Postglacial Relative Sea Level Research Articles

A “broad‐shelf effect” upon postglacial relative sea level history

Recently published analyses of postglacial relative sea level (RSL) histories from coastal locations which are proximate to broad continental shelves that were significantly exposed land at Last Glacial Maximum, have demonstrated that the Holocene RSL variations at such sites are not adequately explained unless particular care is taken to treat the influence of shelf inundation. This is herein shown to require a local adjustment to the water load based upon a mass conservation constraint. Previously revealed misfits of theory to data from such regions therefore do not signify a requirement for lateral heterogeneity in model viscoelastic structure. The quality of the revision to the theory required to account for this “broad shelf effect” is illustrated through analysis of RSL histories from the east coast of the South American continent and from Northwestern Europe.

On the postglacial isostatic adjustment of the British Isles and the shallow viscoelastic structure of the Earth

Summary Observations of postglacial relative sea level (RSL) history at sites beyond the margins of the main accumulations of Wu¨rm–Wisconsin–Devensian ice may be invoked to test the accuracy of global models of the glacial isostatic adjustment process. The existing database of RSL time-series for the British Isles forms a continuous network of sites that sample the coasts of Scotland, England and Wales fairly uniformly, and which may be especially important in this regard. Not only are these 14C data for the Holocene epoch of very high quality in terms of our knowledge of the coastal environment and relationship to palaeo-sea level of each age measurement, but they are especially numerous and relate to an island that was itself partly glaciated in the north, essentially ice-free in the south, and located well outboard of the massive ice-sheet that covered Fennoscandia at the Last Glacial Maximum (LGM). Taken together, the time-series from 55 distinct locations allow us to perform a fairly stringent test of the recently formulated ICE-4G (VM2) global model of the isostatic adjustment process. Even without modification, this model has been shown to fit most of the time-series in the UK data set rather well. In order to satisfy very recently published a priori constraints on the maximum thickness of the Scottish ice-sheet, however, we find that the lithospheric thickness, the feature of the radial viscoelastic structure to which the rebound data from Scotland are most sensitive, must be somewhat reduced. Whereas the assumed thickness of the lithosphere in ICE-4G (VM2) was near to 120 km, best fits to the Scottish data require a lithospheric thickness near to 90 km. This reduced value for lithospheric thickness agrees exceedingly well with the latest estimates of this Earth property based on the inversion of oceanic geoid and topography data, as the asymptotic thickness characteristic of the oldest ocean floor. This would appear to be the best possible a priori estimate for the thickness of the cratonic lithospheres upon which the largest ice-sheets of the late Pleistocene are known to have rested. We also demonstrate explicitly that inferences of the shallow viscoelastic structure beneath the British Isles are strongly influenced by the details of the history of deglaciation at remote locations, at Antarctica in particular.

Postglacial sea level: energy method

The equation governing changes in sea level caused by the redistribution of ice and water masses on the earth's surface is rederived based on the least potential energy principle. This energy method deepens our understanding of the coupled ice–sea–Earth system, and lays the foundation of a global solution used in this study. A phenomenological parameterization of `realistic' viscosity models is proposed based on microphysical considerations. Continuously varying viscosity structures are determined using a few characteristic viscosity values (parameters) at seismically identified boundaries. By using a set of admissible `realistic' viscosity models and two models of deglaciation histories, we find that satisfactory convergence can be reached for a global solution for postglacial relative sea level (RSL) at about harmonic degree 50, and the convergence appears independent of ice model. This relatively lower tolerable truncation level is a consequence of global nature of the ice–sea–Earth system. We further examine the sensitivity of postglacial sea level to `realistic' viscosity structure and the lithospheric thickness combined. We find that variation of lithospheric thickness does not alter our previous conclusion (Fang, M., Hager, B.H., 1996. The sensitivity of post-glacial sea level to viscosity structure and ice-load history for realistically parameterized viscosity profiles. Geophys. Res. Lett. 23, 3787–3790) that there is a correlation of RSL sensitivities between ice history and viscosity structures, i.e., at sites less sensitive to the ice model, the resolving power for viscosity structure is also less. Furthermore, models having a thicker lithosphere tend to permit better resolution of lower mantle viscosity.

