instrumental records extend back for decades to at most
about 100 years and measure only the response long after
the end of the last major deglaciation phase. They are,
therefore, likely to also include the Earth’s response to
any recent changes in the residual ice sheets.
The geological evidence for sea-level change is particularly important since it extends back in time, primarily
to the time of the Last Glacial Maximum (LGM) but with
scattered evidence for earlier periods. The observation is
of the change in sea-level relative to the land surface and
is, therefore, a relative measurement and includes both
the crustal deformation associated with the changes in surface loads and the changes in the shape of gravitational
equipotential surfaces. A raised shoreline could mean
either that there has been land uplift, that there has been
a reduction in ocean volume, or that there has been a redistribution of water within the ocean basin. The glaciohydro-isostatic effect is a combination of all three. Observations of sea-level therefore contain information on the
Earth’s rheology as well as on the ice history, and on any
other processes, principally of tectonic and geodynamic
origin.
Models for predicting the Earth’s response to glacial
cycles and the concomitant water loading have been well
developed over recent decades with a focus on linear viscoelastic models for the mantle rheology and in particular
on a Maxwell rheology with realistic radial elastic and
density stratification (Peltier, 1974; Cathles, 1975). The
rheological parameters are usually expressed as depthdependent effective viscosities and as an effective elastic
thickness of the lithosphere that provide a formulation that
appear to work remarkably well in explaining much of the
observational evidence even though the relations between
these parameters and laboratory evidence for mantle
behavior under stress remains unclear. A critical element
in these models is the treatment of the water loading on a
deforming Earth such as to be gravitationally consistent
and conserve ice-water mass in which the water load itself
is a function of the deformation of the Earth and of the
redistribution of the water (Farrell and Clark, 1976;
Mitrovica and Milne, 2003).
The pattern of global sea-level change
The pattern of sea-level induced by the growth and decay
of ice sheets is conveniently divided into three zones:
areas within or near the margins of the former ice sheets
(near-field), areas far from these ice sheets (far-field),
and an intermediate-field. During deglaciation, the dominant contribution to sea-level change in the near-field is
the crustal rebound as mantle materials flow back beneath
once ice-loaded continents. This is the glacio-isostatic
component. For the largest ice sheets, perhaps up to
4,000 m thick and 1,000 km or more radius, the total
rebound since the LGM approaches the local isostatic
limit of 800–1,000 m but only the rebound that occurred
after the region became ice-free is preserved in the record.
Rates of crustal uplift of up to 1 cm/year still occur in areas
such as the Hudson Bay (Canada) or the Gulf of Bothnia
(Fennoscandia). The other not insignificant contributions
are from changes in the gravitational attraction in the
Earth-ocean-ice system, the change in the ocean volume
and the deformation of the Earth caused by the changing
water load. The net result is that where there once were
large ice sheets, sea-level has been falling and continues
to fall long after the last of the ice has vanished. Toward
the edge of the ice sheet, the crustal rebound component
is reduced, and the other contributions begin to dominate
the signal such that the relative sea-level signal may oscillate in time between rising and falling.
In far-field regions, it is the change in the ocean volume
and the deformation of the Earth caused by the water loads
that are the dominant contributors. With increasing water,
the ocean floor subsides, approaching the local isostatic
limit in mid-oceans of about 25 % of the change in the
ocean depth. Along continental margins, because of the
elastic properties of the lithosphere and the viscous flow
in the mantle, the subsidence will be about half this. This
is the hydro-isostatic component. Since the LGM, ~50 Â
10
6 km
3 of ice has been added into the oceans raising globally averaged sea-level by ~135 m but by less along the
continental far-field margins because of the isostatic
effects. At the end of glaciation, the ocean volume remains
constant but the isostatic effects continue and sea levels
fall, producing small-amplitude (up to ~3 m) highstands
peaking at around 6,000 years ago that will be strongly
dependent on nearby coastline geometry.
In the intermediate-field, the deformational, gravitational, and ocean volume contributions are often of similar
magnitudes but opposing signs, such that the sea-level signal can be spatially complex. But the dominant feature of
this response is the combination of the subsidence of a
broad zone of crustal uplift that developed around the ice
sheet during the loading phase and the change in the gravitational attraction between ice and ocean. Together, these
factors result in sea levels outside of the ice margins rising
throughout the post-glacial phase at rates that are a function of the ice sheet dimensions and of the distance from
the ice margin. This pattern will extend out to several
thousand kilometers beyond the center of the largest ice
sheets.
Viewed globally, the glacio-hydro-isostatic response
can therefore be expected to present a complex spatial
and temporal variability, and this is indeed observed
(Figure 1). For Hudson Bay and the Gulf of Bothnia, near
the centers of the North American and Fennoscandian ice
sheets respectively, sea-level has been falling over past
millennia, and observations of the timing and rates of this
fall provide constraints on both the mantle response function and local ice thickness. In the northwestern Norway
example of Andøya in Figure 1, sea levels since the
respective areas became ice-free have fallen and risen in
time due to competing contributions from the crustal
rebound and ocean volume increase, and observations
from these locations provide insight into the local details
of the ice sheet. The Andøya record, for example,
GLACIO(HYDRO)-ISOSTATIC ADJUSTMENT
295
about 100 years and measure only the response long after
the end of the last major deglaciation phase. They are,
therefore, likely to also include the Earth’s response to
any recent changes in the residual ice sheets.
