Supplementary Information
Acceleration of Jakobshavn Isbræ Triggered by Warm Subsurface Ocean Waters DAVID M. HOLLAND1,*, ROBERT H.THOMAS2, BRAD DE YOUNG3, MADS H. RIBERGAARD4, BJARNE LYBERTH5 1 New York University, New York 10012, USA EG&G, Wallops Flight Facility, Virginia 23337, USA 3 Memorial University, St. John’s A1B 3X7, Canada 4 Danish Meteorological Institute, Copenhagen DK‐2100, Denmark 5 Greenland Institute of Natural Resources, Nuuk 3900, Greenland *e‐mail: [email protected] 2
Background Considerable progress has been made over the last decade in constraining the present mass
budgets of the large ice sheets, Greenland and Antarctica, largely by airborne and satellite
techniquesS1,S2, with both sheets presently considered to be contributing to observed global sealevel rise. The mass budgets are a function of atmospheric and oceanographic conditions near
the sheets. While the atmospheric conditions have been relatively well observed over past
decadesS3, the oceanographic conditions are poorly known, having suffered from much sparser
observations. The presence of ice and ice-calving makes such observations difficult to obtain.
We here provide some brief discussion of the factors that influence glacial dynamics with an
emphasis on the circulation of subsurface ocean waters along the periphery of the ice sheets, and
of the Jakobshavn Isbræ (JI) in particular.
Glaciology The present mass balance of the ice sheets reveals a net gain of mass over the vast interior
surface areas, and a net loss at the edge. Some loss is caused by increased summer melting,
particularly in Greenland, and some by recent acceleration of specific glaciers and ice streams
that transport ice from the slow moving interior to the coast. There, it is discharged into the
ocean, either by calving of icebergs or by melting from beneath floating ice tongues.
Until recently, JI terminated as a floating ice tongue. The JI is unique in having a detailed time
series of glaciological measurements since the early 1990s, spanning the breakup of the floating
tongue and subsequent glacier acceleration. It is also known that the JI calving front retreated
S1
30 km between 1850 and 1964 and subsequently occupied approximately the same location until
the late 1990sS4,S5. Balance calculationsS6 suggest that total loss by surface and basal melt and
ice discharge was slightly less than total accumulation within the drainage basin. Estimates of
anomalous surface melting using positive-degree-day methods are small compared to total
thinning, suggesting that a large part of the thinning is dynamic, associated with increased
longitudinal creep rates, and therefore velocity.
A curious fact is that while the JI is now flowing faster than prior to 1998, and thus now
transporting more land ice towards the sea, it is at the same time retreating; the ice is thinning
more rapidly than can be balanced by the thickening effects of increased advection of thicker ice
from upstream. Extremely rapid thinning results because the ice is so deep at the grounding line,
where removal of the ice tongue allows creep thinning rates to be roughly proportional to the
forth power of the ice thicknessS7. Continued rapid thinning inevitably causes additional
grounding-line retreat over a bed that deepens further inland. Consequently, continued retreat
and probably further acceleration is likely unless a new ice tongue develops, and “jams up” part
of the fjord. This is unlikely, because the heavily-crevassed glacier fragments into small
icebergs close to the calving front, and because continued subsurface warm waters in the fjord
favor rapid melt rates.
Oceanography Relative to other factors influencing the mass balance of the ice sheets, the temperature of the
subsurface waters near the ice edge is, as mentioned, difficult to observe as the ocean there is
frequently covered by ice shelves, icebergs, or sea ice. Nonetheless measurements have been
made using either traditional oceanographic instrumentation or autonomous underwater
vehiclesS8. Available data show the presence of relatively warm, saline waters several hundred
meters below the ocean surface, well separated from the cool, fresh surface waters. Indeed, such
warm, subsurface waters are quite common in the polar oceans. They originate in the extra-polar
oceans and although warmer than the surface waters of the polar oceans, they are saltier and
denser and consequently as they move poleward they sink beneath the cold, fresh polar surface
waters and generally spread laterally at depths of between about 200 and 800 m. Where such
waters come directly into contact with the periphery of an ice sheet, enormous melt rates resultS9.
Floating glacier tongues and the ice shelves into which most Antarctic glaciers flow are
particularly susceptible to increased basal melting, as recent observations show that some of
these ice shelves are already thinningS10,S11.
In the particular instance of the oceanography of West Greenland, changes in the ocean
temperature regime are traditionally described on the basis of measurements made at
oceanographic standard stations, such as SJA3 as discussed in the main text (Fig. 3). In turns out
that from a fisheries-science perspective, these data are too sparse and do not necessarily reflect
the conditions faced by demersal species. In contrast to standard oceanographic stations, bottom
temperatures from fishery trawls sites show the conditions encountered by species on a finer
spatial scale. Interestingly, these data are the most complete set available to demonstrate the
presence of subsurface warm water on the continental shelf. These are precisely the waters that
are critical to monitor to assess the degree of ocean-glacier interaction. Only by serendipity did
the demersal fisheries surveys record a deep ocean temperature jump for the JI in 1997 (Fig. 4).