New insights obtained from joint inversions for the radial profile of mantle viscosity

We discuss recent inferences of mantle viscosity obtained via formal joint inversion of data associated with glacial isostatic adjustment (henceforth ‘GIA’) and mantle convection. The former includes a set of decay times obtained from post-glacial relative sea level histories at sites in previously glaciated regions, while the latter includes spherical harmonic coefficients (up to degree and order 8) of the present-day free-air gravity anomaly field. The analyses described here indicate that many previous inferences of an isoviscous mantle based on GIA data sets may have been significantly biased by a misinterpretation of the resolving power of the so-called ‘Haskell number’ (10 21 Pas). Furthermore, we find that both the GIA and convection data sets are well fit by a radial viscosity profile characterized by a large (∼ 2 orders of magnitude) increase from the base of the lithosphere to the core-mantle-boundary. This demonstration indicates that there is no compelling reason to invoke a transient mantle rheology. Final word in regard to these issues awaits an improvement in the quality of the GIA data base.

Terminating the 100 kyr ice age cycle

We report a simulation of the most recent 100,000‐year glaciationdéglaciation cycle of the late Pleistocene ice age, a simulation that delivers an ice sheet chronology that is in close accord with that inferred from the geological record. Our analyses are performed with a reduced model of the climate system that incorporates significant improvements to the representation of both climate forcing and mass balance response in a previously described theory based upon a coupled one‐level energy balance model (EBM) and vertically integrated ice sheet model (ISM). The theory fully incorporates the influences of orbital insolation forcing, glacial isostatic adjustment and variations in the atmospheric concentrations of greenhouse gases. It correctly predicts the main geographical regions of the northern hemisphere that were glaciated at last glacial maximum 21,000 years ago as well as the abrupt termination of the glacial epoch that occurred subsequently. The latter feature of the ice age cycle is obtained without the need to incorporate unconstrained and therefore controversial physical processes into the model, a limitation of all previous attempts to understand this global scale climate cycle. Our analyses suggest that the radiative impact on surface glaciation due to the changing atmospheric concentration of CO2 Is critical to the ability of the model to deliver a synthetic history of glaciation and déglaciation that is in accord with inferences based upon surface geological and geomorphological evidence. With the incorporation of this influence, model‐predicted ice thickness distributions at last glacial maximum (LGM) are very similar to those of the recently described ICE‐4G reconstruction that was based upon the inversion of postglacial relative sea level histories.

The influence of a finite glaciation phase on predictions of post-glacial isostatic adjustment

We predict relative sea level (RSL) variations, present day three-dimensional (3D) crustal deformation rates and baseline velocities, and a set of anomalies in the gravitational field of the planet, using a suite of models for the space-time geometry of the Pleistocene ice sheets. These models are distinguished on the basis of the details of the glaciation phase, and they include the cases of an infinite glaciation phase (i.e., an assumption of isostatic equilibrium at the time of the last glacial maximum — LGM), a steady uniform glaciation, and a glaciation in which the bulk of the accumulation is limited to the final stages of the growth phase. Comparison of the observational records of postglacial RSL change with predictions based on the assumption of isostatic equilibrium at LGM have commonly been used to infer both the space-time geometry of the final deglaciation event and the mantle viscosity. We conclude that both these applications may be influenced significantly by the incorporation of a finite glacial cycle. As an example, RSL predictions, which assume isostatic equilibrium at LGM, and those which incorporate a relatively rapid glaciation phase, can differ by a factor of 2 or more at some sites within previously glaciated regions when an Earth model characterized by a lower mantle viscosity of 10 22 Pa s is adopted. We have also found that differences in the adopted model for the glaciation event are capable of explaining entirely the discrepancies ( ∼ 0.2 mm/yr) between previously published predictions of present day tangential velocities in North America. These discrepancies are characterized by a dominant long-wavelength signal and, hence, they do not necessarily influence (strongly) predictions of baseline velocities. Variations in the glaciation model can also have an important effect on predictions of peak gravity anomalies over previously glaciated regions, and secular variations in the zonal harmonics of the Earth's geoid. In both cases the effect may be sufficient to alter significantly the signal from other geophysical processes which may need to be invoked to reconcile the observational constraints.

Ice Age Paleotopography

A gravitationally self-consistent theory of postglacial relative sea level change is used to infer the variation of surface ice and water cover since the Last Glacial Maximum (LGM). The results show that LGM ice volume was approximately 35 percent lower than suggested by the CLIMAP reconstruction and the maximum heights of the main Laurentian and Fennoscandian ice complexes are inferred to have been commensurately lower with respect to sea level. Use of these Ice Age boundary conditions in atmospheric general circulation models will yield climates that differ significantly from those previously inferred on the basis of the CLIMAP data set.