The geological evidence for sea-level change is particularly important since it extends back in time, primarily
to the time of the Last Glacial Maximum (LGM) but with
scattered evidence for earlier periods. The observation is
of the change in sea-level relative to the land surface and
is, therefore, a relative measurement and includes both
the crustal deformation associated with the changes in surface loads and the changes in the shape of gravitational
equipotential surfaces. A raised shoreline could mean
either that there has been land uplift, that there has been
a reduction in ocean volume, or that there has been a redistribution of water within the ocean basin. The glaciohydro-isostatic effect is a combination of all three. Observations of sea-level therefore contain information on the
Earth’s rheology as well as on the ice history, and on any
other processes, principally of tectonic and geodynamic
origin.
Models for predicting the Earth’s response to glacial
cycles and the concomitant water loading have been well
developed over recent decades with a focus on linear viscoelastic models for the mantle rheology and in particular
on a Maxwell rheology with realistic radial elastic and
density stratification (Peltier, 1974; Cathles, 1975). The
rheological parameters are usually expressed as depthdependent effective viscosities and as an effective elastic
thickness of the lithosphere that provide a formulation that
appear to work remarkably well in explaining much of the
observational evidence even though the relations between
these parameters and laboratory evidence for mantle
behavior under stress remains unclear. A critical element
in these models is the treatment of the water loading on a
deforming Earth such as to be gravitationally consistent
and conserve ice-water mass in which the water load itself
is a function of the deformation of the Earth and of the
redistribution of the water (Farrell and Clark, 1976;
Mitrovica and Milne, 2003).
The pattern of global sea-level change
The pattern of sea-level induced by the growth and decay
of ice sheets is conveniently divided into three zones:
areas within or near the margins of the former ice sheets
(near-field), areas far from these ice sheets (far-field),
and an intermediate-field. During deglaciation, the dominant contribution to sea-level change in the near-field is
the crustal rebound as mantle materials flow back beneath
once ice-loaded continents. This is the glacio-isostatic
component. For the largest ice sheets, perhaps up to
4,000 m thick and 1,000 km or more radius, the total
rebound since the LGM approaches the local isostatic
limit of 800–1,000 m but only the rebound that occurred
after the region became ice-free is preserved in the record.
Rates of crustal uplift of up to 1 cm/year still occur in areas
such as the Hudson Bay (Canada) or the Gulf of Bothnia
(Fennoscandia). The other not insignificant contributions
are from changes in the gravitational attraction in the
Earth-ocean-ice system, the change in the ocean volume
and the deformation of the Earth caused by the changing
water load. The net result is that where there once were
large ice sheets, sea-level has been falling and continues
to fall long after the last of the ice has vanished. Toward
the edge of the ice sheet, the crustal rebound component
is reduced, and the other contributions begin to dominate
the signal such that the relative sea-level signal may oscillate in time between rising and falling.
In far-field regions, it is the change in the ocean volume
and the deformation of the Earth caused by the water loads
that are the dominant contributors. With increasing water,
the ocean floor subsides, approaching the local isostatic
limit in mid-oceans of about 25 % of the change in the
ocean depth. Along continental margins, because of the
elastic properties of the lithosphere and the viscous flow
in the mantle, the subsidence will be about half this. This
is the hydro-isostatic component. Since the LGM, ~50 Â
10
6 km
3 of ice has been added into the oceans raising globally averaged sea-level by ~135 m but by less along the
continental far-field margins because of the isostatic
effects. At the end of glaciation, the ocean volume remains
constant but the isostatic effects continue and sea levels
fall, producing small-amplitude (up to ~3 m) highstands
peaking at around 6,000 years ago that will be strongly
dependent on nearby coastline geometry.
In the intermediate-field, the deformational, gravitational, and ocean volume contributions are often of similar
magnitudes but opposing signs, such that the sea-level signal can be spatially complex. But the dominant feature of
this response is the combination of the subsidence of a
broad zone of crustal uplift that developed around the ice
sheet during the loading phase and the change in the gravitational attraction between ice and ocean. Together, these
factors result in sea levels outside of the ice margins rising
throughout the post-glacial phase at rates that are a function of the ice sheet dimensions and of the distance from
the ice margin. This pattern will extend out to several
thousand kilometers beyond the center of the largest ice
sheets.
Viewed globally, the glacio-hydro-isostatic response
can therefore be expected to present a complex spatial
and temporal variability, and this is indeed observed
(Figure 1). For Hudson Bay and the Gulf of Bothnia, near
the centers of the North American and Fennoscandian ice
sheets respectively, sea-level has been falling over past
millennia, and observations of the timing and rates of this
fall provide constraints on both the mantle response function and local ice thickness. In the northwestern Norway
example of Andøya in Figure 1, sea levels since the
respective areas became ice-free have fallen and risen in
time due to competing contributions from the crustal
rebound and ocean volume increase, and observations
from these locations provide insight into the local details
of the ice sheet. The Andøya record, for example,
GLACIO(HYDRO)-ISOSTATIC ADJUSTMENT
295