The 2018 M_w 7.9 Offshore Kodiak, Alaska, Earthquake: An Unusual Outer Rise Strike‐Slip Earthquake

Abstract

The rupture process of the 2018 Mw 7.9 offshore Kodiak (Alaska) earthquake is still in hot dispute because of a lack of offshore observations, thus causing difficulties for understanding seismogenic tectonics and tsunami hazards. In this study, teleseismic body waves, high‐rate GPS, seismic waves preceding tsunami waves, static GPS, and tsunami data are jointly used to resolve the faulting geometry and coseismic slip distribution of the offshore Kodiak earthquake. Tests of a series of finite fault rupture models illustrate that an optimal five‐fault model can reconcile all available data sets well. The results reveal that the asperity on each fault segment is located near the hypocenter, with peak slip of ~7.8 m. The aftershock loci appear to be complementary to the mainshock slips, demonstrating the velocity‐strengthening regions that predominantly slip aseismically. Based on a tectonic stress field analysis, we propose that the 2018 Kodiak earthquake is attributed to a combination of enhanced tensional stress following the 1964 Mw 9.2 Alaska earthquake and compressional stress produced by the collision of the Yakutat terrane with North America.

Consecutive ruptures on a complex conjugate fault system during the 2018 Gulf of Alaska earthquake

“…Our inversion results indicate that the main rupture stage (0–27 s after origin) affected segments oriented both N-S and E-W, suggesting that the 2018 Gulf of Alaska earthquake ruptured a conjugate fault system, as proposed in previous studies 4 – 8 . Our source model suggests that the rupture occurred along weak zones in the sea floor: fracture zones extending E-W and plate-bending faults parallel to N-S magnetic lineaments 29 , 30 .…”

“…(4) and (5) was set such that min k m q = 1 (Table S3). Based on the aftershock distribution, the 2018 Gulf of Alaska earthquake is considered to have occurred on a quasi-orthogonal multiple-fault system [4][5][6][7][8] . To cover the high point density area of aftershocks within one week of the event 1 (Fig.…”

“…For the inversion analysis, we adopted a potency-rate density function on each knot, each representing a linear combination of B-splines at an interval of 0.8 s. The maximum rupture-front velocity, which defines the rupture starting time at each knot, was set to 7.0 km/s to account for the possibility of supershear rupture propagation. The rupture ending time at each knot was set to 65 s from the origin time based on previous inversion results 4 , 6 . We evaluated the sensitivity of our model by perturbing the model parameters, and the robustness of the new method (see Supplementary Material S2, and Figs.…”

“…Several pioneering studies that built finite-fault models based on the aftershock distribution demonstrated that the 2018 Gulf of Alaska earthquake ruptured a quasi-orthogonal multiple-fault system oriented approximately N-S and E-W 4 – 8 . However, it is difficult to adopt a reasonable fault model because the fault model parametrization, number of fault segments, and fault geometries differ by study, partly due to the spatial spread of the aftershock distribution (Fig.…”

“…Based on the static slip distribution estimated from Global Navigation Satellite System and tsunami data, major slips occurred on E-W-striking segments 5 , 7 , 8 . Finite-fault inversions estimated that the maximum slip occurred around the boundary between the crust and uppermost mantle in the N-S-oriented segment 4 , 6 , which would have played a significant role in tsunami generation. However, it remains challenging to adequately explain the complex characteristics of the observed teleseismic body waveforms by conventional finite-fault inversion methods due to the uncertainty on the fault geometry, which lead to significant model errors.…”