Today "Statistical emulation of landslide-induced tsunamis at the Rockall Bank, NE Atlantic" was publishing in The Proceedings of the Royal Society A. The paper is writen by D. M. Salmanidou (UCD School of Mathematics and Statistics) S. Guillas (University College London), A. Georgiopoulou (UCD School Of Earth Sciences) and F. Dias (UCD School of Mathematics and Statistics).
The team built a statistical emulator that approximates the deterministic codes used to simulate the submarine sliding and tsunami generation that occurred 21,000 years ago at the Rockall Bank, NE Atlantic Ocean. This is the first uncertainty quantification of a realistic tsunami for a past (but uncertain) landslide event. Uncertain landslide characteristics are obtained by calibrating the landslide model outputs against observed run-out distances. Since numerical tsunami simulations are expensive to run many times, the team run the fast emulator to efficiently propagate source uncertainties to uncertainties in resulting tsunami waves.