Introduction to PTHA

One use of tsunami modeling is to perform hazard assessment for a community or structure in relation to possible future tsunamis. Since it is impossible to know the details of future tsunamis, this is often done either by

• A scenario-based approach, in which one or more hypothetical future events are considered (for example one event that is thought to be a “worst case”, or a few hypothetical events of varying severity).

• A probabilistic approach where some probability distribution is assumed for future events and then we attempt to determine the resulting probability distribution of some quantity of interest, such as the inundation depth at a location or as a map over an entire community.

There is often some overlap between these approaches, e.g. by a “worst case” event we might mean one that has at least a 2500-year return time according to some probabilistic model. Or to approximate the probability distribution of a quantity of interest we may sample a large number of scenarios from the underlying probability distribution and perform a tsunami simulation of each.

Review paper and Tutorial¶

The review paper of Grezio et al. 2017, “Probabilistic Tsunami Hazard Analysis: Multiple Sources and Global Applications” provides an overview of PTHA. As part of this, a tutorial was created in the form of Jupyter notebooks. These notebooks have been adapted for these notes can can be find in the following sections. See PTHA Tutorial from Review Paper for more information and links.

A self-contained repository containing these notebooks can be found at https://github.com/rjleveque/ptha_rog.