THE UTILITY OF 2D ELASTIC SCREW AND EDGE DISLOCATIONS FOR MODELING CRUSTAL DEFORMATION ASSOCIATED WITH SHEAR ZONES

Elastic edge and screw dislocations are some of the most prolific models in the active tectonics literature for describing present-day accumulation of ephemeral strain within Earth’s crust. In tectonic regimes where faulting is the prevailing mode of deformation, edge and screw dislocations may be used to model the interseismic portion of the seismic cycle for a single fault or for a network of faults. In this work, we focus on the application of screw and edge dislocations to describing interseismic strain accumulation on a single locked or partially locked fault in 2D. In this case, the dislocation line which is the locus of deformation, is typically said to be the location at depth where the frictional properties of the fault change from stick-slip (velocity-weakening) behavior above to steady sliding (velocity-strengthening) below. Geodetic studies typically infer the location of this line to be between 5 km to 20 km depth. However, it is well documented in the literature that at mid-crustal depths shear is likely accommodated across broad zones involving cataclastic flow, crystal plastic deformation, and/or other forms of ductile shear. These shear zones may span several kilometers in width, and are distinct from the discrete fault surface represented by an edge or screw dislocation. In this work, we present a method for combining the deformation from a large number of dislocations to approximate shear across a finite width ductile shear zone at depth. We develop a rigorous quantitative method for comparing the deformation predicted using a model for a finite width shear zone with the deformation predicted using the traditional infinitesimally thin edge or screw dislocation. We demonstrate that to within the typical precision of geodetic measurements, the difference in predicted deformation at the surface is indistinguishable for the finite- and infinitesimal-width models for strike-slip, reverse and normal faulting modes. Additionally, we demonstrate that regardless of whether the shear zone or discrete fault formulation of the model is chosen, the salient features of the model are identical and therefore the single fault model, which has fewer parameters, is preferable in most interseismic applications.