S U M M A R YTheoretical approaches to computing gravitationally self-consistent sea-level changes in consequence of ice growth and ablation are comprised of two parts. The first is a mapping between variations in global sea level and changes in ocean height (required to define the surface load), and the second is a method for computing global sea-level change arising from an arbitrary surface loading. In Mitrovica & Milne (2003) (Paper I) we described a new, generalized mapping between sea-level change and ocean height that takes exact account of the evolution of shorelines associated with both transgression and regression cycles and time-dependent marine-based ice margins. The theory is valid for any earth model. In this paper we extend our previous work in three ways. First, we derive an efficient, iterative numerical algorithm for solving the generalized sea-level equation. Secondly, we consider a special case of the new sea-level theory involving spherically symmetric earth models. Specifically, we combine our iterative numerical formulation with viscoelastic Love number theory to derive an extended pseudo-spectral algorithm for solving the new sea-level equation. This algorithm represents an extension of earlier methods developed for the fixed-shoreline case to precisely incorporate shoreline migration processes. Finally, using this special case, we quantitatively assess errors incurred in previous efforts to extend the traditional (fixed shoreline) sea-level equation of Farrell & Clark (1976) to treat time-dependent shorelines. We find that the approximations adopted by Johnston (1993) and Milne (1998) to treat transgression and regression at shorelines introduce negligible (∼1 per cent) error into predictions of post-glacial relative sea-level histories. In contrast, the errors associated with the Peltier (1994) sea-level equation are an order of magnitude larger, and comparable to the error incurred using the traditional sea-level theory. Furthermore, our numerical tests verify the high accuracy of the Milne (1998) approximation for treating the influence of grounded, marine-based ice.Predicting gravitationally self-consistent sea-level changes driven by the melting of ice sheets on a deformable earth is a complex undertaking. Under the assumption of an equilibrium, or hydrostatic, theory (e.g. Dahlen 1976), the sea-level redistribution will be governed by the gravitational field of the planet, since the sea surface is constrained to remain on an equipotential. However, this field is, in turn, perturbed by the redistribution of ocean mass, both through the direct attraction of the total (ice plus ocean) surface mass load and by solid earth deformations driven by this mass loading. As a further complexity, load-induced perturbations to the rotation vector of the planet will also have an impact on sea level, both through the re-orientation of the rotational potential and the deformation that results from this re-orientation. In addition, the ocean loading is defined by the local geometry of the shorelines,...

The 8.2 ka cooling event was an abrupt, widespread climate instability. There is general consensus that the episode was likely initiated by a catastrophic outfl ow of proglacial Lakes Agassiz and Ojibway through the Hudson Strait, with subsequent disruption of the Atlantic meridional overturning circulation. However, the total discharge and fl ux during the 8.2 ka event remain uncertain. We compute the sea-level signature, or "fi ngerprint," associated with the drainage of Lakes Agassiz and Ojibway, as well as the expected sea-level signal over the same time period due to glacial isostatic adjustment (GIA) in response to the Late Pleistocene deglaciation. Our analysis demonstrates that sites relatively close to the lakes, including the West and Gulf Coasts of the United States, have small signals due to the lake release and potentially large GIA signals, and thus they may not be optimal fi eld sites for constraining the outfl ow volume. Other sites, such as the east coast of South America and western Africa, have signifi cantly larger signals associated with the lake release and are thus better choices in this regard.

The importance of including lateral Earth structure in the analysis of Fennoscandian glacial isostatic adjustment (GIA) is investigated using a finite volume numerical formulation. Comparing output from radially‐varying 1‐D Earth models and models which account for the presence of plate boundaries, lateral variations in lithospheric thickness and viscosity heterogeneities in the upper and lower mantle, we find that perturbations to present‐day rates of surface deformation due to the inclusion of 3‐D Earth structure significantly exceed current observational uncertainties. Predicted residuals between 1‐D and 3‐D Earth models may be improved with the use of a 1‐D model which approximates the local depth‐dependent mean of the 3‐D model. However, the remaining misfit is still large enough to significantly bias inferences of Earth structure and ice history. We conclude that lateral variations at both global and regional scales must be accounted for when interpreting GPS observations from Fennoscandia.

