Glacial Isostatic Adjustment with 3D Earth models in North America

The Canadian landmass of North America, which was covered by the large Laurentide Ice Sheet during the Last Glacial Maximum, has been a key area for Glacial Isostatic Adjustment (GIA) studies. Previous GIA studies have applied 1D models of Earth's interior viscoelastic structure in such analyses, however, seismic tomography, field geology and more recent studies reveal the potential importance of 3D models of this structure. Here, using the latest quality-controlled deglacial sea level database from Engelhart and Horton (2012, QSR), Engelhart et al (2015, QSR) and Vacchi et al (2018, QSR), Global Navigation Satellite System (GNSS) data and Gravity Recovery And Climate Experiment (GRACE) data in North America, to investigate the effects of 3D structure on GIA predictions in North America. We explore scaling factors in the upper mantle (B_UM) and lower mantle (B_LM) and the 1D background viscosity model (V_o) while considering both the ICE-6G_C and ICE-7G_NA glaciation/deglaciation models of Peltier et al (2015, JGR) and Roy and Peltier (2017, GJI).

The Normal Mode Method and Coupled Laplace-Finite Element Method both compute gravitationally self-consistent relative sea level histories with time dependent coastlines and rotational feedback. Overall,an ensemble of 3D GIA models are found that can fit the deglacial sea-level, GPS and GRACE data simultaneously. However, different regions (e.g. East Canadian coast, US Atlantic coast, Pacific coast) prefer different combinations of (V_o, B_UM, B_LM) to provide the best fits. To further simulate the local structures, more complex geological settings (e.g. low viscosity asthenosphere, subduction zone) are investigated and the results will be presented. Another issue, to what extent the 3D models can improve the fit when compared with 1D internal structure, will also be discussed.