Inferring Ice Fabric Anisotropy Using Polarimetric Radar in Different Flow Regimes

Abstract:

Ice is strongly anisotropic, yet the majority of model projections for polar ice sheets use an isotropic rheology. Reasons for this shortcoming are manifold, one of them being the lack of spatially coherent data that can be used to develop and constrain an anisotropic flow law. Here we aim to apply polarimetric ice-penetrating radar data (ApRES) as a tool that can derive ice-fabric parameters independent of ice cores.

Ice anisotropy can be detected in radar data by interpreting phase and intensity differences caused by birefringence and anisotropic scattering. Building on previous research (the ‘polarimetric coherence method’ in particular), we designed a robust approach to infer the depth-variability of the ice-fabric orientation, ice-fabric strength, and ice-fabric variability. We do this using ApRES data in conjunction with a new multi-variate optimization scheme. We validate our inferences at two different ice-core deep drilling sites in East Antarctica, namely EPICA Dome-C (EDC) and EPICA Dronning Maud Land (EDML). Both ice cores are located in contrasting flow regimes (i.e. dome vs. flank flow at transient ice divide). Using additional ApRES surveys at EDC, we image the ice-fabric development in the transition zone from dome to flank.

We find strongly contrasting ice properties at the two ice-core sites. Most notably, ice appears more anisotropic and more depth-variable at the EDML site. In addition to mapping out the spatial variability in ice-fabric characteristics, we explore the possibility to fully reconstruct magnitudes and depth-dependence of all three Eigenvalues using characteristics from anisotropic scattering. If successful, this approach can be used to infer ice-fabric properties along radar transects and provide the required constraints to move towards validating anisotropic flow laws.