5.4 Constraints on the Updip Coseismic Slip Distribution and Postseismic Mechanisms
The updip portion of the published coseismic slip models varies considerably (Figures 2, 7), and it remains un which model better resolves the up-dip portion of the coseismic slip. The Ye et al. (2022) and Liu et al. (2023) models used tsunami data to iteratively adjust the coseismic model. These models differ from the Elliott et al. (2022), Liu et al. (2022) and Mulia et al. (2022) model in having a distinct slip patch that has very large slip magnitude (~ 10m) and a rake of 45 degree under Chirikof island. Aside from that shallow patch, the Ye et al. (2022) and Liu et al. (2023) models overall restrict slip to greater depths than do the Elliott et al. (2022) model, or the models shown here. However, the solutions fit the tsunami data are not unique. Liu et al., (2023) Figure S7 shows that the Elliott et al., (2022) model, which has about ~1m slip extending slightly beyond the continental shelf break, also fits the tsunami data well. Brooks et al. (2023) modeled the coseismic (and 2.5 month postseismic) displacement from a GNSS-Acoustic site seaward of the Chignik rupture and have also argued for greater slip extending to shallower depths than some of the models proposed, although the slip could be in the form of rapid shallow afterslip.
Its more difficult to constrain the up-dip post-seismic slip compared to the downdip portion. The main limitation is that we only have one GPS continuous site, AC13, at the updip end of the coseismic rupture area. Stress-driven afterslip models do not predict the very large postseismic displacement at AC13, unless they include a large coseismic slip patch very close to AC13 (like the Ye et al., (2022) and Liu et al., (2023) models). However, considering the magnitude of this earthquake and the location of site AC13, the effects of poroelastic relaxation, depth variations in the frictional parameters (e.g., Tian et al. (2023)), the potential existence of a weak sub-slab oceanic mantle layer could also affect the model prediction. Thus, there could be considerable non-uniqueness in the model. Due to the limited number of GPS sites located at the updip end of the coseismic rupture, it is difficult to fully separate the postseismic mechanisms, making it also difficult to assess the exact contribution of shallower afterslip and the shallower portion of coseismic slip of this event. A significant expansion of seafloor geodesy will be required to answer those remaining questions.