Paleoshorelines and prehistory: an investigation of method

Preface. Introduction. PART I: STABLE LAND. ISLANDS. Introduction. Lucayan Settlement Patterns and Recent Coastal Changes in the Bahamas. Paleoshorelines and the Prehistory of Barbuda, West Indies. MAINLAND. Introduction. A Geoarchaeological Methodology for Studying Prograding Coastal Sequences: Beach Ridge Geomorphology in Kotzebue Sound, Alaska. Paleoshorelines and the Sambaquis of Brazil. UNDERWATER. Introduction. Recent Archaeogeophysical Studies of Paleoshorelines of the Gulf of Mexico. Inundated Prehistoric Sites in Apalachee Bay, Florida and the Search for the Clovis Shoreline. PART II: UNSTABLE LAND. ISLANDS. Introduction. Modeling Paleoshorelines in Geologically Active Regions: Applications to the Shumagin Islands, Southwest Alaska. Paleoshorelines and Prehistoric Settlement on Simeonof and Chernabura Islands, Outer Shumagin Islands, Alaska. MAINLAND. Introduction. Postglacial Relative Sea Level History of the Labrador Coast and the Interpretation of the Archaeological Record. Space Shuttle Imagery of Recent Catastrophic Change Along the Arid Andean Coast. CONCLUSION. GLOSSARY. BIBLIOGRAPHY. INDEX.

Ice‐3G: A new global model of Late Pleistocene deglaciation based upon geophysical predictions of post‐glacial relative sea level change

A new high resolution global model of late Pleistocene deglaciation is inferred on the basis of geophysical predictions of postglacial relative sea level variations in which the ice‐ocean‐solid Earth interaction is treated in a gravitationally self‐consistent fashion. For the purpose of these analyses the radial viscoelastic structure of the planet is assumed known on the basis of previously published sensitivity tests on solutions of the forward problem. Only radiocarbon controlled relative sea level histories from sites that were actually ice covered (with one or two additions) are employed to constrain the model, leaving relative sea level (RSL) data from sites that were not ice covered to be employed to confirm its consistency. Results for these confirmatory analyses are reported elsewhere. Here the new deglaciation model, referred to as ICE‐3G, is compared to previous models derived by several independent means and tested against a number of additional observations other than sea level histories, including geologically controlled retreat isochrones, oxygen‐isotope data from deep‐sea sedimentary cores, and coral terrace elevations. The latter two observations strongly constrain the net sea level rise that has occurred since the onset of deglaciation and therefore the mass of ice that melted during the last glacial‐interglacial transition.

Une transgression marine holocène au Spitsberg nord-occidental (Svalbard) : une origine eustatique ou glacio-isostatique

In the Brøgger Peninsula area, northwest Spitsbergen, the postglacial glacio- isostatitc uplift has induced a rapid land emergence during the Early Holocene and relative sea level fell below its present position. This regression was followed by approximately 15 m of relative sea level rise during the second part of the Holocene. The postglacial relative sea level variations are explained in terms of glacio-isostatic adjustments. The migration of a peripheral forebulge, initially induced by ice-loading on eastern Svalbard, at the end of the Late Weichselian glaciation, is likely the cause of a transgression which followed the postglacial emergence in that area.

Lithospheric thickness, Antarctic deglaciation history, and ocean basin discretization effects in a global model of postglacial sea level change: A summary of some sources of nonuniqueness

The global model of postglacial relative sea level variations that has been developed over the past decade is employed to investigate the constraints that it may be invoked to place on the timing of the deglaciation of West Antarctica. The analyses presented here confirm the suggestion of P. Wu and W. R. Peltier (1983, Geophysical Journal of the Royal Astronomical Society 74, 377–450) that the model of this event employed in J. A. Clark and C. S. Lingle (1979 Quaternary Research 11, 279–298) may be simply modified to rectify the misfits between theory and observation that are otherwise obtained at Southern Hemisphere sites. A large number of Southern Hemisphere relative sea level data are shown to require that the retreat of Antarctic ice substantially lagged the retreat of Northern Hemisphere ice if the deglaciation of Antarctica was abrupt. The time of onset of Antarctic deglaciation is thereby shown to coincide with the time of most rapid Northern Hemisphere deglaciation. Sensitivity tests are performed which demonstrate that this result is relatively insensitive to the discretization employed to represent the ocean basins; the only exception to this general rule obtains at some coastal sites at which a trade-off is revealed between the delay of West Antarctic melting and the thickness of the lithosphere required to reconcile the observed local variations of relative sea level. At such sites, which are all in the far field of the ice sheets, some attention must be paid to the accuracy of the local finite element representation of the oceans and to the details of the Antarctic deglaciation history.

