Old ice investigation along the Dome C ridge using a 2.5D thermomechanical ice flow model
Olivier Passalacqua (1), Frédéric Parrenin (1), Olivier Gagliardini (1,2), Catherine Ritz (1), and Fabien Gillet-Chaulet (1)
(1) Laboratoire de glaciologie et géophysique de l’environnement (CNRS-UJF), Grenoble, France
(2) Institut universitaire de France, Paris, France
The presented model is to consider a 2D steady-state ice flow along the
ridge to Vostok, that accounts for the widening of the flow tube (2.5D model)
and thermal advection and diffusion.
Such a site does exist at ~40 km southwest from the Dome C, straight under
the ridge to Vostok. A secondary dome
relief is situated at 30 km from the main
dome, where the speed of the ice is
expected to be null.
The 2.5D model does not hold if the width of the tube highly diverges.
Accordingly, we will run the model only on the downstream part, starting from
the secondary dome (last 15 km on the right).
Relative width
of the ice flow
-3°C
The geothermal ux warms the base
but the melting point is not reached.
The heat equation is modified as well, to account for the width of the tube,
and allows to compute the basal melting.
-50°C
Q g = 65 mW/m2
Q g = 70 mW/m2
3e+06
age
2e+6
1e+6
0
Model characteristics
-2°C
The geothermal ux melts the basal ice.
To be done
• Effect of anisotropy and sliding.
• Precise geothermal flux estimation.
• 3D runs will be performed, with evolving geometry of the surface.
Vertical velocity (m/a, downward)
0.0433
Acknowledgments
0.0193
0.04
0.01
0.03
0.02
0.001
0.01
0.00017
0
Geothermal flux : 65 mW/m2
Accumulation : 0.019 m/a
Temperature (K)
271
260
240
N
220
213
• Determining the radius R is not straightforward, and the 2.5D model only
works when the tube width evolves gently.
• The ice is frozen at the bed or encounters little basal melting. Our first
simulations show that ice older than 1 million years is conserved, but
uncertainties remain.
• Uniform geothermal flux
• Isotropic ice
• Free surface, accounting for the mean past accumulation
• Null speed on the left boundary, and a vertical profile on the right
Horizontal velocity (m/a)
Take-home message
• The ice travels only ~10 km from a secondary dome to the investigated coring
site.
3 Computed ice flow and temperature
Surface speed (0.1m/a)
An advection-reaction equation is solved to compute the age of the ice, for
several geothermal fluxes and accumulation values. Contour lines correspond
to 800,000 years and 1,500,000 years.
-50°C
H < ~2700 m
Heat escapes
H > ~2700 m
Too much a thick ice acts as an
insulator, so that the basal layer would
reach the melting point (Pattyn, 2010).
Considering the probable values of the
geothermal flux in Antarctica, an
appropriate thickness ranges from
2500 m to 3000 m.
The symmetric shape of the surface leads us to use a 2.5D model (Reeh, 1988 ;
Schøtt Hvidberg, 1993). The 2D Stokes equations are written to account for
the widening of the flow tube W, and implemented in the Elmer/Ice finite
element model. Instead of W, the equations rather depend on the radius R of
the surface contour lines, determined thanks to the digital elevation model
Bedmap 2. W and R are linked as follows :
Momentum and mass
conservation
1 Location of the site
We look for a site near Dome C where
the basal layer does not encounter
melting.
4 Computed age
a = 0.019 m/a
Fourty kilometers from the dome lies a bedrock relief that makes the ice
thinner (2700 m), so that the bottom ice could be prevented from
encountering basal melting. We show that due to a small dome
configuration leading to null horizontal velocities the ice at this possible
drilling site mainly comes from 10 km upstream only.
2 Ice flow model
a = 0.022 m/a
One of the main present-day challenges in ice core sciences, as fixed by the
IPICS (International Partnerships in Ice Core Sciences), consists in finding a
continuous ice archive as old as 1.5 million year. The previous oldest ice core
was drilled at Dome C, on the East Antarctic plateau (800,000 years), and
some observations seem to indicate that even older ice could be retrieved in
the vicinity of the dome.
• The horizontal velocity is low (a few
centimeter per year).
• The vertical velocity near the bedrock
depends wether the melting point is
reached or not.
We're beholden to the Norwegian Polar Institute for the Quantartica 2 QGIS distribution. For large
scale computations, we used the Bedmap 2 digital elevation model (Fretwell et al., 2013). We also
thank Luca Vittuari (Università di Bologna) for making the surface speed data available.
References
Fretwell et al (2013). Bedmap2 : improved ice bed, surface and thickness datasets
for Antarctica. The Cryosphere , 7 (1), 375-393.
Pattyn, F. (2010). Antarctic subglacial conditions inferred from a hybrid ice
sheet/ice stream model. Earth and Planetary Science Letters , 295 (3), 451461.
Reeh, N. (1988). A flow-line model for calculating the surface profile and the
velocity, strain rate, and stress fields in an ice sheet. Journal of Glaciology , 34
(116).
Schøtt Hvidberg, C. (1993). A thermo-mechanical ice flow model for the
centre of large ice sheet. Ph.D. thesis, University of Copenhagen.
Contact : [email protected]
Find this poster on the web !