Journal of Hydraulic Engineering

Bank erosion plays an important role in the hydro-morphodynamics and evolution of natural rivers. Therefore, it is essential to have a reliable bank erosion model for accurate simulation of hydro-morphodynamic processes. We developed and successfully implemented a new, feasible bank erosion model in Delft3D software. The developed model considers physical bank erosion processes to a greater extent than previous models. Model performance was assessed by comparison with a previously reported experiment in a mobile-bed-and-bank laboratory open-channel bend flume. The results from our developed model were compared with those from the standard Delft3D and angle of repose bank erosion models. We showed that progressive lateral bank erosion as well as the corresponding hydro-morphodynamics of the channel were better predicted by the developed bank erosion model. The results of this study provide insight into bank erosion prediction with Delft3D, and they suggest that the developed model will improve the performance of the Delft3D model for short- and long-term hydro-morphodynamic simulation of natural meandering rivers.

When estimating design local scour depths at structures where the design flows have short durations, equilibrium scour depths can be too conservative. Attempts to develop methods to accurately estimate scour rates and times to reach near equilibrium scour depths have only been moderately successful. The objective of the work reported here was to develop an improved, but still conservative, mathematical model for estimating local scour depth as a function of time for a known flow event. This model also estimates scour rates and the time required for the scour to reach near equilibrium (80%–90%) depths. When used in conjunction with the recently published equilibrium local scour depth model, Sheppard et al., it can be applied to both simple and complex bridge pier structures. When using the scour rate equations presented here, changes in sediment properties with sediment depth can be included in the analysis. Due to the lack of local scour rate data at large structures subjected to live-bed flow conditions, the proposed methodology is currently limited to clear-water scour conditions.

Turbulent flow around a circular pier on a gravel bed is investigated experimentally, using particle image velocimetry. The turbulent kinetic energy (TKE), Reynolds shear stress (RSS), and vorticity are examined to understand the effect of flow depth ($h$) to pier diameter ($d$) ratio and flow intensity. Additionally, spectrum analysis is carried out to examine the vortex intensity over various frequencies of eddies. Mixing of the horseshoe vortex and surface roller is observed in the upstream of pier for $h/d<4.33$. Consequently, two opposite rotating vortices reduced the turbulence strength. The critical limit of $h/d$ on a gravel bed is found to be three times more than that of a sand bed. Both TKE and RSS increased approximately linearly with flow intensity. Further, vortex shedding frequency and energy of vortices increased with $h/d$ which attributed to an increase in the normalized equilibrium scour depth. The present study will be resourceful for researchers in developing scour models for gravel bed by giving more insights into the flow hydrodynamics with respect to approach flow parameters.

Unsteady internal water pressure (UIWP), such as water hammer and mass oscillation (rigid column effects) induced by load regulation of hydropower plants arranged with surge tanks, leads to crack damage and structural instability in hydropower tunnels. In this study, an analytical model for the propagation of UIWP along the cracks in hydropower tunnel walls was developed by analytically solving the one-dimensional (1D) unsteady seepage flow equation in a cylindrical coordinate system. The analytical solution, which was validated to be in good agreement with prototype monitoring data, was then applied to an engineering case to investigate the propagation characteristics of UIWP along the cracks in the tunnel wall. UIWP was found to be attenuated along the propagation direction because of the flow resistance, indicating that the surface layer of the tunnel wall may be more vulnerable to UIWP. Moreover, the maximum propagation depth of mass oscillation was greater than the water hammer. Time lag occurred and resulted in an additional seepage force on the inner layers of cracks, referred to as hydraulic impact, which was expected to be one of the main drivers of instability in the tunnel structure. The hydraulic impact became more pronounced with the increase in period and amplitude. Based on the comprehensive comparison of the propagation characteristics between water hammer and mass oscillation, mass oscillation is concluded to have a more significant influence on structural damage than water hammer and, therefore, should receive more attention.