Author(s): A. Atzeni; A. Balzano; R. A. Falconer; B. Lin; Y. Wu
Linked Author(s):
Keywords: Moving boundaries; Numerical models; Tidal flows; Wetting anddrying
Abstract: Three two-dimensional shallow water flow models and four methods for simulating flooding and drying processes are compared and evaluated. The comparison has been carried out for the significant case study represented by Poole Harbour, Dorset, UK, a natural harbour whose wet plain surface is nearly halved between high and low tide. After a brief description of the hydrodynamic models, preliminary simulations are carried out, with each model being implemented with its own wetting and drying algorithm. On this base, one model is chosen to implement all four methods for flooding and drying in it for testing them. The four methods are then compared pointing out both rough and detailed differences of results. It is shown that two of the methods tested, as expected, actually yield what appears to be an effectively inaccurate evaluation of the storage volume of the water body, resulting in significant modification of the peak values of the computed results. Further simulations carried out with the other two methods yield significant improvement of the results compared to the original couplings between hydrodynamic model and flooding and drying method. In particular, it is shown that the depth tolerance value, often concerned with in several criteria for the checking of flooding and drying conditions, can be actually chosen to be one order of magnitude smaller than previously used in the DIVAST model, with use of the WWU method proposed by Balzano (1998) (10-2 m or less). This result is achieved without wiggles or other irregularities of merely numerical origin being introduced in the computed results. Also, bottom elevations are accounted for which are consistent with actual bathymetry, implying a correct evaluation of the storage volume of the cell in the context of the linear bottom elevation model assumed.
Year: 2001