Iceshelf melting in Antarctica

1
Ice-shelf melting in Antarctica
Eric Rignot1 and Stanley Jacobs2 1 University of
California Irvine and Jet Propulsion Laboratory,
Pasadena CA 2 University of Columbia, Lamont
Observatory, NY.
2
Importance of ice-shelf melting

• Ice shelves buttress the flow of continental ice
  into the ocean.
• Ice shelf removal can destabilize Antarctica
  Larsen A/B (3-8 times acceleration), Pine Island
  shelf (74 in 30 years).
• Ice shelf melting is 10-100 times larger than
  surface mass balance.
• Ice shelf melting is controlled by the ocean
  temperature above the freezing point.
• What fraction of ablation is controlled by B vs
  calving?
• What is the status of the southern ocean?
• How will it change in the future?
• How will this affect ice-shelf melt rates?

3
Bottom melting of Petermann Gletscher, Greenland
4
Bottom melting heat source and effect of
pressure.

• Mode 1 Thermohaline circulation induced by sea
  ice formation and drainage of dense saline water.
  Melting point decreases as much as 1oC due to
  0.00075 oC/dbar-1
• Mode 2 Direct inflow of intermediate-depth
  warm water from the slope-front region (e.g.
  Circumpolar Deep Water intrusion through deep
  troughs).
• Mode 3 Ice-front interactions (tidal pumping,
  coastal currents)

AASW
CDW
ISW
Jacobs et al., 1992

• Ice shelf water may refreeze to form marine ice,
  or sink to participate in the formation of
  Antarctic bottom water, which regulates global
  climate.
• Mode 1 dominates for large ice shelves in the
  Weddell, and Ross seas.
• Mode 2 melts large volumes of ice where deep
  water has access to glacier grounding lines, e.g.
  Amundsen Sea.

5
Ice shelf melting vs iceberg calving

• Ice shelf melting near GL vs T-Tm at 0.8 Hmax
  (Rignot and Jacobs, 2002).
• Linear relationship, large spread.
• Uncertainties
• What GZ?
• What Hmax?
• What T?
• Here, (T-Tm) vs B averaged over z.
• Better/more velocity (Radarsat, ALOS PALSAR),
  topography (ICESAT-1), SMB (RACMO2/ANT).

1 m/year/ 0.1oC
6
Antarctic Ice VelocityERS-1/2 (1996), Radarsat-1
(2000-2007), ALOS PALSAR (2006)
7
Ice thickness

• Ice thickness, H, from surface elevation
  a.m.s.l., h, density, and firn depth correction,
  ?H
• H(h-?H)(?sea)/(?sea-?ice)
• Error 10 m (ISR), 80 m (most glaciers), 100 m
  (AP, WL), 120-200 m (Byrd, Larsen B).
• Small glaciers not included because of DEM
  resolution.

2,465 points, ASE sector
Mean error 1463 m.
0 40 m
ERS ICESAT topography
Firn depth correction
8
Antarctic grounding lines
9
Ice-shelf surface mass balance

• RACMO2/ANT (van den Broeke et al. 2006 van de
  Berg et al., 2005) only 6 difference with
  earlier maps on average but gt 100 on wet coastal
  basins (Getz).
• Precision 5 (dry) to 30 (wet).
• No change in integrated snowfall 1980-2004.
• Antarctic GIS 205960Gt/yr
• Antarctic GISIce Shelves 252160Gt/yr
• Year 2000 GL flux 219792Gt/yr

10
Estimates of Bottom Melting

• B (?F - SMBIS ) / area (steady state or dhdt
  0)
• ?SMB 10 (dry) to 30 (wet).
• Flux ?H 80m, ?V 5-50m/yr,
• Grounding line ?F/F ?H/H?V/V80/10005/600
  9
• Ice Front ?F/F80/2505/600 30
• dB/B 34
• Ross Ice Shelf 309cm/yr vs PIG 257m/yr
• If ice shelf is thickening by dhdt, then B
  Bsteady - dhdt

11
Ronne/Filchner Ice Shelves
Melt 98 Gt/yr
12
Ross
Melt 77 Gt/yr
13
Abbot and Amery
14
Totten
Melt 89 Gt/yr
15
Queen Maud Land, East Antarctica
Melt 50 Gt/yr Calving 45 Gt/yr
16
Antarctic Peninsula
IS SMB 18 Gt/yr Melt 10 Gt/yr
14 Gt/yr
Total Melt 67 Gt/yr Melt GVI 57 Gt/yr Calving
28 Gt/yr
17
Pine Island and Getz ice shelves
Melt 123 Gt/yr Calving 148 Gt/yr
Melt 86 Gt/yr Calving 44 Gt/yr
18
Ice shelf melt water production
19
Ice shelf melt rates
20
Ice shelf melting vs ocean thermal forcing
Pine Island
Thwaites
Dotson
Getz
Ross
LarsenC
Filchner
21
Ice shelf melting vs ocean thermal forcing
To get those ice shelves to fall on
curve George VI T- Tm 1.2 oC instead of
NA Venable, Cosgrove, Abbot 1 to 1.5 oC
instead of 2 to 2.4 oC, which means only top
400 m Ronne 0.4 oC instead of 0.26 oC
Filchner 0.3 oC instead of 0.08 oC (station
too cold) Amery, QML, Fimbul 0.5 oC instead of
NA. West and Shackleton ice shelves 0.9 to 1.5
oC instead of NA Totten and Moscow Un. Ice
shelves 2.1 to 1.5 oC (!) instead of 0.8 oC
Pine Island
Thwaites
Dotson
Getz
Ross
LarsenC
Filchner
22
Conclusions

• Ice shelf melting controls gt 50 of the ice
  sheet/ice shelf mass balance.
• This neglects ice-front sub-aqueous melting.
• Low melt on large ice shelves (far from CDW),
  Queen Maud Land, East Peninsula.
• High melt on West Peninsula (CDW), Amundsen,
  Bellingshausen sea (CDW), Wilkes Land (?).
• Quadratic dependence on temperature where CDW
  fuels high melt.
• Elsewhere, linear relationship might still hold
  but not enough ocean temperature data near GL.

23
(No Transcript)