Mantle dynamics and slab rheology constrained by numerical modeling, structural and source seismology

I use geodynamic models to increase our understanding of the fate of subducted lithosphere as it contributes and responds to surface plate motion and upper mantle flow, as well as its effect on surface processes, using various surface observables. In Chapter 2, I compare predicted stress tensors from mantle circulation models to centroid moment tensor solutions from seismic events in order to constrain to first order the strength of upper mantle slabs. I find that moderately strong slabs produce the best match to the observations and that the net rotation of the lithosphere with respect to the lower mantle is important in generating the global-scale asymmetry in intermediate depth deformation. Using the global constraints and mantle flow computations from Chapter 2, I focus on the western Mediterranean in Chapter 3, where earthquake patterns and seismic tomography are insufficient to precisely resolve the density structure beneath the Alboran Sea that is reflected in the complex surface tectonics. Here, I predict upper mantle anisotropy and SKS splitting resulting from the flow models for a range of suggested structures. I show that SKS splitting observations in the Alboran Sea region are best reproduced by mantle flow models that include a deeply extending, slab-like structure beneath the Alboran Sea, elongate along the Iberian margin from Granada to the Gibraltar arc, where it curves southward toward the High Atlas. I also find that the absolute reference frame choice when prescribing surface velocities is crucial due to the opposing surface flow orientation in this region between two end-member absolute plate motion models. The southwest-directed surface flow in the hot-spot reference frame (HS3), when combined with sublithospheric deflection in response to structure in the Alboran region, generates a north-south shear and NNW splitting orientations most similar to the patterns observed along Gibraltar, without invoking slab rollback. Along with seismicity, regional uplift patterns and magmatism have been used to infer the relationship between subducting slabs and surface processes. In Chapter 4, I develop three dimensional, thermochemical, dynamic subduction models guided by seismicity and geomorphology from the Banda arc to explore the fate of subducted slabs after continental collision. I confirm the importance of buoyancy contrasts in exerting a first order control on the generation of slab tears in the upper mantle. I find that rheology exerts a strong control on the timing of slab tearing while continental geomorphology may explain complex surface processes in regions of incipient continental collision. From my modeling, I infer that seismicity patterns beneath the Banda Arc represent a westward propagating slab tear in response to the collision of the buoyant Australian shelf, and an eastward propagating slab tear in response to the collision of the Scott Plateau to the west. Further utilization of surface observations along with continued advances in geodynamic modeling continue to bridge the gap between lithospheric and deeper mantle processes.