VERTICAL PALEOGEODETIC CONSTRAINTS ON MEGATHRUST RUPTURE IN GREAT EARTHQUAKES AT THE CASCADIA SUBDUCTION ZONE

Coastal marshes record a 6,500-year history of coseismic displacements in great earthquakes at the Cascadia subduction zone. We first compile published estimates of vertical coseismic displacement for past megathrust events, using correlations with megathrust-triggered turbidites. Age-correlated marsh data are compatible with event rupture extents defined by the published turbidite record, and a recurrence interval (averaged over 6,500 years) that increases northwards from ~230 years in the south to ~480 years for full-margin ruptures. We then compare the coseismic subsidence pattern with the predictions of elastic dislocation models for great earthquake rupture. Within the marsh data uncertainties, the coseismic rupture zone constrained by marsh subsidence generally agrees with the interseismic locked zone downdip width inferred from geodetic and thermal data. However, in southernmost Cascadia, where the model does not include the complex deformation near the Mendocino triple junction, the coastal data may be better fit by a model with a ~25% narrower rupture than inferred from regional geophysical data. We also estimate megathrust slip based on comparisons of marsh coseismic displacements with elastic dislocation model predictions. At each site, displacements tend to be averaged from event to event, independent of the time since the previous event. Slip in the 1700 earthquake was consistent with the preceding interval of strain accumulation (~200 years) only at the northern and southern ends of the margin. In southern Washington/northern Oregon, greater-than-average slip in 1700 may indicate a catch-up event to make up for slip deficit in the preceding event. Overall agreement between the geophysically-constrained dislocation models and the marsh data for most of the margin implies that such models can be usefully applied to rupture and ground shaking predictions of future events. The modeling results also highlight areas where marsh data have the most potential to provide improved constraints on rupture. Future studies should focus on transfer function analysis of microfossil data sets to increase the quality and reduce the uncertainty of coseismic subsidence estimates, particularly in areas where better constraints would result, e.g., southern Oregon.