S2
In general, it must be acknowledged that the monitoring of the subsurface ocean waters near the
ice sheets and beneath ice shelves remains an acute observational problem.
Supplementary References S1. Rignot. E. & Thomas, R.H. Mass balance of polar ice sheets, Science, 297, 1502 (2002).
S2. Shepherd, A. & D. Wingham, D. Recent Sea-Level Contributions of the Antarctic and Greenland Ice Sheets,
Science, 315, 1529-1532 (2007).
S3. Monaghan, A.J. et al., Science, 313, 827-831 (2006).
S4. Weidick, A. Satellite Image Atlas of Glaciers of the World: Greenland, United States Geological Survey,
professional paper 1386-C, edited by R.S. Williams and J. Ferrigno, U.S. Government Printing Office,
Washington, DC, 141 pp (1995).
S5. Sohn, H-G., Jezek, K. & van der Veen, C. Jakobshavn glacier, west Greenland: 30 years of spaceborne
observations, Geophys. Res. Lett., 25, 2699-2702 (1998).
S6. Echelmeyer, K., Harrison, W., Clarke, T. & Benson, C. Surficial glaciology of Jakobshavns Isbrae, west
Greenland: part II. Ablation, accumulation and temperature, J. Glaciol., 38, 169-181 (1992).
S7. Weertman, J. Deformation of floating ice shelves, J. Glaciol., 3, 38-42 (1957).
S8. Nicholls, K.W. et al. Measurements beneath an Antarctic ice shelf using an autonomous underwater vehicle,
Geophys. Res. Lett., 33, L08612 (2006).
S9. Motyka, R., Hunter, L., Echelmeyer, K.A. & Connor, C. Submarine melting at the terminus of a temperate
tidewater glacier, Annals Glaciol., 36, 57-65 (2003).
S10. Zwally, H.J. et al. Mass changes of the Greenland and Antarctic ice sheets and shelves and contributions to sealevel rise: 1992– 2002, J. Glaciol., 51, 509– 527 (2005).
S11. Shepherd, A., Wingham, D. & Rignot, E. Warm ocean is eroding West Antarctic Ice Sheet, Geophys. Res.
Lett., 31, L23402 (2004).
S12. Valeur, H.H., Hansen, C., Hansen, K.Q., Rasmussen, L. & Thingvad, N. Physical environment of eastern Davis
Strait and northeastern Labrador Sea, Danish Meteorological Institute Technical Report 97-09, Copenhagen
(1997).
S13. Ribergaard, M.H., Kliem, N. & Jespersen, M. HYCOM for the North Atlantic Ocean with special emphasis on
West Greenland Waters, Danish Meteorological Institute Technical Report 96-07, Copenhagen (2006).
S14. Haven, K. Greenland’s Ice Island Alarm. NASA Earth Observatory Feature.
http://earthobservatory.nasa.gov/Study/Greenland/greenland.html (2007)
S15. Ribergaard, M.H. Oceanographic Investigations off West Greenland 2005, NAFO Scientific Council
Documents 06/001 (2006).
S16. Climatology/Meteorology Research Group, http://www.giub.unibe.ch/klimet/wanner/nao.html (2008).
S17. Hurrell, J. Decadal trends in the North Atlantic Oscillation: Regional temperatures and precipitation, Science,
269, 676–679 (1995).
S18. Blindheim, J., Toresen, R. & Loeng, H. Fremtidege klimatiske endringer og betydningen for foskeressuresene,
Havets miljo, 73-78 (2001).
S19. Ribergaard, M.H., Olsen, S.M. & Mortensen, J. Oceanographic Investigations off West Greenland 2007, NAFO
Scientific Council Documents, 08/003 (2008).
S20. Smed, J. Fluctuations in the temperature of the surface water in the areas of the northern North Atlantic, 18761975. In: Danish Meteorological Institute Climatological Papers, 4, 205-210 (1978).
S21. Drinkwater, K.F. The regime shift of the 1920s and 1930s in the North Atlantic, Prog, Oceanogr., 68, 134-151
(2006).
Supplementary Figures S3
Figure S1. Schematic of water masses near the continental shelf break, West Figure
S1 Schematic of water masses near the continental shelf break West
GreenlandS12,S13. The waters on the continental shelf here consist of two distinct water masses. The surface layer consists of Polar Water (PW), a cold, fresh water mass originating in the Arctic Ocean, but with local input due to freshwater runoff from coastal glacier melt. The subsurface layer consists of Irminger Water (IW), a relatively warm, saline water mass originating in the North Atlantic. As IW advects around the southern tip of Greenland, it subducts beneath the PW. Along the west coast of Greenland the core of the subducted IW is typically found at depths of about 400 600 m and located at
core of the subducted IW is typically found at depths of about 400‐600 m, and located at the continental shelf break.
