Incelli, Giorgio
(2016)
A fully-coupled coastal hydro-morphodynamical numerical solver.
PhD thesis, University of Nottingham.
Abstract
This research work aims at using a fully-coupled hydro-morphodynamical numerical solver to study the beachface evolution at the storm time-scale.
The proposed model originates from that of Briganti et al. (2012a), who considered a system comprising the Nonlinear Shallow Water Equations and the Exner one (bed-load only). Suspended load, bed diffusion and infiltration are now included, following Zhu (2012) and Dodd et al. (2008) approaches.
The original version of the numerical scheme (TVD-MCC) is modified to deal with the aforementioned additional physics, while the infiltration computation is implemented at the end of each time step (see Dodd et al., 2008). A new treatment for the wet / dry front is adopted, following the previous work of Hubbard & Dodd (2002).
About model validation, enhanced results are obtained in both the fluvial dune and the dam break tests with respect to those of Briganti et al. (2012a). In the uniform bore test with bed-load the results confirm those of the previous version (see Zhu et al., 2012), while in the case with combined load they show an overall good agreement with the reference solution, even though the maximum run-up is underestimated. Single swash on fixed slope experiments are reproduced as well. In the impermeable case the results are improved on those of Briganti et al. (2011), while in the permeable one the overall performance is thought to be reasonable (better the uprush than the backwash).
Although the maximum predicted inundations are smaller than measured, hydrodynamic results compare quite well with field data for real single swash events, thus confirming that one-dimensional, depth-averaged description of the swash is reasonable. The final computed bed changes show the correct order of magnitude but are generally underestimated and the predicted pattern is not always observed in the data. The sensitivity analyses indicate that this discrepancy is probably due to the initial (unknown) distributions of pre-suspended sediment concentration and velocity.
The morphodynamic evolution of two beaches at the storm time-scale is studied. In the bed-load test, results compare very well with the reference ones from Dodd et al. (2008) and Sriariyawat (2009) and, in general, the sensitivity analyses for the permeable beach case confirm previous findings. In the combined load test, the Meyer-Peter and Müller formula is applied excluding the threshold for sediment movement. This assumption is not expected to have a significant impact on the morphodynamic evolution, in the limits of the chosen parameters and settings. Increased efficiency in the entrainment rate for suspended load is found to promote onshore transport, extending Pritchard & Hogg (2005) observation for single swash events to the case of multiple ones. Variations in the incoming wave period and height yield different final bed change profiles from the default one (three long-shore bars and generally deposition seaward and erosion landward), showing differences in the number of formed bars and in the morphodynamic pattern, with sometimes accretion in the upper beach.
Beside this, new seaward boundary conditions (REBCs) are derived. They do not alter flow and bed level patterns generated by nonlinear standing waves on mobile bed, do converge to the hydrodynamic conditions on virtually-fixed bed and perform reasonably well in the demanding morphodynamic bore test.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
