Howell, A.; Penney, C.; Kaiser, A.E.; Fry, B. 2023 Modelling ground motions in the Greater Wellington region from multi-fault earthquakes in central Aotearoa New Zealand. Lower Hutt, NZ: GNS Science. GNS Science report 2023/45. 19 p.; doi: 10.21420/Z9SM-0G27
Abstract
Combined rupture of the Hikurangi subduction interface with crustal faults like the Wellington or Wairarapa faults may represent a significant source of seismic hazard for the Greater Wellington region. However, until now, it has been challenging to incorporate combined subduction-crustal ruptures in seismic hazard analyses, mainly due to a lack of empirical constraints on the likely characteristics of such ruptures. Recently, multi-cycle physics-based earthquake simulators – which do not rely on empirical constraints – have allowed modelling of combined subduction-crustal earthquakes. These earthquake simulators produce synthetic earthquake catalogues that contain thousands of hypothetical (but realistic) combined subduction-crustal and multi-fault crustal earthquakes. In this report, we model ground shaking from a range of synthetic combined subduction-crustal ruptures to: (1) investigate a method for inclusion of subduction-crustal earthquakes in future seismic hazard assessments; (2) assess the relative contributions to ground shaking from the subduction and crustal components of a combined rupture; and (3) illustrate potential spatial distributions of ground shaking from hypothetical future subduction-only, crustal-only and combined subduction-crustal earthquakes. We show that modelling the ground shaking due to movements of crustal and subduction faults separately and re-combining the results may be an effective way to include subduction-crustal earthquakes in seismic hazard models, but more work is required to validate the approach. We also find that, for combined ruptures in general, the movement of crustal faults dominates modelled ground shaking in the area immediately adjacent to their surface trace. Further from the crustal faults (>5–15 km away), ground motions are dominated by the subduction component of the earthquake. This finding can apply even when the magnitude of the subduction component of the earthquake is much greater than the magnitude of the crustal component. Finally, we show that crustal and subduction earthquakes can both cause significant ground accelerations across the Greater Wellington region at a range of frequencies. This result suggests that it may be challenging to determine likely sources of past earthquakes from off-fault proxies, such as the spatial distribution of landslides. (auths)
