Investigating the tsunamigenic potential of crustal faults in the Strait of Georgia

Coastal communities around the Strait of Georgia, between Vancouver Island, British Columbia and the mainland of North America, may be relatively sheltered from tsunamis sourced from both distant and local subduction zone earthquakes. However, these low-elevation communities, located in the forearc of the northern Cascadia subduction zone, are likely vulnerable to local tsunamis generated by the rupture of submarine crustal faults. Active crustal faults within and surrounding the Strait of Georgia remain relatively poorly understood, but several show evidence of vertical seafloor displacements. These faults likely have long recurrence intervals, but their close proximity to communities means there would not be much warning time between an earthquake and the arrival of waves. Ruptures along multiple fault splays or triggering of landslides during an event could also create larger localized wave heights. In this study we focus on three recently-identified crustal faults that show vertical Holocene offsets and likely accommodate transpression. These include the NW-SE trending Birch Bay and Sandy Point faults on the south side of Boundary Bay, and the E-W trending Skipjack Island fault zone that crosses Georgia Strait through Pender Island and north of Orcas Island. It remains a challenge to constrain the detailed structure and rupture mechanisms for these faults, which have not hosted large historical earthquakes. Previous onshore paleoseismic investigations have revealed sudden north-side-up Holocene offsets on both the Birch Bay and Sandy Point faults, and provide constraints on the amount of slip per event. Some of these offsets may be coincident with tsunami deposits found in Boundary Bay. Our analysis of cross-border magnetic and bathymetric data suggests that both of these faults continue offshore further north than previously mapped. Previous reflection seismic lines across the Skipjack Island fault zone show significant north-side-up offsets of late Quaternary sediments, but the possible slip per event can only be constrained using empirical scaling relations. We present preliminary model results for surface displacements from a number of rupture scenarios of these three faults and resultant tsunami waves.