Towards integrating information about strain rates in PSHA models in Europe: comparison between seismic moment rates from ESHM20 model and geodetic estimates

Most national and international seismic regulations require quantifying seismic hazard based on probabilistic seismic hazard assessment (PSHA) methods. The probabilities of exceeding ground-motion levels at sites of interest over a future time window are determined by combining a source model and a ground-motion model. This research work aims at understanding how the measurement of strain rates by geodesy can provide constraints on the source model.

Earthquake catalogs, merging instrumental and historical data, are usually used to establish earthquake recurrence models. Although these catalogs extend over several centuries, the observation time windows are often short with respect to the recurrence times of moderate-to-large events and in some regions the recurrence models can be weakly constrained.

Here, we compute different realizations of strain rates maps over Europe using a combined velocity field (Piña Valdes et al., JGR submitted). These strain rates are compared to the source model of the new European seismic hazard model (ESHM20, Danciu et al. 2021). More precisely, the moment rates estimated from the earthquake recurrence models are compared to the geodetically-derived moment rates.

We explore the different uncertainties in both models. For geodesy, we integrate uncertainties on the velocities at each station, and the epistemic uncertainties on the different steps of the computation of the geodetic moment rate : filtering of the velocity field (outliers’ removal and smoothing) (Piña-Valdés et al, JGR submitted), parameters used to drive the strain inversion with VISR software (Shen et al. 2015), constants and formula used for the moment computation. The goal is to quantify the impact of each parameter uncertainty or decision on the moment computation.

The first results show that a correlation exists between the seismically and geodetically derived moment rates. In general, the main uncertainty is on the velocities at each stations, followed by the depth taken into account for the moment computation. In areas characterized by high activity, such as Betics or Apennines for example, the moment rates derived by both methods are comparable. In areas of lower activity, such as at the interior of plates, the error associated with geodetic measurements is of the same order of magnitude as the measured strain, and the relation between catalog-based and strain-based moment is not straightforward.