Gangfeng Ma

Modeling of SMF tsunami hazard along the upper US East Coast: detailed impact around Ocean City, MD

With support from the US National Tsunami Hazard Mitigation Program (NTHMP), the authors have been developing tsunami inundation maps for the upper US East Coast (USEC), using high-resolution numerical modeling. These maps are envelopes of maximum elevations, velocity, or momentum flux, caused by the probable maximum tsunamis identified in the Atlantic oceanic basin, including from far-field coseismic or volcanic sources, and near-field Submarine mass failures (SMFs); the latter are the object of this work. Despite clear field evidence of past large-scale SMFs within our area of interest, such as the Currituck slide complex, their magnitude, pre-failed geometry, volume, and mode of rupture are poorly known. A screening analysis based on the Monte Carlo simulations (MCS) identified areas for possible tsunamigenic SMF sources along the USEC, indicating an increased level of tsunami hazard north of Virginia, potentially surpassing the inundation generated by a typical 100-year hurricane storm surge in the region, as well as that from the most extreme far-field coseismic sources in the Atlantic; to the south, the MCS indicated that SMF tsunami hazard significantly decreased. Subsequent geotechnical and geological analyses delimited four high-risk areas along the upper USEC where the potential for large tsunamigenic SMFs, identified in the MCS, was realistic on the basis of field data (i.e., sediment nature and volume/availability). In the absence of accurate site-specific field data, following NTHMP’s recommendation, for the purpose of simulating tsunami hazard from SMF PMTs, we parameterized an extreme SMF source in each of the four areas as a so-called Currituck proxy, i.e., a SMF having the same volume, dimensions, and geometry as the historical SMF. In this paper, after briefly describing our state-of-the-art SMF tsunami modeling methodology, in a second part, we parameterize and model the historical Currituck event, including: (1) a new reconstruction of the SMF geometry and kinematics; (2) the simulation of the resulting tsunami source generation; and (3) the propagation of the tsunami source over the shelf to the coastline, in a series of nested grids. A sensitivity analysis to model and grid parameters is performed on this case, to ensure convergence and accuracy of tsunami simulation results. Then, we model in greater detail and discuss the impact of the historical Currituck tsunami event along the nearest coastline where its energy was focused, off of Virginia Beach and Norfolk, as well as near the mouth of the Chesapeake Bay; our results are in qualitative agreement with an earlier modeling study. In a third part, following the same methodology, we model tsunami generation and propagation for SMF Currituck proxy sources sited in the four identified areas of the USEC. Finally, as an illustration of our SMF tsunami hazard assessment work, we present detailed tsunami inundation maps, as well as some other products, for one of the most impacted and vulnerable areas, near and around Ocean City, MD. We find that coastal inundation from near-field SMF tsunamis may be comparable to that caused by the largest far-field sources. Because of their short propagation time and, hence, warning times, SMF tsunamis may pose one of the highest coastal hazards for many highly populated and vulnerable communities along the upper USEC, certainly comparable to that from extreme hurricanes.

Modeling coastal tsunami hazard from submarine mass failures: effect of slide rheology, experimental validation, and case studies off the US East Coast

We perform numerical simulations to assess how coastal tsunami hazard from submarine mass failures (SMFs) is affected by slide kinematics and rheology. Two types of two-layer SMF tsunami generation models are used, in which the bottom (slide) layer is depth-integrated and represented by either a dense Newtonian fluid or a granular flow, in which inter-granular stresses are governed by Coulomb friction (Savage and Hutter model). In both cases, the upper (water) layer flow is simulated with the non-hydrostatic 3D σ-layer model NHWAVE. Both models are validated by simulating laboratory experiments for SMFs made of glass beads moving down a steep plane slope. In those, we assess the convergence of results (i.e., SMF motion and surface wave generation) with model parameters and their sensitivity to slide parameters (i.e., viscosity, bottom friction, and initial submergence). The historical Currituck SMF is simulated with the viscous slide model, to estimate relevant parameters for simulating tsunami generation from a possible SMF sited near the Hudson River Canyon. Compared to a rigid slump, we find that deforming SMFs of various rheology, despite having a slightly larger initial acceleration, generate a smaller tsunami due to their spreading and thinning out during motion, which gradually makes them less tsunamigenic; the latter behavior is controlled by slide rheology. Coastal tsunami hazard is finally assessed by performing tsunami simulations with the Boussinesq long wave model FUNWAVE-TVD, initialized by SMF tsunami sources, in nested grids of increasing resolution. While initial tsunami elevations are very large (up to 25 m for the rigid slump), nearshore tsunami elevations are significantly reduced in all cases (to a maximum of 6.5 m). However, at most nearshore locations, surface elevations obtained assuming a rigid slump are up to a factor of 2 larger than those obtained for deforming slides. We conclude that modeling SMFs as rigid slumps provides a conservative estimate of coastal tsunami hazard while using a more realistic rheology, in general, reduces coastal tsunami impact.

