ALBERT

Description: The Mw = 9.0 earthquake of 11 March 2011 at the Japan Trench and its devastating tsunami underscore the importance of understanding seismogenic behavior of subduction faults and realistically estimating the potential size of future earthquakes and tsunamis. For the Cascadia subduction zone (Figure 1a), a critical knowledge gap is the level of microseismicity offshore, especially near the megathrust, needed to better understand the state of the locked zone. In 2010 the first detailed seafloor earthquake monitoring campaign along the northern Cascadia subduction zone recorded nearby earthquakes in the local magnitude (ML) range from possibly around zero to 3.8 (Figures 1b and 1c) and larger earthquakes from outside this region. Preliminary analyses indicate that the network appears to have yielded a fairly complete catalog for events with ML 〉 1.2. Only a few tens of these events occurred beneath the continental shelf and slope (Figure 1a). The majority of the earthquakes were located along the margin-perpendicular Nootka fault zone. The relatively low seismicity away from the Nootka fault is consistent with a fully locked megathrust. Land-based GPS measurements cannot resolve the question of whether the offshore part of the megathrust seismogenic zone is narrow and fully locked or wider and only partially locked (slowly creeping). If it were only partially locked, the seafloor seismometer data should show many more small earthquakes along the interface than were actually detected.

Description: At 8:04 P.M. Pacific daylight time (PDT) on 27 October 2012 (03:04 universal time (UT), 28 October), Canada's second largest instrumentally recorded earthquake rocked Haida Gwaii (formerly Queen Charlotte Islands) and the mainland coast of British Columbia. The M 7.7 event off the west coast of Moresby Island caused a tsunami with local runup of more than 7 meters and amplitudes up to 0.8 meter on tide gauges 4000 kilometers away in Hawaii. Shaking was felt as far away as the Yukon, Alberta, Washington, and Montana, up to 1500 kilometers away. Little damage was caused, as the immediate region is an uninhabited National Park Reserve. The closest point of the rupture zone, as defined by aftershocks (Figures 1a and 1c), was 50 kilometers from the nearest community, Queen Charlotte, where damage was confined to a few chimneys and slumped roads.

Description: We developed thermal models for the Chile subduction zone along two profiles at 38.2°S and 42°S within the rupture area of the 1960 M = 9.5 Valdivia earthquake and south of the 2010 M = 8.8 Maule earthquake. The age difference of the subducting Nazca Plate has a major impact on the thermal regime, being much younger and hotter in the south. Seafloor heat flow observations confirm this difference but also indicate that in the southern area, heat advection at the outer rise cools the incoming plate. Heat flow values derived from the depth of gas hydrate bottom-simulating reflectors are in general agreement with probe and borehole measurements. The positions where the plate interface reaches temperatures of 100–150°C and 350–450°C differ between the two profiles. If these temperatures control the updip and downdip limits of the interplate seismogenic zone, the seismogenic zone widens and shifts landward to greater depths from south to north. Observed microseismicity, however, seems to fade at temperatures much lower than 350–450°C. This discrepancy can be explained in three alternative ways: (1) deformation in a thick subduction channel controls the seismic/aseismic transition; (2) microseismicity recorded over a limited time period does not represent the rupture depth of large interface earthquakes; or (3) the serpentinized mantle wedge controls the downdip limit.

Description: For assessing tsunami hazard in northernmost Cascadia, there is an urgent need to define tsunami sources due to megathrust rupture. Even though the knowledge of Cascadia fault structure and rupture behaviour is limited at present, geologically and mechanically plausible scenarios can still be designed. In this work, we use three-dimensional dislocation modelling to construct three types of rupture scenarios and illustrate their effects on tsunami generation and propagation. The first type, buried rupture, is a classical model based on the assumption of coseismic strengthening of the shallowest part of the fault. In the second type, splay-faulting rupture, fault slip is diverted to a main splay fault, enhancing seafloor uplift. Although the presence or absence of such a main splay fault is not yet confirmed by structural observations, this scenario cannot be excluded from hazard assessment. In the third type, trench-breaching rupture, slip extends to the deformation front and breaks the seafloor by activating a frontal thrust. The model frontal thrust, based on information extracted from multichannel seismic data, is hypothetically continuous along strike. Our low-resolution tsunami simulation indicates that, compared to the buried rupture, coastal wave surface elevation generated by the splay-faulting rupture is generally 50–100% higher, but that by trench-breaching rupture is slightly lower, especially if slip of the frontal thrust is large (e.g. 100% of peak slip). Wave elevation in the trench-breaching scenario depends on a trade-off between enhanced short-wavelength seafloor uplift over the frontal thrust and reduced uplift over a broader area farther landward.

Description: What controls subduction megathrust seismogenesis downdip of the mantle wedge corner (MWC)? We propose that, in the region of the 2010 Mw=8.8 Maule, Chile, earthquake, serpentine minerals derived from the base of the hydrated mantle wedge exert a dominant control. Based on modeling, we predict that the megathrust fault zone near the MWC contains abundant lizardite/chrysotile‐rich serpentinite that transforms to antigorite‐rich serpentinite at greater depths. From the MWC at 32–40 km depth to at least 55 km, the predominantly velocity‐strengthening megathrust accommodated dynamic propagation of the 2010 rupture but with small slip and negative stress drop. The downdip distribution of interplate aftershocks exhibits a gap around the MWC that can be explained by the velocity‐strengthening behavior of lizardite/chrysotile. Interspersed velocity‐weakening and dynamic weakening antigorite‐rich patches farther downdip may be responsible for increased abundance of aftershocks and possibly for some of the high‐frequency energy radiation during the 2010 rupture.

