Practical implications of the geometrical sensitivity of elastic dislocation models for field geologic surveys

Geodetic observations of both coseismic and interseismic surface deformation fields in the vicinity of subduction zones are frequently interpreted using simple elastic dislocation models (EDMs). The geometry of the megathrust interface used in these models has a first order effect on their predicted surface deformation. Here, we systematically explore the sensitivity of the surface velocity field predicted by EDMs both early and late in the seismic cycle, to parameterizations of megathrust interface geometry, effective subducting plate thickness, and gradual transitions in apparent plate coupling. We focus on how these parameterizations affect the hingeline - the location where vertical velocities switch from subsidence to uplift - as well as the location of the peak uplift rates. We find that these surface observables are much less sensitive to uncertainties in dip at the downdip end of the seismogenic zone for realistic curved faults in comparison to planar faults. For realistic megathrust geometries (planar or curved) having gradual transitions in apparent plate coupling, we find that the extent of locking is best approximated at the surface by the location of peak uplift rates. Therefore, the common notion - based on shallow-dipping planar faults - that the hingeline is located directly above the maximum depth extent of the locked plate interface is generally incorrect. Using the hingeline as the basis for coupling may lead to a significant underestimation of seismic hazard early in the cycle, as well as during the interseismic period. This analysis also demonstrates the importance of considering both vertical and horizontal velocities for determining seismic source extents, as well as interseismic coupling, on the megathrust. The tradeoffs presented here between the geometry of the megathrust and fault coupling along its surface can assist in the planning of campaign-GPS or field geologic surveys, and help improve seismic hazard estimates in active subduction zones. (C) 2012 Elsevier B.V. All rights reserved.