since 1992. Moreover, as the ice shelves exert a buttressing
effect for the upstream grounded ice, their disintegration can
cause the destabilization of a large part of the ice sheet.
There are different methods to measure the amount of ice
stored in an ice sheet:
(a) The measurement of inflow (snow accumulation) and
outflow (ablation, discharges to the ocean). Snow
accumulation is often estimated from annual layers in
ice cores and interpolated between different drilling
sites. The use of high resolution atmospheric models is
also becoming more common. Discharges of ice to the
ocean are estimated from seismic or radar measurements of the ice thickness and from measurements of
ice flow velocity. Ablation is usually determined from
ice models forced with atmospheric reanalyses, climatology or outputs from global climate models calibrated
with surface observations. The loss of mass beneath the
ice shelves remains very difficult to quantify. In general,
inflows and outflows cannot be estimated with a margin
error of less than 5%, which implies uncertainties of 40
and 140 Gt/yr on the estimate of the mass balance of
Greenland and Antarctica respectively.
(b) Remote-sensing techniques. These include altimetry
measurements from radio-echo sounding, interferometry and gravimetry. Interferometry provides information
on ice flow velocities. The altimetric measurements
provide information on the spatial and temporal variations of the topography, which makes it possible to
trace the ice volume variations, after correcting for the
altitude of the bedrock and for variations in the thickness and density of the firn. These measurements are
strongly dependent on the nature of the terrain (flat,
sloping or hilly surfaces) and the snow surface conditions (density, viscosity, etc.) and this may, in some
cases, make interpretation difficult. Finally, satellite
measurements of the gravity field provide, for the first
time, estimates of the ice mass changes. However, there
are still significant uncertainties, especially regarding
the cause of the change in mass, as this could come
from the ice sheet itself, the air column, the evolution of
the subglacial bedrock or even from masses near the ice
sheet (i.e. ocean mass, masses of water or snow contained on nearby continents). These different effects are
evaluated and then corrected, but a significant uncertainty remains, mainly on how to correct for the altitude
of the underlying Antarctic bedrock. Other sources of
uncertainty relating to satellite measurements come
from the fact that they do not provide complete coverage of the ice sheets. Moreover, in Greenland, for
example, remote sensing techniques may underestimate
the extent of the ablation area and the increased rainfall
over some parts of the ice sheet during rainfall events.
Past published estimates of Greenland and Antarctic ice
sheets obtained with these methods have often diverged.
This was due to the way in which the different sources of
uncertainty were estimated and to the fact that the different
measurements did not cover the same regions and the same
time periods. Within the IMBIE framework (Ice-sheet Mass
Balance Intercomparison Exercise), scientists made a huge
effort to combine, over common survey periods and common
regions, various observations from satellite geodetic techniques (altimetry, interferometry, radio-echo sounding and
gravimetry) with simulated surface mass balance estimates
inferred from regional atmospheric models. This allowed for
a reconciliation of the apparent disparities between the different methods and for a consistent picture of ice-sheet mass
balance to be created (Shepherd et al. 2012).
For Greenland, this compilation effort confirmed with a
high confidence level that the ice sheet is continuously losing
mass and that this process now affects all sectors of the ice
sheet (Fig. 31.6a–c). However, after a record mass loss in
summer 2012, Greenland has seen a slight decrease in the
short-term mass loss trend. The mass loss is partitioned
between surface melting and dynamic ice discharges. Shepherd et al. (2012) estimate that the ice mass loss is
about −142 ± 49 Gt/yr over the IMBIE time period (1992–
2011) with an acceleration of the mass loss rate as illustrated
by the comparison between the estimations made for 1992–
2000 (–51 ± 65 Gt/yr) and 2005–2010 (–263 ± 30 Gt/yr).
Using gravimetry observations, Velicogna et al. (2014) provide a more recent estimate of the Greenland ice sheet mass
loss for the 2003–2013 decade of 280 ± 58 Gt/yr with an
acceleration of the loss rate bringing it to 25.4 ± 58 Gt/yr
2 .
The case of the Antarctic ice sheet (Fig. 31.6b) is a bit
different since recent observations have shown that mass loss
was mainly driven by dynamic ice discharge resulting from
enhanced ice flow of marine-terminating glaciers. The main
region experiencing mass loss is the West Antarctic ice sheet
(WAIS), especially in the Amundsen/Bellingshausen Sea
sectors (e.g. Thwaites and Pine Island glaciers) and, to a
lesser extent, in the Antarctic Peninsula. According to a
recent update of the IMBIE estimates (IMBIE team, 2018),
the mass loss from the Amundsen and Bellingshausen Sea
sectors increased from 53 ± 29 Gt/yr to 159 ± 26 Gt/yr
over the 1992–2017 period, and from 7 ± 13 Gt/yr to
33 ± 16 Gt/yr in the Antarctic Peninsula. It has long been
considered that the East Antarctic ice sheet (EAIS) was
gaining mass due to enhanced precipitation, despite no firm
31 From the Climates of the Past to the Climates …
455
effect for the upstream grounded ice, their disintegration can
cause the destabilization of a large part of the ice sheet.