Description: Highlights • Hypocenters within the subducted Explorer plate indicate slab deformation. • The oceanic slab is bending downward toward the northwest. • A complex sequence of focal mechanisms also indicates plate deformation. • Decreased seismic activity in the overriding plate indicates decoupling to the NW. • Deformation and decoupling could limit megathrust rupture propagation. Abstract At the northernmost extent of the Cascadia subduction zone, the Explorer plate subducts at approximately 2 cm/yr, less than half the rate of the Juan de Fuca plate to the south. The boundary between these two plates is known as the Nootka fault zone, which is one of the focuses of the Seafloor Earthquake Array Japan-Canada Cascadia Experiment (SeaJade). During this survey, an 6.4 earthquake occurred on 24 April 2014. This event and the subsequent aftershocks (referred to as the Nootka Sequence) reveal an approximately 40-km-long subducted fault within the Explorer Plate to the north of the Nootka fault zone. We infer that the fault is a subducted conjugate fault because of its nearly identical orientation to those seaward of the subduction front within the Nootka fault zone. The depth distribution and focal mechanisms of the aftershocks indicate significant margin-parallel deformation of the subducting plate. The subduction interface at the Nootka Sequence fault has been deflected downward to the northwest from a depth of approximately 15 – 25 km over a distance of 25 km. We propose two possible scenarios that are modified from previously suggested slab-tear model with induced margin-parallel mantle flow to explain the significant deformation of the young, warm subducting Explorer plate. To the northwest of this change in slab geometry, a lack of seismic activity above the plate interface indicates that the Explorer plate has partially decoupled from the overriding North America plate. We conclude that the geometric variation separating the southern Explorer plate from the north, along with decoupling and a possible intraslab tear, may be a significant combination to resist the propagation of a megathrust rupture across this boundary.

Description: At the northern Cascadia subduction zone, the subducting Explorer and Juan de Fuca plates interact across a transform deformation zone, known as the Nootka fault zone (NFZ). This study continues the Seafloor Earthquake Array Japan Canada Cascadia Experiment to a second phase (SeaJade II) consisting of nine months of recording of earthquakes using ocean-bottom and land-based seismometers. In addition to mapping the distribution of seismicity, including an M W 6.4 earthquake and aftershocks along the previously unknown Nootka Sequence Fault, we also conducted seismic tomography, which delineates the geometry of the shallow subducting Explorer plate (ExP). We derived hundreds of high-quality focal mechanism solutions from the SeaJade II data. The mechanisms manifest a complex regional tectonic state, with normal faulting of the ExP west of the NFZ, left-lateral strike-slip behaviour of the NFZ, and reverse faulting within the overriding plate above the subducting Juan de Fuca plate. Using data from the combined SeaJade I and II catalogs, we have performed double-difference hypocentre relocations and found seismicity lineations to the southeast of, and oriented 18° clockwise from, the subducted NFZ, which we interpret to represent less active small faults off the primary faults of the NFZ. These lineations are not optimally oriented for shear failure in the regional stress field, which we inferred from averaged focal mechanism solutions, and may represent paleo-configurations of the NFZ. Further, active faults interpreted from seismicity lineations within the subducted plate, including the Nootka Sequence Fault, may have originated as conjugate faults within the paleo-NFZ.

Description: Highlights • Stress in cratons is limited by tectonic force from distant plate boundaries. • Average stress in cratonic lithosphere decreases as competent thickness increases. • Craton lithosphere undergoes mostly elastic deformation. • Brittle yielding and seismicity are restricted to very shallow crustal levels. • Effective elastic thickness is large even though average differential stress is low. Whether lithospheric stress can reach the maximum level predicted by the Christmas-tree strength envelope is a fundamental question but with controversial answers. There is little controversy that a deforming lithosphere in high heat flow regions is likely critically stressed, i.e., at full yield at all depths, as described by the envelope. But different conceptual frameworks offer opposite views for very cold lithosphere, either at full yield or far below yield. Here, we use simple numerical models to investigate stresses in end-member cold cratonic lithosphere (e.g., Canadian Craton) in comparison with end-member warm plate-boundary lithosphere (e.g., Canadian Cordillera). The two key elements of our modelling are (1) that lithospheric stress builds up elastically with horizontal tectonic loading not only in the elastic–frictional brittle regime but also in the viscoelastic ductile regime, and (2) that the stress level is limited by the available tectonic force. In a cratonic lithosphere, the limiting tectonic force is sustained by competent rock material over a large depth range, represented by the competent thickness Tc that exceeds 90 km. The lithosphere undergoes mostly elastic deformation at a stress level of a few tens of MPa. While weakly stressed strong lithosphere can still produce limited earthquakes at shallow depths due to structural and stress heterogeneity, the lithospheric stress under horizontal tectonic loading is theoretically predicted to be orders of magnitude lower than predicted by the Christmas-tree envelope. Stresses in a real lithosphere may substantially deviate from this theoretical level because of spatiotemporal variations in rheology and structure. For example, the stress memory of past loading history in cold lithosphere may or may not be erased by more recent tectonic stresses. Because much of previous scientific debates on lithospheric stress levels and comparison with seismicity were focused on topographically induced flexural stress, we also investigate the effect of vertical loading. We show that the effective elastic thickness Te derived from the flexural response is a reasonable proxy for Tc derived from horizontal tectonic loading; a very large Te such as 〉 80 km is generally associated with very low tectonic stress far below yield. However, the flexure-induced bending stress is not directly comparable with seismicity because it may either enhance or suppress seismogenic stress in the crust.