Erratum: Postglacial relative sea-level change, Port au Port area, west Newfoundland

We first report pollen and foraminifera analyses and radiocarbon dates from two cores taken from salt-marsh deposits bordering Port au Port Bay, southwestern Newfoundland. Results show that relative sea level (RSL) stood at 2.8 m below present higher high-water level (HHWL) at 2770 ± 300 years BP and at −1.8 m at 2365 ± 175 years BP at the core sites. They permit calculation of a rate of late Holocene RSL change from western Newfoundland. We then report other available dates bearing on the earlier RSL record of this area.A date of 5800 ± 200 years BP fixes the age of minimum RSL in Port au Port Bay at 11–14 m below present. A date of 9350 ± 120 years BP from St. George's provides a minimum age for the passage of sea level below present there. A date of 12 600 ± 140 years BP from Stephenville fixes a sea level at 29 m above present, whereas one of 13 600 ± 110 years BP from Abrahams Cove dates the marine limit at 44 m. These geographically restricted data closely constrain a curve of postglacial RSL change...

Postglacial relative sea-level change, Port au Port area, west Newfoundland

We first report pollen and foraminifera analyses and radiocarbon dates from two cores taken from salt-marsh deposits bordering Port au Port Bay, southwestern Newfoundland. Results show that relative sea level (RSL) stood at 2.8 m below present higher high-water level (HHWL) at 2770 ± 300 years BP and at −1.8 m at 2365 ± 175 years BP at the core sites. They permit calculation of a rate of late Holocene RSL change from western Newfoundland. We then report other available dates bearing on the earlier RSL record of this area.A date of 5800 ± 200 years BP fixes the age of minimum RSL in Port au Port Bay at 11–14 m below present. A date of 9350 ± 120 years BP from St. George's provides a minimum age for the passage of sea level below present there. A date of 12 600 ± 140 years BP from Stephenville fixes a sea level at 29 m above present, whereas one of 13 600 ± 110 years BP from Abrahams Cove dates the marine limit at 44 m. These geographically restricted data closely constrain a curve of postglacial RSL change in the Port au Port Bay – northern St. George's Bay area. The form of the curve supports a recent model predicting sea-level response to wastage of a limited late Wisconsinan ice load in the wider region.

The deglaciation of Atlantic Canada as reconstructed from the postglacial relative sea-level record

The post-Wisconsinan relative sea-level record from Atlantic Canada is used to reconstruct the morphology of late Wisconsinan age ice cover during its retreat from the Atlantic region. The proposed reconstruction has little or no grounded ice in the southern Gulf of St. Lawrence, an ice dome over the north shore of the St. Lawrence, and thin ice, often less than 1 km thick, over much of the rest of the area. A sensitivity analysis shows that the proposed reconstruction is not unique in its ability to account for the relative sea-level record but that the thickness of ice in any individual area of the reconstruction is unlikely to be in error by more than a factor of two. The exact position of the ice margin in some areas is not well constrained by the model; an example is in southeastern Newfoundland.The numerical model used to relate ice morphology to postglacial relative sea level assumes that the ice sheets are isostatically equilibrated at the glacial maximum and, therefore, that load changes associated with earlier ice-sheet growth may be ignored. This assumption is shown to be reasonable. The same rapid relaxation of the Earth that allows one to ignore the effects of glacial accumulation, however, prohibits one from recognizing the effects of large-scale ablation that may have occurred prior to the assumed glacial maximum. For this reason the proposed reconstruction may be representative of only a late stage in the ablation of much more extensive and thicker ice sheets.Surfaces of relative sea level are presented for Atlantic Canada at various times in the past. These surfaces coincide with observational data where such data exist and are felt to provide reasonable estimates of relative sea level at all other locations for at least the last 13 000 years.

A comparison of observed and theoretical postglacial relative sea level in Atlantic Canada

Two extreme models of late Wisconsinan ice cover in Atlantic Canada and the northeastern U.S.A. are shown to produce postglacial relative sea level curves that bracket existing field observations at six sites throughout the region. This suggests that the true late Wisconsinan ice distribution is probably intermediate to the two contrasting reconstructions proposed. Both ice models predict the existence of four relative sea level zones: an innermost zone closest to the centre of glaciation in which relative sea level falls continuously throughout postglacial time; an outermost zone in which it rises continuously; and two transitional zones in which it first falls and then rises in varying proportions according to the distance from the ice margin. The distinctive forms of the relative sea level curves are probably representative of each of the zones and are unlikely to be significantly perturbed even by large local ice readvances. They, therefore, establish patterns with which future field data are expected to conform. The form that the geological record of relative sea level change is likely to take within each zone is discussed and promising settings for the collection of new data are proposed. The common practice of separating relative sea level into an isostatic and a eustatic component is analysed and shown to be incorrect as usually applied. The practice is also shown to be unnecessary because the models discussed in this paper predict changes in relative sea level that can be compared directly with the observations.