The thermochemical structure of West Antarctica from multi-observable probabilistic inversion

The interaction between ice sheets and mantle dynamics is crucial to understanding the present-day topography in many regions (Antarctica, Patagonia, North America, Scandinavia) and recent ice mass losses on a large scale. A better knowledge of mantle rheology and the physical properties beneath these regions will improve our understanding of this interaction. To better characterize these processes, we investigate the present-day thermochemical structure (temperature and major-element composition) of the lithospheric and sub-lithospheric mantle. The thermal structure provides indirect information on variations in mantle viscosity, key parameter in glacial isostatic adjustment models (GIA). Recent geophysical studies in Antarctica show a relationship between mantle viscosity inferred from GIA and seismic velocity anomalies. Here we use a 3-D multi-observable probabilistic inversion method to retrieve estimates of the thermal and lithological structures (velocities and densities) beneath West Antarctica at a resolution of 1°x1°. The method is based on a probabilistic (Bayesian) formalism and jointly inverts Rayleigh wave dispersion data, bouguer gravity anomalies, satellite‐derived gravity gradients, geoid height, absolute elevation and surface heat flow. With the Markov chain Monte Carlo procedures applied here, we use highly optimized forward problem solvers to sample the parameter space and determine geological structure and feature with full characterization of their uncertainties. In this presentation, we will discuss the main results, interpretation in terms of mantle rheology, and its implication for GIA model in this region.