A high-resolution numerical model investigation into the response of a channelized salt marsh to a storm surge event

Date
2014
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
In this work, the hydrodynamics of a salt marsh located along the coast of Delaware Bay are investigated through both a field study and numerical model. In order to obtain data for validation of the model, a field study was conducted on the main channel of Brockonbridge Marsh, which opens to Delaware Bay, in Kent County, Delaware. The study encompassed a full neap-spring tidal cycle, where pressure was measured at six locations in the main channel, while velocity profiles were recorded at the channel mouth. The velocity profiles are used along with a cross-sectional profile of the mouth to compute an estimate of total flux through the inlet. Analysis of the tidal signals shows a fast rising tide, accompanied by a more slowly falling ebb tide, which is indicative of an estuary dominated by tidal variations in channel depth, rather than variations in estuary width. This suggests the wave crest diffuses landward faster than the trough. During deployment, a low-pressure system with strong winds directed onshore initiated a multi-day surge event, causing the duration asymmetry to become even more pronounced. In addition to pressure and velocity measurements, two bathymetric surveys of the main channel were conducted to obtain a detailed map of the main channel. The data from these surveys are combined with LiDAR data to create a high-resolution, topo-bathy Digital Elevation Model (DEM) of the marsh. Two important characteristics of Brockonbridge Marsh are observed from the DEM, which separate it from the typical accepted definition for the geometry of a salt marsh. First, the marsh platform experiences a decrease in elevation with distance increasing from the bay. Second, the channel depth does not monotonically decrease further inland. Instead, it is shallow near the mouth, then suddenly deepens, followed by a general decrease in depth moving inland. The maximum attainable water level at any point in the main channel does not appear to be influenced by the surface elevation at the inlet. Rather, the maximum is governed by the local geometry of the marsh in the immediate vicinity of the channel, which is a direct consequence of the negatively sloping marsh platform. The high-resolution DEM and field data are used with the Nearshore Community Model (NearCoM), a quasi-3D numerical model, to perform a validation. A small domain which does not encompass the entirety of Brockonbridge Marsh is first used. The small domain represents water levels and total flux through the mouth very well for the neap phase of the data, when the tidal flats are seldom inundated. However, with the arrival of the storm, the small domain fails to adequately account for the sudden surge of water onto the marsh platform. When the domain is expanded to encompass nearly all of the floodable surface of Brockonbridge Marsh, the total discharge computed by the model under-predicts the field observations during the storm. Furthermore, the expanded domain exhibits difficulty in draining the marsh. Computed water levels in more landward areas of the channel are higher in the model when compared to field data. Finally, it is shown how different marsh geometries can have profoundly different effects on tidal distortion. Ideal channelized salt marshes are classified into four categories based on their tidal flat configuration. Each type is modeled with NearCoM using different friction factors. For short estuaries, it is found that the presence of tidal flats serves as one of the major mechanisms in distorting the tide, leading to ebb/flood dominance. Furthermore, maximum attainable tidal amplitude is shown to be highly dependent on the direction in which the marsh platform increases in elevation. When compared to tidal propagation through a marsh with positively sloping tidal flats in the landward direction, water levels in the idealized marsh with negatively sloping flats exhibit trends much more similar to the field observations.
Description
Keywords
Citation