T21D-2848
Fragility of Forearc Stresses as a Consequence of Extreme Weakness of Megathrust Faults
Tuesday, 15 December 2015
Poster Hall (Moscone South)
Kelin Wang1, Lonn Nathaniel Brown2, Jiangheng He3 and Tianhaozhe Sun2, (1)Geological Survey of Canada, Sidney, BC, Canada, (2)University of Victoria, Victoria, BC, Canada, (3)Pacific Geoscience Centre, Geological Survey of Canada, Sidney, BC, Canada
Abstract:
There is mounting evidence that subduction megathrusts are extremely weak. The weakness is based on a spatial and temporal average. Spatially, a seismogenic megathrust may host interspersed stronger and weaker patches due to variations in pore fluid pressure, gouge properties, and fault zone structure. In the 2011 M=9 Tohoku-oki earthquake, one strong patch underwent a local stress drop of several tens of MPa, although the rupture-zone average of the stress drop is less than 5 MPa on the basis of all the (> 20) published rupture models for this earthquake that we have examined. Temporally, megathrust strength fluctuates in earthquake cycles, punctuated by coseismic weakening or strengthening of different patches. Using finite element modeling, we demonstrate that the weakness of the megathrust leads to a fragile state of stress in the overlying forearc wedge, where compression due to plate coupling and tension due to gravity are in a subtle balance that can be tipped by small perturbations. Prior to the Tohoku-oki earthquake, the Japan Trench forearc was predominantly under margin-normal compression, a state that can be modeled using an effective friction coefficient of 0.032 for the megathrust. In a coseismic deformation model, an average stress drop of about 4 MPa on the megathrust changes the offshore forearc into tension. This is consistent with the observed stress reversal in this region as a result of the Tohoku-oki earthquake. The same level of coseismic stress drop would not cause the observed forearc stress reversal if the megathrust was assumed to have a higher strength such as 0.045. The state of stress in the offshore forearc is so fragile that large changes can be caused by other seemingly benign perturbing factors. For example, without the ocean water compressing the continental slope, much of the offshore forearc would no longer be in compression even if the megathrust strength were twice the value of 0.032. If the slope angle of the continental slope were assumed to be lower than the actual value by a couple of degrees, the offshore forearc would be in compression even if megathrust friction were as low as 0.015. The fragility of forearc stresses has profound implications on the interplay between megathrust earthquakes and long-term upper plate deformation that operate on very different timescales.