Data-driven tectonic regionalization of Antarctica: appreciate the similarity

Antarctica is the continent with the least constrained lithospheric architecture. Geophysical and geological data are sparse, and often associated with larger uncertainties than elsewhere. Such limitations have challenged the robustness of global tectonic regionalization models and schematic continental scale geological maps in Antarctica. Some published crustal models simply exclude high latitudes.

We collect an ensemble of robust datasets that covers the Antarctic continent. We match those Antarctic observables with global compilations to detect the most similar geological and tectonic setting. Most datasets are derived from satellite potential field surveys, seismic surface wave tomography and geological observations. We aim to take advantage of all and any existing data for Antarctica with a reasonable coverage and quality, and with a global equivalent reference observable.

The degree of similarity between Antarctic locations and global reference locations is used to generate a similarity rating map. We use this map to extract the suggested segment type classes from a number of published global tectonic regionalization models. The uncertainty of classification is expressed as information entropy for each location in Antarctica.

We present a number of tectonic regionalization models for Antarctica that can replace or extrapolate the global counterpart for high latitudes. Reference global models used include legacy studies, seismic segmentation, and lithological maps.