Earth’s Wandering Rotation Axis as a Diagnostic for Global Mantle Convection Models

Abstract

Dynamic topography is the change in topography that arises from viscous flow within the Earth’s mantle. As such, dynamic topography is sensitive to past mantle flow states. Making predictions of dynamic topography through time often relies on complex mantle convection models. To better constrain mantle convection models, we compare their implied True Polar Wander (TPW) paths for a range of model parameters. TPW is the re-orientation of a planetary solid body with respect to its rotation axis and may be produced by large scale mass redistributions on the Earth’s surface or within the mantle that perturb the Earth’s moment of inertia.Here we compare TPW histories estimated from two global plate tectonic reconstructions that were assimilated into the TERRA mantle convection code: (1) the widely-used Earthbyte global plate model (‘corrected R’ Matthews et al., 2016); and (2) TOMOPAC-22, a newly developed global plate tectonic model of the circum-Pacific using structurally-restored slabs from mantle seismic tomography (Wu et al., 2022). The time series of geodynamically-modeled mantle states are used to calculate synthetic TPW paths from perturbations in components of Earth’s moment of inertia from mass redistribution within the mantle; multiple (>10) viscosity-depth profiles were considered. We test these modeled TPW paths by comparing them against published paleomagnetic observations (Torsvik et al., 2012; Besse and Courtillot, 2002). Predicted TPW for plate Model 1 ranges widely (~90°) in azimuth from 120°W to 59°E with no consistent pattern across viscosity profiles. TPW rates reach maximums of 1.1°/Myr with excursions of ~25°. In contrast, predicted paths for Model 2 cluster within a smaller ~30° azimuthal range centered around ~29°E irrespective of the viscosity profile.  Predicted maximum rates were up to ~2°/Myr with excursions of up to 30°. Temporally, predicted paths for Model 2 drift toward northern Russia and then veer towards Greenland. Depending on the viscosity profile used some predicted TPW paths undergo stillstands from ~80 to ~30 Ma.  Ultimately, most model scenarios show longitudinal misfits up to 60° with observed paleomagnetic data; modeled TPW rates were within observed and theoretical ‘speed limits'. We discuss similarities and differences between our preliminary TPW history results and paleomagnetic observations, with a goal of developing an effective TPW test for constraining geodynamic parameters, plate tectonic reconstructions, and dynamic topography through time.