A Multivariate Approach for Mapping Lithospheric Domain Boundaries in East Antarctica

Abstract

Beneath the ice of East Antarctica lies a continent that is likely to be as geologically complex as its neighbors in Gondwana. An improved model of the heterogeneous lithosphere is required to progress research on Antarctica's tectonic evolution and support interdisciplinary studies of cryosphere and solid Earth interaction. We make use of multiple data sets, which were updated following the field campaigns and compilations of the International Polar Year of 2007/2008. Seismic tomography results, gravity anomalies, and surface elevation are used in a novel method, which combines spatial multivariate data to map possible boundaries as projected likelihood functions. Six multivariate combinations are tested and compared with sparse geological observations in East Antarctica. The resulting lithospheric domain boundaries contribute to our understanding of the deep continental structure. New boundaries are suggested in the interior, and models agree with likely surface expressions of crustal tectonic boundaries exposed along the coast.

Geothermal Heat Flux in Antarctica: Assessing Models and Observations by Bayesian Inversion

“…We suggest instead to combine this approach with a definition of preliminary tectonic provinces as for example by multivariate analysis (e.g., Stål et al, 2019). In the combined set-up data from the formerly adjacent continents can be used as proxies to adjust our prior parameter ranges, at least to study GHF in East Antarctica.…”

The Antarctic Crust and Upper Mantle: A Flexible 3D Model and Software Framework for Interdisciplinary Research

“…Reproducible models of the Antarctic crust and upper mantle are needed to progress interdisciplinary studies such as those relating to GIA, heat flow and topography. A better understanding of the solid Earth is achieved by combining multiple data sources (Begg et al, 2009;Pappa et al, 2019;Stål et al, 2019). Populating models with current data presents a challenge, especially given the present rate of new data releases that have the potential to improve existing results.…”

“…With the large uncertainties involved in Antarctic solid Earth research, probabilistic tools are essential to progress in the understanding of the Antarctic lithosphere. A productive way forward is to embrace the uncertainties and build probabilistic models (e.g., Stål et al, 2019). The computational framework that is presented here is well-suited to this task and provides an environment where data and associated uncertainties, probabilities and likelihoods can be processed.…”

A grid for multidimensional and multivariate spatial representation and data processing.

“…It is advantageous that cells of a grid model can be populated with such allied information, together with the dataset. agrid was initially developed for studies of the Antarctic lithosphere [33,34], and pre-processing of geophysical data for visualisation purposes [22], but with updates as presented here, it can be used in any discipline, geographical region, projection, dimensionality and any resolution. This initial release of the code is presented with tutorial notebooks that demonstrate its usage.…”

“…for dynamic updating. A few example-specific methods are also distributed together with the module[22,33,34].…”