5.1 The downdip limit of the coseismic rupture
All of the models that predict the postseismic displacements well have a slip distribution that is located well offshore. They also have a relatively abrupt downdip termination of slip, and a clean separation between the zones of coseismic slip and afterslip (Figures 6, 9). Because the basic characteristics of the slip models are not very sensitive to assumed fault depth, and there is some uncertainty in the geometry and depth of the plate interface, the horizontal location of the downdip end of the rupture is determined more precisely than the depth.
Figure 1 shows comparisons between the co-seismic slip region of the Elliott et al. (2022) model and our preferred 10km fault width model, along with the aftershock distribution. The aftershock region matches more closely to the slip area with our preferred narrower fault than with the original Elliott et al. (2022) model. Based on our preferred coseismic rupture model, the downdip afterslip did not trigger a significant number of downdip aftershocks. This suggests that the coseismic rupture extended in depth to the deepest extent of the velocity-weakening friction, or beyond it given stress shadowing effects (e.g., Lindsey et al., 2021).
There are several clusters of aftershocks updip of the coseismic rupture, which we interpret to be triggered by the updip afterslip. he existence of these aftershock clusters suggests that some of the fault plane updip of the coseismic rupture likely has velocity-weakening frictional behavior.