GeoClaw Validation against the Great Tohoku Tsumani of 11 March 2011

点击翻译按钮获取中文摘要

The Great Tohoku Tsunami of 11 March 2011 has generated massive quantities of data that will be invaluable in helping to validate tsunami models such as the GeoClaw software being developed by the PI and his collaborators. This RAPID project will provide short-term support for two postdoctoral fellows and two graduate students to assist in collecting data and performing simulations. The main goals are to further validate the software and constituent numerical algorithms as tools for tsunami modeling, and to compare a variety of proposed earthquake source mechanisms for this event that have been developed by different groups of seismologists. Data sources will include NOAA DART buoys, tide gauges, videos and photographs, and post-tsunami surveys, both in the near field and far field. The postdoctoral fellows are experts in tsunami science who have experience in working with a variety of tsunami data, also have experience working with the GeoClaw software in their thesis work. The software implements high-resolution finite volume methods for the nonlinear shallow water equations, with provisions to robustly model inundation. Adaptive mesh refinement is incorporated to allow the efficient solution of the problem over the vast range of spatial scales required.Tsunamis pose a severe threat to many coastal regions of the United States, particularly in the Pacific Northwest. The offshore Cascadia Subduction Zone (CSZ) running from Northern California up to British Columbia periodically produces a magnitude 9 earthquake, generating a tsunami of magnitude similar to that seen in the Indian Ocean in 2004 or from the Japanese tsunami of 2011 (The last major CSZ tsunami was in 1700). The entire West Coast, Hawaii, and Alaska are also at risk from tsunamis arising from many other potential sources. Hazard mitigation and early warning systems require accurate computer simulation tools based on mathematical models of the generation and propagation of tsunamis and the subsequent flooding and inundation along the coast. It is important to test these models against real events, and the 11 March 2011 Great Tohoku Tsunami provides a wealth of data from a variety of sources. Performing a careful series of computational experiments will assist in validating the computational model developed at the University of Washington and will guide the development of further improvements to the model.