Upper mantle structure, intraplate volcanism and the unique anisotropy of the Antarctic Plate from waveform tomography

Abstract

Antarctica’s unique geological setting makes it an important location for understanding the Earth’s crust and upper mantle, and its connections to the past and future evolution of the Earth. The Antarctic continent has remained relatively stationary at the South Pole over the past 80 million years, and is covered by the largest ice sheet in the world. The ice sheet is a major potential source driving global sea level rise, and is impacted by processes occurring within solid earth that are modulated by the lithospheric and upper mantle parameters. The challenging conditions and remote location has historically limited the direct data sampling of Antarctica, consequently, researchers have been turning to indirect geophysical methods such as seismology to better understand the region. Over the past two decades, the quantity and quality of seismic stations deployed globally—both permanent and seasonal—have dramatically increased. The significant rise in the the number of temporary deployments have enhanced the station coverage of Antarctica, leading to an increasing number of seismic studies of Antarctica and surroundings. However, the seismic data coverage is still heterogeneous, and our understanding of the Antarctic lithosphere and man- tle is still limited. In this work, we present a new anisotropic Sv velocity model of the Antarctic Plate, AP2024, that includes the lithosphere and underlying mantle down to 660 km depth beneath both the continental and oceanic portions of the plate. To augment the limited seismic station coverage of Antarctica, we assemble very large regional and global data sets, comprising all publicly available broadband seismic data. The model is generated using 785,000 waveform-fitted seismograms from over 27,000 events and 8,700 stations, and constrained by both body and Rayleigh surface waves, ensuring the dense data sampling of the en- tire upper mantle depth range. We invert the waveforms of S-, multiple S- and surface waves using Automated Multimode Inversion (AMI), producing sets of linearly independent equations with uncorrelated uncertainties for each source- station path. These equations describe the path-averaged perturbations of P- and S-wave velocities along approximate sensitivity kernels. The equations are then combined into a large linear system where the 3-D distributions of P- and S-wave velocities and ‘2Ψ′ azimuthal anisotropy is solved for, to obtain a preliminary model. We then exploit the mutual consistency of data arising from the massive dataset to remove the least mutually consistent measurements. A further manual error identification process, aimed at identifying and removing probable artefacts due to instrumentation errors, completes the process. The tomographic inversion is global but focused on the Antarctic Plate, with the data sampling maximised in the Southern Hemisphere, and with the regularisation tuned for the region. The model is parameterised on a triangular grid with a target lateral inter-knot spacing of 125 km, and is validated via a series of resolution tests. The model is consistent with previous studies, and reveals that the upper mantle of the Antarctic continent exhibits a bimodal nature. A sharp boundary, along the Transantarctic Mountains divides the cratonic eastern from tectonic western Antarctica. The bimodality also extends to the oceanic part of the plate, with the older oceanic lithosphere beneath the southern Indian Ocean and the Weddell and Enderby basins showing higher shear velocities. The continental lithosphere in East Antarctica shows high velocity anomalies, with significant lateral heterogeneity, similar to those beneath stable cratons elsewhere around the world. Low-velocity anomalies underlie most of West Antarctica at 100 km depth, indicative of thin lithosphere and warm asthenosphere, including a prominent low velocity channel along the southern front of the West Antarctic Rift System. The seismic structure beneath the Antarctic Peninsula, confirms the post-subduction origins of volcanism along the peninsula. The two largest volcanic provinces in West Antarctica—Marie Byrd Land and Erebus—are underlain by prominent low velocity anomalies at asthenospheric depths, indicative of partial melting that feeds the volcanic processes. Combining our observations with whole mantle tomography models, we find strong evidence of deep-mantle origins for West Antarctica. The azimuthal seismic anisotropy reveals interesting fast-propagation directions oriented E-W in the older oceanic lithosphere, in a surprising circular pattern around the continent, in contrast to the younger oceanic lithosphere and elsewhere around the world. We identify complex anisotropy within the cratonic East Antarctica, indicative of the layering of deformation in the course of the craton formation. Strong azimuthal anisotropy associated with the Phoenix-Antarctic and South American-Sandwich subduction zones reveal clues of pressure-driven flow in the oceanic asthenosphere.