Slab window geodynamics: towards an integrated understanding of upper mantle dynamics and observations

Abstract

The generation of a slab window impacts the spatio-temporal evolution of subduction zones and promote complex mantle flow pattern where slabs once descended. The origin of slab windows is attributed to processes such as mid-ocean ridge subduction, slab tearing and/or break-off. The interaction between mid-ocean ridges and trenches is a common process affecting the geodynamic history of the margins around the Pacific, at least, during the Cenozoic and generated several modern slab windows. These intriguing features have notable effects on the upper mantle where temperature anomalies develop due to the asthenospheric upwelling and complex toroidal flow patterns through and around slab windows. There are profound effects on the overriding plate for the surface heat flow, geochemistry and spatial distribution of magmatic activity, seismicity and topographic relief. However, these manifestations evolve through space and time depending on the ridge axis-trench geometry, inducing the continuous slab window opening during its subduction.In this contribution, we derived a simplified expression for the slab window angle and then conducted 3D geodynamic modeling to link slab windows dynamics with geochemical and geophysical observables. The numerical models were conducted with fixed geometries in steady-state (using finite elements), compared with time-dependent solutions (using the I3ELVIS code) and then compared with observations from modern slab windows along the eastern Pacific. The analytical solution for the plan projection of the slab window depends on three parameters: the ratio between the half-spreading rate to the velocity of the overriding plate, the subduction angle and the obliquity of the ridge axis respect to the trench.Fast spreading or slow plate convergence promotes a wide (> 90°) slab window while slow spreading or fast convergence narrows this gap (< 90°). The slab dip and ridge obliquity have a second order control on the plan projection of the slab window but affect the existence of a steady-state solution. The implementation of this geometry into 3D steady-state models was used to generate a novel methodology to estimate mantle/melt upwelling and temperature anomalies in the upper mantle for a wide range of tectonic settings. Preliminary results on 3D time-dependent models reproduce a self-consistent opening of the slab window by only imposing spreading at the mid-ocean ridge and a subduction velocity with respect to the overriding plate. The ratio and absolute magnitude of these velocities controls the timing of the opening as well as the lateral and depth extent of the subducting plates. This timing also influences the development of upwelling and toroidal flow patterns around the slab edges. Finally, observations in modern slab windows along the eastern Pacific are consistent with the temperature and velocity field of the models. Variations in temperatures in the upper mantle are consistent with mantle shear wave speeds anomalies, while the flow field is correlated with the azimuthal anisotropy. In terms of magmatism, variables degrees of melting are consistent with the generation of tholeiitic to alkaline magmas in backarc areas.