A backbone seismic ground motion model for strike-slip earthquakes in Southwest Iceland and its implications for near- and far-field PSHA

The earthquake hazard and seismic risk in Iceland are highest in the Southwest due to the transform faulting in the South Iceland Seismic Zone (SISZ) and Reykjanes Peninsula Oblique Rift (RPOR) being in close proximity to a large part of the population. Reliable probabilistic seismic hazard assessment (PSHA) is therefore critical in this region which in turn requires two of its key elements: the most appropriate ground motion models (GMMs) and the specification of seismic sources. In this study, we address this by employing a suite of new hybrid Bayesian empirical GMMs and a new physics-based finite-fault system model for the SISZ–RPOR. By ranking the GMMs using the deviance information criterion against the Icelandic strong-motion dataset we propose backbone GMMs (i.e., the most appropriate models), which along with their backbone Δ -factors not only comprehensively capture the characteristics of Icelandic data but also the epistemic uncertainties of their ground motion prediction. We then simulate suites of synthetic finite-fault earthquake catalogues that are consistent with the physics-based fault system and long-term seismic activity of the region. We carry out a Monte-Carlo PSHA for two representative near-fault (the town of Selfoss) and far-field (the capital city of Reykjavik) sites in Southwest Iceland, and compare the results with those of a classical PSHA. We show that the backbone PSHA is consistent with classical PSHA point estimates but more importantly, the backbone approach reveals the “body” of the hazard i.e., the majority of realistically expected hazard values, which dwarfs any differences in the results from the two approaches. The PGA results show that the body of hazard values varies from 0.07-0.13 g and 0.4-0.6 g for Reykjavik and Selfoss, respectively. The scale factor is distance- and period-dependent that results in a constant body of the backbone hazard values for a far-field site at all periods while at long periods and decreasing annual exceedance rates, the body progressively increases. Moreover, the disaggregation showed that the modal event for a 475 year return period is a Mw ~ 6–6.5 at distance ~ 14–26 km for the far-field site, but at ~ 6 km for the near-fault site. We conclude that the backbone approach and the models used in this study make ideal candidates for improved PSHA for Iceland.