The effect of elastic layering on inversions of GPS data for coseismic slip and resulting stress changes:: Strike-slip earthquakes

We investigate the effect of depth-dependent elasticity on slip inversions and coseismic stress change estimates for large strike-slip earthquakes, using a series of hypothetical models and the 1999 Izmit, Turkey earthquake as examples. Slip inversions are performed using both semianalytical and finite-element solutions for surface displacements due to a shear dislocation in layered and uniform elastic Earth models. We find that incorporating realistic increases to shear modulus (mu) with depth in our inversions increases recovered centroid depth and seismic potency relative to uniform elastic half-space models. Recovered seismic moment is up to 40% greater for models incorporating depth-dependent mu than it is for uniform elastic half-space models. Incorporating depth-dependent mu also increases our estimate of the maximum slip depth for the Izmit, Turkey, earthquake (to at least 20 km). Our estimates of coseismic stress change in the upper crust do not change significantly when we incorporate depth-dependent elasticity in our inversions, as long as slip at depth is tuned (increased) to match surface displacements. Coseismic differential stresses in the lower crust increase by up to a factor of 3 in the near field, but further from the fault, stresses from layered and uniform elastic models are approximately equal. With increasing depth in the mantle, the ratio of modeled differential stresses for the layered and uniform elastic models approaches the ratio of mantle p values for these two models. We conclude that models of postseismic viscoelastic relaxation following large strike-slip earthquakes should incorporate depth-dependent elasticity, but that uniform elastic half-space models are adequate to calculate coseisimic Coulomb stresses in the upper crust for most triggering studies.