There are different methods to measure the amount of ice
stored in an ice sheet:
(a) The measurement of inflow (snow accumulation) and
outflow (ablation, discharges to the ocean). Snow
accumulation is often estimated from annual layers in
ice cores and interpolated between different drilling
sites. The use of high resolution atmospheric models is
also becoming more common. Discharges of ice to the
ocean are estimated from seismic or radar measurements of the ice thickness and from measurements of
ice flow velocity. Ablation is usually determined from
ice models forced with atmospheric reanalyses, climatology or outputs from global climate models calibrated
with surface observations. The loss of mass beneath the
ice shelves remains very difficult to quantify. In general,
inflows and outflows cannot be estimated with a margin
error of less than 5%, which implies uncertainties of 40
and 140 Gt/yr on the estimate of the mass balance of
Greenland and Antarctica respectively.
(b) Remote-sensing techniques. These include altimetry
measurements from radio-echo sounding, interferometry and gravimetry. Interferometry provides information
on ice flow velocities. The altimetric measurements
provide information on the spatial and temporal variations of the topography, which makes it possible to
trace the ice volume variations, after correcting for the
altitude of the bedrock and for variations in the thickness and density of the firn. These measurements are
strongly dependent on the nature of the terrain (flat,
sloping or hilly surfaces) and the snow surface conditions (density, viscosity, etc.) and this may, in some
cases, make interpretation difficult. Finally, satellite
measurements of the gravity field provide, for the first
time, estimates of the ice mass changes. However, there
are still significant uncertainties, especially regarding
the cause of the change in mass, as this could come
from the ice sheet itself, the air column, the evolution of
the subglacial bedrock or even from masses near the ice
sheet (i.e. ocean mass, masses of water or snow contained on nearby continents). These different effects are
evaluated and then corrected, but a significant uncertainty remains, mainly on how to correct for the altitude
of the underlying Antarctic bedrock. Other sources of
uncertainty relating to satellite measurements come
from the fact that they do not provide complete coverage of the ice sheets. Moreover, in Greenland, for
example, remote sensing techniques may underestimate
the extent of the ablation area and the increased rainfall
over some parts of the ice sheet during rainfall events.
Past published estimates of Greenland and Antarctic ice
sheets obtained with these methods have often diverged.
This was due to the way in which the different sources of
uncertainty were estimated and to the fact that the different
measurements did not cover the same regions and the same
time periods. Within the IMBIE framework (Ice-sheet Mass
Balance Intercomparison Exercise), scientists made a huge
effort to combine, over common survey periods and common
regions, various observations from satellite geodetic techniques (altimetry, interferometry, radio-echo sounding and
gravimetry) with simulated surface mass balance estimates
inferred from regional atmospheric models. This allowed for
a reconciliation of the apparent disparities between the different methods and for a consistent picture of ice-sheet mass
balance to be created (Shepherd et al. 2012).
For Greenland, this compilation effort confirmed with a
high confidence level that the ice sheet is continuously losing
mass and that this process now affects all sectors of the ice
sheet (Fig. 31.6a–c). However, after a record mass loss in
summer 2012, Greenland has seen a slight decrease in the
short-term mass loss trend. The mass loss is partitioned
between surface melting and dynamic ice discharges. Shepherd et al. (2012) estimate that the ice mass loss is
about −142 ± 49 Gt/yr over the IMBIE time period (1992–
2011) with an acceleration of the mass loss rate as illustrated
by the comparison between the estimations made for 1992–
2000 (–51 ± 65 Gt/yr) and 2005–2010 (–263 ± 30 Gt/yr).
Using gravimetry observations, Velicogna et al. (2014) provide a more recent estimate of the Greenland ice sheet mass
loss for the 2003–2013 decade of 280 ± 58 Gt/yr with an
acceleration of the loss rate bringing it to 25.4 ± 58 Gt/yr
2 .
The case of the Antarctic ice sheet (Fig. 31.6b) is a bit
different since recent observations have shown that mass loss
was mainly driven by dynamic ice discharge resulting from
enhanced ice flow of marine-terminating glaciers. The main
region experiencing mass loss is the West Antarctic ice sheet
(WAIS), especially in the Amundsen/Bellingshausen Sea
sectors (e.g. Thwaites and Pine Island glaciers) and, to a
lesser extent, in the Antarctic Peninsula. According to a
recent update of the IMBIE estimates (IMBIE team, 2018),
the mass loss from the Amundsen and Bellingshausen Sea
sectors increased from 53 ± 29 Gt/yr to 159 ± 26 Gt/yr
over the 1992–2017 period, and from 7 ± 13 Gt/yr to
33 ± 16 Gt/yr in the Antarctic Peninsula. It has long been
considered that the East Antarctic ice sheet (EAIS) was
gaining mass due to enhanced precipitation, despite no firm
31 From the Climates of the Past to the Climates …
455