal and Spatial Extrapolation

Boat wave induced erosion can be a major concern in rivers and other navigable waterways. When they reach the banks, boat-generated waves can produce velocities and shear stresses near the water surface much larger than those generated from streamflow. This is mainly due to the high level of turbulence produced by wave breaking. Recurrent wave action along cohesive banks due to boat traffic may cause undercutting and eventual failure. Estimation of the relative contribution of boat waves to stream bank erosion requires the knowledge of the spatial and temporal distribution of instances of boat passage in addition to the boat wave properties. Despite of the substantial contribution of boat-induced waves to the bank erosion, the availability of historic boat-wave data for hydraulic modeling purposes is extremely limited. In this study, a partially deterministic method for generating long-term boat-traffic data from a field campaign is presented. Boat-generated wave data (water depth, wave height and wave period) was collected for four months between May, 2015 and September 2015 at three sites along a reach of the Connecticut River, MA. The collected data was then used to predict 15-year boat traffic and boat-wave properties at 25 sites along the same reach of the river, by combining measured daily and weekly variations of boat traffic, together with information on seasonal variations, boat ownership statistics and historic rainfall data obtained form various sources.

PIV-PTV Measurements of a Tailings Dam-Break Flow

To experimentally study transient flows such as dam-break flows, imaging techniques are often used due to the constraints imposed by the highly transient nature of such phenomena. In this paper an experimental laboratory study of tailings dams made with a combined PIV + PTV algorithm is presented. In this algorithm the PIV technique is first applied to determine an estimator for the sediment layer velocity field. The obtained estimator is then used to compute a displacement estimator for the PTV approach. In this method the first step (PIV) provides a global estimation of the sediments velocity field and in the second step (PTV) a particle-by-particle analysis is obtained leading to an increased spatial resolution of the measurements. The obtained algorithm is applied to a laboratory tailings dam-break flow to obtain relevant information about the behavior of the tailings. This paper is a summary of the research presented in other papers by the authors.