Tsunamis are some of the most devastating natural disasters than can occur. In just the last 15 years, two tsunamis - the 2004 Indian Ocean tsunami and the 2011 Japan tsunami - killed hundreds of thousands of people and destroyed billions of dollars of property. Despite the importance of understanding these dangerous waves, there is still much we do not understand about how tsunamis are generated.
The largest tsunamis are caused by megathrust earthquakes in subduction zones, when shallow coseismic slip between tectonic plates causes the seafloor to deform, uplifting the ocean surface and initiating a tsunami. Tsunamis can also be caused by earthquakes with smaller magnitude that are more efficient at generating tsunamis. These are called “tsunami earthquakes,” and they may result from slip along high angle splay faults or through a very compliant wedge of sedimentary materials in the trench.
When an earthquake generates a tsunami, it also excites a wide range of fast-propagating seismic and ocean acoustic waves, some of which get trapped in the ocean and may contain valuable information about the size of the tsunami. These trapped waves could potentially be useful for improving tsunami early warning systems.
To better understand these types of problems, we use numerical models that fully couple dynamic rupture on the fault to the elastic response of the earth and ocean. This means that we can model the full seismic, ocean acoustic, and tsunami wavefield that results from a subduction zone earthquake. This way we can explore and investigate some of the complexities of tsunami generation.