ALBERT

Description: Characterizing surface deformation throughout a full earthquake cycle is a challenge due to the lack of high-resolution geodetic observations of duration comparable to that of characteristic earthquake recurrence intervals (250–10,000 years). Here we approach this problem by comparing long-term geologic slip rates with geodetically derived fault slip rates by sampling only a short fraction (0.001%–0.1%) of a complete earthquake cycle along 15 continental strike-slip faults. Geodetic observations provide snapshots of surface deformation from different times through the earthquake cycle. The timing of the last earthquake on many of these faults is poorly known, and may vary greatly from fault to fault. Assuming that the underlying mechanics of the seismic cycle are similar for all faults, geodetic observations from different faults may be interpreted as samples over a significantly larger fraction of the earthquake cycle than could be obtained from the geodetic record along any one fault alone. As an ensemble, we find that geologically and geodetically inferred slip rates agree well with a linear relation of 0.94±0.09. To simultaneously explain both the ensemble agreement between geologic and geodetic slip-rate estimates with observations of rapid postseismic deformation, we consider the predictions from simple two-layer earthquake-cycle models with both Maxwell and Burgers viscoelastic rheologies. We find that a two-layer Burgers model, with two relaxation timescales, is consistent with observations of deformation throughout the earthquake cycle, whereas the widely used two-layer Maxwell model with a single relaxation timescale, is not, suggesting that the earthquake cycle is effectively characterized by a largely stress-recoverable rapid postseismic stage and a much more slowly varying interseismic stage.

Description: We explore the potential geodetic signature of mechanical stress shadows surrounding inferred major seismic asperities along the Japan-Kurile subduction megathrust. Such stress shadows result from a decrease in creep rates late in the interseismic period. We simplify the rupture history along this megathrust as the repeated rupture of several asperities, each with its own fixed recurrence interval. In our models, megathrust creep throughout the interseismic period evolves according to velocity strengthening friction, as opposed to common kinematic back-slip models of locked or partially locked (i.e. coupled) regions of the megathrust. Such backslip models are usually constrained by onshore geodetic data and typically find spatially extensive and smooth estimates of plate coupling, a likely consequence of model regularization necessitated by poor model resolution. Of course, these large coupled regions could also correspond to seismogenic asperities, some of which have not experienced a significant earthquake historically. A subset of existing kinematic models of coupling along the Japan Trench, particularly those that use both horizontal and vertical geodetic data, have inferred a surprisingly deep (~100 km) locked zone along the megathrust or have called upon complex, poorly constrained megathrust processes, such as subduction erosion, to explain the geodetic observations. Here, we posit two scenarios for distributions of asperities on a realistic 3-D megathrust interface along the Japan-Kurile Trench off NE Japan. These scenarios reflect common assumptions made before and after the 2011 M w 9 Tohoku-oki earthquake. We find that models that include two shallow M 9-class asperities (one corresponding to the 2011 Tohoku-Oki earthquake and one offshore of Hokkaido) and associated stress-shadows can explain geodetic observations of interseismic strain along the eastern halves of Honshu and Hokkaido. Specifically, models including localized fault creep can explain most of the observed long-term vertical subsidence in this region during the past century and thus appealing to processes such as deep locking or subduction erosion may not be required.

Description: We investigate the rupture process of the 25 April 2015 M w  7.9 Gorkha, Nepal, earthquake and its biggest aftershock on 12 May 2015, based on joint inversion of teleseismic body waves, Interferometric Synthetic Aperture Radar, and Global Positioning System measurements. The Gorkha earthquake propagated unilaterally to the southeast along the Himalayan thrust fault (HTF), with coseismic slip separating into patches up-dip and down-dip of the hypocenter. Slip in the up-dip patch initially surrounded a region on the fault that did not slip. About 15 s after being surrounded, this region of the HTF then slipped, filling in the initial slip deficit. The delayed slip accounts for ~20% of the moment release in the Gorkha earthquake. The inferred coseismic slip in the Kodari earthquake is localized to one patch on the HTF, extending to the south and southeast from the hypocenter and 20–30 km to the northeast of the main slip patch in the Gorkha earthquake. The maximum coseismic slip in both the Gorkha and Kodari earthquakes is ~4.5 m.

Description: We explore the impact of deep ductile shear zones on post-seismic deformation following a finite length strike-slip earthquake. We show that the pattern of post-seismic vertical surface deformation surrounding the fault is a discriminant for the existence of high viscosities immediately below the seismogenic layer, regardless of whether the model contains purely distributed creep or also includes a component of localized creep at subseismogenic depths. Post-seismic deformation characterized by initially fast relaxation followed by a slower relaxation is predicted by models that include both localized creep in a subseismogenic shear zone and distributed creep in the surrounding region, even if they only contain steady Maxwell viscoelasticity. This post-seismic deformation is similar to that in models that approximate the ductile lithosphere and/or asthenosphere with Burgers viscoelasticity. We find that the post-seismic deformation following the 1997 M w 7.6 Manyi, China, earthquake, is consistent with a post-seismic model composed of a lower Maxwell viscoelastic region with viscosity 10 19 Pa s and a 5 km wide, Maxwell viscoelastic shear zone with viscosity 10 18 Pa s beneath the fault.

Description: Post-seismic deformation is commonly attributed to viscoelastic relaxation and/or afterslip, although discerning between the two driving mechanisms can be difficult. A major complication in modeling post-seismic deformation is that forward models can be computationally expensive, making it difficult to adequately search model space to find the optimal fault slip distribution and lithospheric viscosity structure that can explain observable post-seismic deformation. We propose an inverse method which uses coseismic and early post-seismic deformation to rapidly and simultaneously estimate a fault slip history and an arbitrarily discretized viscosity structure of the lithosphere. Our method is based on an approximation which is applicable to the early post-seismic period and expresses surface deformation resulting from viscoelastic relaxation as a linearized function with respect to lithospheric fluidity. We demonstrate this approximation using two-dimensional earthquake models. We validate the approximation and our inverse method using two three-dimensional synthetic tests. The success of our synthetic tests suggests that our method is capable of distinguishing the mechanisms driving early post-seismic deformation and recovering an effective viscosity structure of the lithosphere.

Description: The M w  7.8 Ecuador earthquake on 16 April 2016 is the sixth earthquake larger than M w  7 to rupture the subduction megathrust between the Nazca and South American plates since the M w  8.8 Colombia–Ecuador earthquake in 1906. We use Interferometric Synthetic Aperture Radar (InSAR) images from Sentinel-1A to determine coseismic surface displacements associated with this earthquake. The interferograms exhibit a relatively simple pattern of deformation, with maximum displacements of 0.7 and 0.3 m on the descending and ascending images, respectively. We invert the interferograms for both rupture geometry and slip distribution in the earthquake. We find that the data are best described by slip on a fault dipping 17° to the east and that these InSAR data cannot uniquely constrain the strike. The maximum inferred slip is just over 2.5 m at about 20 km depth, with the main slip in the depth range of about 10–25 km. The geodetic moment of our preferred slip model is 7.15 x 10 20 N·m, equivalent to M w  7.87. Our results suggest that there is little, if any, partitioning of the oblique plate convergence. The 2016 Ecuador earthquake is coincident with the location of the 1942 M w  7.8 earthquake, with both earthquakes most likely rupturing an asperity that also failed in the 1906 M w  8.8 earthquake.