Long-term slip history discriminates among occurrence models for seismic hazard assessment

We propose a Bayesian framework for the combination of catalogs of large earthquakes and dated cumulative slip data. It provides a quantitative way of discriminating or ranking the different renewal models. Indeed, once the datasets and priors are chosen, no additional expert opinion is required, and the ranking comes out of the Bayes factors directly. This way, the experience of experts is valorized, but the effects of group dynamics do not directly influence the weighing of the different candidate models