Antarctic Glacial Isostatic Adjustment from an inversion of satellite and in-situ observations

Glacio-isostatic adjustment (GIA) has, until recently, been estimated using forward models that attempt to determine how the mantle and lithosphere respond to changes in ice loading through time. These models require knowledge of the Earth structure, including mantle viscosity, and ice load history, both of which have large uncertainties for Antarctica. Furthermore, 3-D Earth models are required to adequately accommodate the substantial variations in crustal thickness between West and East Antarctica. An alternative approach is to use observations of crustal motion from GPS, combined with mass trends from GRACE to invert for GIA. However, this is an undetermined problem. Here, we present a novel solution to this problem using the latest methods in statistical modelling of spatio-temporal processes. We use Bayesian hierarchical modelling and employ stochastic partial differential equations to allow us to solve, simultaneously, for ice mass trends and GIA. Here, we focus on the GIA solution derived from a combination of ICESat, ENVISAT, GRACE, InSAR, GPS and regional climate model output data for the period 2003-2009, assuming that GIA is time-invariant over this time frame. We compare our results with forward models and other inverse approaches. Our rates differ substantially regionally compared with the latest forward model solutions. For the Pine Island basin, for example, we obtain a rate close to 1 mm/yr as opposed to over 4 mm/yr for the models IJ05v2 and W12a. Over East Antarctica our rates are generally higher than forward models and we obtain positive values over a large area of Dronning Maud Land, where subsidence is predicted from forward modelling. On the other hand, we obtain negative trends over a large portion of Wilkes Land which is in good agreement with the results from IJ05v2, W12a and AGE-1. Compared to IJ05v2, our maximum uplifts are shifted towards the Ross and Ronne Ice Shelves, corresponding to about 4 and 6 mm/yr respectively. Over the Antarctic Peninsula we predict higher GIA uplifts than forward models, rating between 3 and 6 mm/yr over the entire region.