FN Archimer Export Format
PT J
TI A numerical scheme for coastal morphodynamic modelling on unstructured grids
BT 
AF GUERIN, Thomas
   BERTIN, Xavier
   DODET, Guillaume
AS 1:1;2:1;3:2;
FF 1:;2:;3:;
C1 Univ La Rochelle, Inst Littoral & Environm, CNRS, LIENSs,UMR 7266, 2 Rue Olympe de Gouges, F-17000 La Rochelle, France.
   Inst Univ Europeen Mer, CNRS, LETG, GEOMER,UMR 6554, Pl Nicolas Copernic, F-29280 Plouzane, France.
C2 UNIV LA ROCHELLE, FRANCE
   UBO, FRANCE
IF 3.341
TC 18
UR https://archimer.ifremer.fr/doc/00606/71851/70480.gif
   https://archimer.ifremer.fr/doc/00606/71851/70481.gif
   https://archimer.ifremer.fr/doc/00606/71851/70482.gif
   https://archimer.ifremer.fr/doc/00606/71851/70483.gif
LA English
DT Article
CR DYNAMO
BO Haliotis
DE ;Morphodynamic modelling;Unstructured grid;WENO;Diffusion;Coastal environments;Exner equation
AB Over the last decade, modelling systems based on unstructured grids have been appearing increasingly attractive to investigate the dynamics of coastal zones. However, the resolution of the sediment continuity equation to simulate bed evolution is a complex problem which often leads to the development of numerical oscillations. To overcome this problem, addition of artificial diffusion or bathymetric filters are commonly employed methods, although these techniques can potentially over-smooth the bathymetry. This study aims to present a numerical scheme based on the Weighted Essentially Non-Oscillatory (WENO) formalism to solve the bed continuity equation on unstructured grids in a finite volume formulation. The new solution is compared against a classical method, which combines a basic node-centered finite volume method with artificial diffusion, for three idealized test cases. This comparison reveals that a higher accuracy is obtained with our new method while the addition of diffusion appears inappropriate mainly due to the arbitrary choice of the diffusion coefficient. Moreover, the increased computation time associated with the WENO-based method to solve the bed continuity equation is negligible when considering a fully-coupled simulation with tides and waves. Finally, the application of the new method to the pluri-monthly evolution of an idealized inlet subjected to tides and waves shows the development of realistic bed features (e.g. secondary flood channels, ebb-delta sandbars, or oblique sandbars at the adjacent beaches), that are smoothed or nonexistent when using additional diffusion. 
PY 2016
PD AUG
SO Ocean Modelling
SN 1463-5003
PU Elsevier Sci Ltd
VL 104
UT 000380673200004
BP 45
EP 53
DI 10.1016/j.ocemod.2016.04.009
ID 71851
ER

EF