[1] We investigate the potential impact of lateral variations in mantle viscosity and lithospheric thickness on predictions of present-day relative sea-level change due to glacial isostatic adjustment (GIA). We consider three viscoelastic Earth models. The first is a 1-D model with a lithospheric thickness of 120 km and upper and lower mantle viscosities of 5 Â 10 20 Pa s and 5 Â 10 21 Pa s, respectively. The second model includes global lithospheric thickness variations and lateral heterogeneities in upper mantle viscosity ranging over three orders of magnitude, while the third model includes lateral variations in lower mantle viscosity alone. We find that the impact of 3-D structure is significant. Indeed, the difference between the 3-D and 1-D model predictions at $300 sites with tide gauge records longer than 40 years duration is greater than 0.2 mm/yr and 0.5 mm/yr for 50% and 25% of the sites, respectively. The maximum difference exceeds several mm/yr. We conclude that efforts to decontaminate tide gauge records for ongoing GIA, to determine the rate and origin of global sea-level rise, should incorporate 3-D mantle structure into the GIA modelling. Citation: Kendall, R. A., K. Latychev, J. X. Mitrovica, J. E. Davis, and M. E. Tamisiea (2006), Decontaminating tide gauge records for the influence of glacial isostatic adjustment: The potential impact of 3-D Earth structure, Geophys.

ABSTRACT. When Rupert's Land and the North-Western Territory became a part of Canada as the Northwest Territories in 1870, the islands of James Bay were included within the new territorial boundaries. These same islands became a part of Nunavut in 1999, when the new territory was created from the eastern region of the Northwest Territories. Although the James Bay islands remain part of Nunavut, the western James Bay Cree assert that the western James Bay islands, including Akimiski Island, were part of the Cree traditional territory and that these islands have never been surrendered through treaty. This land-claim issue is further complicated by the fact that glacial isostatic adjustment (GIA) is occurring in the James Bay region and that the islands of James Bay may one day become part of mainland Ontario or Quebec. We used numerical models of the GIA process to predict how shorelines in James Bay will migrate over the next 1000 years as a result of post-glacial sea-level changes. These predictions, which were augmented by an additional contribution associated with sea-level rise due to global warming, were used to determine whether the islands in James Bay will ever become part of the mainland. The predictions for the islands are sensitive to the two primary inputs into the GIA predictions, namely the models for the geometry of the ancient Laurentide ice sheet and the viscoelastic structure adopted for the solid earth, as well as to the amplitude of the projected global warming signal. Nevertheless, it was found that many of the smaller and larger islands of James Bay will likely join the mainland of either Ontario or Quebec. For example, using a global warming scenario of 1.8 mm sea-level rise per year, a plausible range of GIA models suggests that the Strutton Islands and Cape Hope Islands will join mainland Quebec in ~400 years or more, while Akimiski Island will take at least ~700 years to join mainland Ontario. Using the same GIA models, but incorporating the upper boundary of global warming scenarios of 5.9 mm sea-level rise per year, the Strutton Islands and Cape Hope Islands are predicted to join mainland Quebec in ~600 years or more, and Akimiski Island is predicted not to join mainland Ontario. Since Akimiski Island is already being prospected for diamonds and the future ownership of emergent land remains an issue, these findings have great economic importance. Key words: sea-level change, subarctic Canada, post-glacial isostatic adjustment, global warming, islands of James Bay RÉSUMÉ. Quand la Terre de Rupert et le Territoire du Nord-Ouest ont joint les rangs du Canada sous le nom de Territoires du Nord-Ouest en 1870, les îles de la baie James ont été intégrées aux nouvelles frontières territoriales. Ces mêmes îles font maintenant partie du Nunavut depuis 1999, lorsque le nouveau territoire a été créé à partir de la région est des Territoires du Nord-Ouest. Bien que les îles de la baie James fassent toujours partie du Nunavut, les Cris de l'ouest de la baie James soutiennent que les îles du côté oue...