Towards a Quantitative Assessment of the Impact of Transient Mantle Rheology on Future Antarctic Ice-Sheet Stability

Mass transfer between the cryosphere and oceans leads to sea-surface height and topography changes whose timescales, amplitudes, and spatial patterns are controlled by mantle viscoelasticity. This ‘glacial isostatic adjustment’ (GIA) can slow or halt retreat of unstable marine-based ice sheets since ice loss induces gravitational sea-surface lowering and bedrock rebound, reducing water depths around ice-sheet margins and lowering their exposure to melting by warm ocean currents. Despite widespread recognition of this solid Earth–ice-sheet feedback, it has often been assumed that Earth’s mantle is too viscous for GIA to have a measurable impact on ice-sheet dynamics over the next few centuries, with many ice-sheet models used in state-of-the-art intercomparison projects assuming either a rigid bed or millennial viscoelastic bedrock deformation timescales. However, GPS bedrock displacement timeseries suggest very low mantle viscosities exist beneath vulnerable regions of the West Antarctic Ice Sheet (~10¹⁷–10¹⁹ Pa s), implying that bedrock elevations are responding to modern melting on annual-to-decadal timescales, i.e., fast enough to have significant impact on ice-sheet stability over the coming centuries. Interestingly, GPS-inferred viscosities obtained in the same regions, but from bedrock responses to longer-timescale (10² –10⁵-yr) deglacial signals, are at ~10–100 times larger. This result suggests the low effective viscosities obtained for modern signals reflect the operation of transient deformation mechanisms. If confirmed, this transience would have major ramifications for our understanding of future Antarctic ice-sheet stability, since it would introduce a negative feedback whereby mantle viscosities and bedrock uplift rates scale with ice mass loss rates, limiting the speed of subsequent grounding line retreat.

 

Here, we first test whether observed loading-timescale-dependence of GPS-inferred mantle viscosities can be explained using experimentally constrained parameterisations of transient rock deformation across seismic to convective timescales. This analysis is carried out by calibrating these thermomechanical parameterisations for individual seismic tomographic models using both geophysical and experimental observations. Importantly, by adopting a probabilistic inverse method we evaluate parametric uncertainties and propagate them into our estimates of timescale-dependent 3D mantle viscosity. We find that transient and steady-state viscosities predicted by our optimal parameterisations can simultaneously explain the short- and long-timescale GPS signals recorded across the Antarctic Peninsula. Next, we integrate this thermomechanical structure into 1D transient and Maxwell viscoelastic Earth models to quantify the impact of this more complex rheology on rates of Antarctic bedrock uplift and relative sea-level change on deglaciation timescales ranging from years to millenia. Our results show that transient mechanisms have measurable impacts on all submillenial deglaciation timescales but are particularly pronounced over decadal-to-centennial intervals, producing up to ~50% more bedrock uplift and up to ~70% higher maximum uplift rates than steady-state counterparts. We conclude by presenting a thermomechanically self-consistent framework for integrating our calibrated ‘full-spectrum’ rheological parameterisations into coupled GIA–ice-sheet simulations that account for observed transient and 3D viscosity variations. We will present early results from these simulations that will ultimately enable the potential stabilising impact of transient rheology on Antarctic ice-sheet evolution to be quantified under different climatic forcing scenarios, improving projections of future barystatic sea-level change.