Large Vessel and Debris Transport due to Tsunami-Induced Currents

The observations from the recent events revealed that the currents generated by the tsunamis can be as disastrous as waves if not more when the heights of the incoming waves are relatively small to cause any inundation. Tsunamis can generate localized high energy areas within the ports and harbors due to flow convergence caused by the infrastructure and even large vessels near these areas can easily be picked up by the flow. This talk will describe a model developed to predict the motion of the large vessels within the ports if they are detached from their moorings as a result of these strong tsunami currents. Same tool with adequate modifications can also be utilized to model the transport of small boats or the debris as well. With this study, we are seeking to be able to provide quantitative guidance to port and harbor authorities so that efficient mitigation measures can be developed.

This tool consists of a transport module based on the linear ship motion model; a collision solver adopting the concepts of conservation of momentum and impulse, and a hydrodynamic solver, MOST, which is modified to allow for the flow-vessel interaction. In restricted waters, the interaction between the flow and the deep draft vessels become more profound and overseeing this fact can lead to significant magnitudes of error in drag force estimates provided by the model. The impact of the vessels on the predicted flow field will be demonstrated in the talk with the examples from the simulations carried out for the Port of Long Beach.

Moreover, the findings of a sensitivity analysis performed based on an incident took place in Ishinomaki Port, Japan, during the 2011 Tohoku Tsunami involving a 180-meter-long bulk carrier will also be discussed. With this, the aim is to understand the effects of the initial conditions that cannot be known precisely before the incident, like draft of the ship or the exact time of the mooring line failure, on the predicted path of the vessel. The results of the sensitivity analysis indicated that the process is inherently random and, in general, the model outputs are very sensitive to the initial choice of the input parameters. Therefore, pure deterministic methods for this type of analysis are of limited use.