Figure S2. Terminus location for the JI during various years over the period 1851 ‐
2006S14. The glacier has undergone a nearly continuous retreat over the entire period, but at rates that have varied over time. Most notable, the glacier front essentially stagnated during the period 1964 to 2001, and then during 2001‐2006 recorded its largest retreat rate.
a
b
c
Figure S3. Time series from standard oceanographic sections, West Greenland sect
o s, est G ee a d
south of Disko BayS15. a Map of Greenland showing (as red dotted lines) some of the standard oceanographic sections taken annually in early summer. The data from the northernmost section shown in
northernmost section shown in panel b is ‘Hol5’ (near Holsteinborg), and the second northernmost section data shown in panel c is ‘Suk5’ (near Sukkertoppen). b
Temperature and salinity collected at station Hol5 during
collected at station Hol5 during the period 1950‐2005. The data clearly shows a steady increase in Irminger Water temperature in the 400‐600 m depth range since 1996, and in the intermediate waters from 150‐400 m between the Polar b
h
l
water core (~ 50 – 150 m) and the Irminger Water core (~400 – 600 m). c Same as panel b, except for standard section Suk5, further south.
a
Figure S4. Observations of warm, subsurface waters at Jakobshavn during summer 2007. a Blue curve is an AXCTD temperature profile taken in the fjord August 21,2007, at the location marked by blue square
location marked by blue square icon in Fig 2a. Red curves illustrate one dozen CTD temperature profiles taken June 7, 2007, in the vicinity of the mouth of the fjord, marked by red circle icon in Fig 2a. b Same as panel a, except for salinity. c
tf
li it
Thi k
Thickness (in meters) of the observed bottom layer of warm water (defined as having temperature above 1.5 °C) near the mouth of the fjord derived from the CTD data in panel a (red curves). The gray shading indicates a mask for land areas in the vicinity of the mouth of the fjord. The relatively thick, warm water layer sits on the floor of Disko Bay at depths above the nearby sill depth of 350 m at the mouth of the fjord, and thus has direct access to the ocean fjord.
b
c
a
b
Figure S5. Schematic of positive and negative phases of the North Atlantic Oscillation (NAO)S16. The North Atlantic marine climate is l
largely influenced by the so‐called l i fl
db h
ll d
North Atlantic Oscillation (NAO), driven by the pressure difference between the Azores‐High and the Icelandic‐Low pressure cells. This is the dominant mode of variability in the North Atlantic winter climateS17. a Positive NAO index is associated with a more northeasterly storm track. b Negative NAO index is associated with a more southerly, and weaker, storm track.
a
b
Figure S6. Schematic of North Atlantic Ocean subpolar gyre during positive and negative phases of the atmospheric NAOS18. a . a The NAO is known to The NAO is known to
impact the ocean circulation and hydrologic conditions in the North Atlantic region including the western Greenland waters. The North Atlantic Current (NAC) is narrowed and strengthened during periods of high NAO This
during periods of high NAO. This is caused by an increase in westerly wind stress, and the oceanic current response is sketched in panel a. b In periods of low NAO the NAC is widened and the westward branching of currents is increased, including a t i i
d i l di
noticeable increase in the Irminger Current.
a
Fylla2
b
Figure S7. Time series of mean temperature at Fylla BankS19. a Map of Greenland showing (as red dotted lines) some of the standard oceanographic sections taken annually in early summer. The data from the section shown in panel b is station ‘Fylla2’ (near Nuuk). b
Temperature collected at Fylla2 over depths 0–40 m during the middle of June for the period 1950–2007. The red curve is the 3‐year running mean value, and the purple curve extends the time series back to 1875 using Smed’s climatologyS20. While these are direct measurement of the surface temperature these waters are nonetheless modified by
measurement of the surface temperature, these waters are nonetheless modified by Irminger Water en route to Fylla Bank, thus partly reflecting the temperature of the Irminger Water as originating in the North Atlantic.
Figure S8. Northward advance of cod along West Greenland in the early 20th century, during a previous warming periodS21. During the 1920s and 1930s there was a dramatic warming of the North Atlantic Ocean. Warmer‐than‐normal sea temperatures and enhanced Atlantic inflow in the northern oceans continued through to the 1950s and 1960s. Ecosystem changes associated with the warm period included a general northward movement of fish. The maximum recorded movement involved cod which spread approximately 1200 km
The maximum recorded movement involved cod, which spread approximately 1200 km northward along West Greenland, their progression marked by the red lines along the coast.