A quasi-3D storm surge modeling system was developed for forecasting the storm surge and inundation in the Chesapeake Bay. The system was constructed with one large unstructured grid covering the Atlantic Coast from Nova Scotia to Florida and a smaller, limited domain unstructured grid covering the Chesapeake Bay, Virginia Beach, Hampton Roads and the adjacent continental shelf regions. It was demonstrated that, with the large domain grid, the model could simulate the hurricane induced storm surge reasonably well using astronomical tide at the open boundary condition and in turn, provide boundary condition for the limited domain model. Since the difficulty of specifying the open boundary condition in the limited domain was solved using this dual-grid methodology, the model achieved the predictive capability for forecasting storm surge as long as accurate atmospheric wind is provided. The high resolution, limited domain grids could, at the same time, be used to map inundation features in details inside the Chesapeake Bay..
The storm surge modeling system has been linked with different atmospheric models. The simulations with different winds proved that the storm surge simulation is highly dependent on the quality of the winds generated by atmospheric models and ensemble technology is necessary for the forecasting purpose and the uncertainty associate with it. Procedures of ensemble simulations were developed in this study so that the modeling system could be used for forecasting storm surge with ensemble winds. Realistic simulations conducted in this study demonstrated that the high-resolution unstructured grid, which is able to resolve complex bathymetry, topography, and coastline structure, could generate accurate storm surge and inundation results if LiDAR data are incorporated. The high-resolution grid and accurate LiDAR topographic data are essential for generating accurate inundation maps. One of the significant findings in this study is that the coastal Ekman dynamics does play an important role in water exchanges between continetial shelf and the Chesapeake Bay. In this context, it is, thereofer, necessary to use a quasi-3D model, rather than 2D model, in order to resolve vertical varying Ekman transport, which significantly improve the storm surge prediction.
| Identifer | oai:union.ndltd.org:wm.edu/oai:scholarworks.wm.edu:etd-3013 |
| Date | 01 January 2009 |
| Creators | Shen, Tao |
| Publisher | W&M ScholarWorks |
| Source Sets | William and Mary |
| Language | English |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Dissertations, Theses, and Masters Projects |
| Rights | © The Author |
Page generated in 0.0021 seconds