The role of crustal models in the dynamics of the India-Eurasia collision zone

We investigate how different crustal models can affect the stress field, velocities and associated deformation in the India-Eurasia collision zone. We calculate deviatoric stresses, which act as deformation indicators, from topographic load distribution and crustal heterogeneities coupled with density driven mantle convection constrained by tomography models. We use three different crustal models, CRUST2.0, CRUST1.0 and LITHO1.0 and observe that these models have different crustal thickness and densities. As a result, gravitational potential energy (GPE) calculated based on these densities and crustal thicknesses differ between these models and so do the associated deviatoric stresses. For GPE only models, LITHO1.0 provides a better constraint on deformation as it yields the least misfit (both orientation and relative magnitude) with the surface observations of strain rates, lithospheric stress, plate motions and earthquake moment tensors. However, when the stresses from GPE are added to those associated with mantle tractions arising from density-driven mantle convection, the coupled models in all cases provide a better fit to surface observations. The N-S tensional stresses predicted by CRUST2.0 in this area get reduced significantly due to addition of large N-S compressional stresses predicted by the tomography models S40RTS and SAW642AN leading to an overall strike-slip regime. On the other hand, the hybrid models, SINGH_S40RTS and SINGH_SAW that are obtained by embedding a regional P-wave model, Singh et al., in global models of S40RTS and SAW642AN, predict much lower compression within this area. These hybrid models provide a better constraint on surface observations when coupled with CRUST1.0 in central Tibet, whereas the combined LITHO1.0 plus mantle traction model provides a better fit in some other areas, but with a degradation of fit in central Tibet.