An unstructured grid hydrodynamic and sediment transport model for Changjiang Estuary

An unstructured grid hydrodynamic and sediment transport model for Changjiang Estuary is developed in the current paper. The model employs finite volume method to discretize the governing equations. Semi-implicit method originally developed by Casulli is utilized to remove the stability limitations associated with the surface gravity wave. A wetting and drying (WAD) scheme is proposed to account for the moving boundary at the shoals and tidal flats. The model is used to investigate the hydrodynamics and sediment transport in the Changjiang Estuary. Comparisons with the measured data show that the model can predict water level and tidal current very well. The variations of sediment concentration are also reasonably captured by the model.

Numerical study of sediment transport on a tidal flat with a patch of vegetation

To understand how vegetation canopies affect sediment transport on tidal flats, a numerical study of tidal flow and sediment transport on an idealized tidal flat with a patch of vegetation is conducted. The numerical model is firstly validated by laboratory measurements of flow and sediment deposition in a partially vegetated open channel. The idealized study shows that a finite patch of vegetation may produce circulation on the tidal flat with converging flow during flood and diverging flow during ebb. The vegetation patch can also generate a tidal phase lag between the vegetated and bare flats. Tidal currents in both zones are asymmetric, with stronger flood current in the vegetated zone and stronger ebb current on the bare flat. The duration of ebb is longer than that of flood. Computed sediment concentration on the bare flat is higher during ebb due to stronger ebb current and larger bottom shear stress. This is in contrast to the tidal flat without a vegetation canopy, where suspended sediment concentration is higher during flood. On the tidal flat without a vegetation canopy, landward net sediment transport occurs on the upper flat, while seaward net sediment transport occurs on the lower flat and subtidal region. On the partially vegetated tidal flat, however, net sediment transport on both the upper and lower flats are in seaward direction. It increases with increasing vegetation density. Alongshore net sediment flux converges inside the canopy and diverges on the bare flat. Sediment exchange rate between the vegetated and bare flats increases with decreasing vegetation density and sediment settling velocity.

High-Resolution Non-Hydrostatic Modeling of Frontal Features in the Mouth of the Columbia River

Airborne data measured during the recent RIVET II field experiment has revealed that horizontally distributed thermal fingers regularly occur at the Mouth of Columbia River (MCR) during strong ebb tidal flows. The high-resolution, non-hydrostatic coastal model, NHWAVE, predicts salinity anomalies on the water surface which are believed to be associated with the thermal fingers. Model results indicate that large amplitude recirculation are generated in the water column between an oblique internal hydraulic jump and the North Jetty. Simulation results indicate that the billows of higher density fluid have sufficiently large amplitudes to interrupt the water surface, causing the prominent features of stripes on the surface. The current field is modulated by the frontal structures, as indicated by the vorticity field calculated from both the numerical model and data measured by an interferometric synthetic aperture radar.

On the Shoaling of Solitary Waves in the Presence of Short Random Waves

Overhead video from a small number of laboratory tests conducted by Kaihatu et al. at the Tsunami Wave Basin at Oregon State University shows that the breaking point of a shoaling solitary wave shifts to deeper water if random waves are present. The analysis of the laboratory data collected confirms that solitary waves indeed tend to break earlier in the presence of random wave field, and suggests that the effect is the result of the radiationstresses gradientinduced by the randomwavefields.A theoreticalapproachbased on the forced KdV equation is shown to successfully predict the shoaling process of the solitary wave. An ensemble of tests simulated using a state-of-the-art nonhydrostatic model is used to test the statistical significance of the process. The results of this study point to a potentially significant oceanographic process that has so far been ignored and suggest that systematic research into the interaction between tsunami waves and the swell background could increase the accuracy of tsunami forecasting.

Sensor Selection and Integration to Improve Video Segmentation in Complex Environments

Background subtraction is often considered to be a required stage of any video surveillance system being used to detect objects in a single frame and/or track objects across multiple frames in a video sequence. Most current state-of-the-art techniques for object detection and tracking utilize some form of background subtraction that involves developing a model of the background at a pixel, region, or frame level and designating any elements that deviate from the background model as foreground. However, most existing approaches are capable of segmenting a number of distinct components but unable to distinguish between the desired object of interest and complex, dynamic background such as moving water and high reflections. In this paper, we propose a technique to integrate spatiotemporal signatures of an object of interest from different sensing modalities into a video segmentation method in order to improve object detection and tracking in dynamic, complex scenes. Our proposed algorithm utilizes the dynamic interaction information between the object of interest and background to differentiate between mistakenly segmented components and the desired component. Experimental results on two complex data sets demonstrate that our proposed technique significantly improves the accuracy and utility of state-of-the-art video segmentation technique.