Greenland Ice Sheet

J.E. Box¹, J. Cappelen², C. Chen¹, D. Decker¹, X. Fettweis³, T. Mote⁴,
M. Tedesco⁵, R.S.W. van de Wal⁶, J. Wahr⁷

¹Byrd Polar Research Center, The Ohio State University, Columbus, Ohio, USA
²Danish Meteorological Institute, Copenhagen, Denmark
³Department of Geography, University of Liège, Liège, Belgium
⁴Department of Geography, University of Georgia, Athens, Georgia, USA
⁵City College of New York, New York, NY, USA
⁶Institute for Marine and Atmospheric Research Utrecht, Utrecht University, Utrecht, The Netherlands
⁷Cooperative Institute for Research in Environmental Sciences, University of Colorado Boulder, Boulder, Colorado, USA

January 14, 2013

Highlights

The duration of melting at the surface of the ice sheet in summer 2012 was the longest since satellite observations began in 1979, and a rare, near-ice sheet-wide surface melt event was recorded by satellites for the first time.
The lowest surface albedo observed in 13 years of satellite observations (2000-2012) was a consequence of a persistent and compounding feedback of enhanced surface melting and below normal summer snowfall.
Field measurements along a transect (the K-Transect) on the western slope of the ice sheet revealed record-setting mass losses at high elevations.
A persistent and strong negative North Atlantic Oscillation (NAO) index caused southerly air flow into western Greenland, anomalously warm weather and the spatially and temporally extensive melting, low albedo and mass losses observed in summer 2012.

Surface Melting and Albedo

In 2012, ice sheet surface melting set two new, satellite era records - melt extent and melt index - according to passive microwave observations made since 1979 (e.g., Tedesco, 2007, 2009). Melt extent is the fractional area (in %) of the surface of the ice sheet where melting was detected. The melt index (MI) is the number of days on which melting occurred multiplied by the area where melting was detected.

Melt extent over the Greenland ice sheet reached record values during 11-12 July, covering as much as ~97% of the ice sheet on a single day (Figs. 5.8 and 5.9, and, e.g., Nghiem et al., 2012). Confirmed by different methods for analyzing passive microwave observations (e.g., Mote and Anderson, 1995; Tedesco, 2009), the almost 100% melt extent is nearly four times greater than the ~ 25% average melt extent that occurred in 1981-2010.

Fig. 5.8. Surface melt extent on the Greenland Ice Sheet on 1 July 2012 (a) and 12 July 2012 (b) detected by the SSM/I passive microwave sensor. Figure is after Tedesco et al. (2007).

Fig. 5.9. Surface melt extent on the Greenland Ice Sheet detected by the SSM/I passive microwave sensor. Figure is after Tedesco et al. (2007).

The standardized melt index (SMI) for 2012 was about +2.4, almost twice the previous record of about +1.3 in 2010 (Fig. 5.10). Melting in 2012 began about two weeks earlier than average at low elevations and, for a given elevation, was sustained longer than the previous record year (2010) for most of June through mid-August. Melting lasted up to 140 days (20-40 days greater than the mean value) at low elevations in some areas of southwest Greenland. The 2012 anomaly for the number of melting days (i.e., number of melting days in 2012 minus the 1980-2010 average) exceeded 27 days in the south and 45 days in the northwest. Some areas in northwest Greenland between 1400 and 2000 m a.s.l. had nearly two months more melt than during the 1981-2010 reference period.

Fig. 5.10. Greenland ice sheet standardized melting index (SMI). The index is calculated by subtracting the melt index (MI) from the 1979 - 2012 average and dividing by its standard deviation (Tedesco, 2007). MI is the number of days on which melting occurred multiplied by the area where melting was detected.

The regions of extended melt duration coincide with areas of anomalously low albedo (or reflectivity, where a surface with low albedo/reflectivity will absorb more solar radiation and warm more than a surface with high albedo/reflectivity). The albedo anomalies across the ice sheet in June-August 2012, when solar irradiance is highest and the albedo is lowest in magnitude, are illustrated in Fig. 5.11. Negative albedo anomalies were widespread across the ice sheet, but were particularly low along the western and northwestern margins in areas where darker bare ice was exposed after the previous winter's snow accumulation had melted completely away. The low albedo was compounded by a persistent feedback of enhanced surface melting due to relatively warm air temperatures and below normal summer snowfall.

Fig. 5.11. Summer (JJA) albedo (reflectivity) anomaly in 2012 relative to the 2000-2011 reference period. Data were derived from MODIS (Moderate Resolution Imaging Spectroradiometer) observations. Figure is after Box et al. (2012).

While Fig. 5.11 shows that there is strong spatial variation in albedo, Fig. 5.12 shows that the area-averaged albedo of the entire ice sheet has continued to decline during the period of MODIS observations (2000-2012). The area-averaged albedo for 2012 was a new record low, and occurred only one year after the previous record of 2011.

Fig. 5.12. Area-averaged albedo of the Greenland ice sheet during June-August each year of the period 2000-2012. Data are derived from MODIS MOD10A1 observations. Figure is after Box et al. (2012).

Equilibrium Line Altitude Along the K-Transect

The 150 km long K-Transect is located near Kangerlussuaq at 67°N between 340 m and 1500 m above sea level (a.s.l.) on the western flank of the ice sheet (van de Wal et al., 2005). The equilibrium line altitude (ELA), the highest altitude at which winter snow survives, is a convenient indicator of the competing effects of surface mass loss from melting and surface mass gain from snow accumulation. The mass balance measurements along the K-transect in 2012 confirm the extensive surface melting observed by satellite (see the previous section on Surface Melting and Albedo).

In 2012, estimates from ground observations placed the ELA far above the height of the ice sheet topographic divide near this latitude (2687 m a.s.l.) and an unprecedented 3.7 times the standard deviation above the 21-year mean ELA value (Fig. 5.13). The satellite-derived snowline, a close proxy of ELA, at the end of the K-transect melt season also occurred at a record high elevation according to MODIS observations made since 2000 (Fig. 5.13) (Box et al., 2009a; van de Wal et al., 2012).

Fig. 5.13. The equilibrium line altitude (ELA), the highest altitude at which winter snow survives, from ground observations (solid line) and the firn line (a proxy for ELA) from MODIS observations (broken line) along the K-transect. It is located near Kangerlussuaq on the western flank of the ice sheet at 67°N between 340 m and 1500 m above sea level (a.s.l.). The difference between the ground and satellite observations in 2012 is because MODIS is not as sensitive to the ELA as it is to the firn line. ELA data are from van der Wal et al. (2012). MODIS data are from Box et al. (2009a).

B_clim (the difference between annual snow accumulation and runoff - see also the Glaciers and Ice Caps essay) during 2011-2012 along the K-transect was characterized by exceptional melt at high elevations. At the highest elevation site (S10, elevation 1847 m, almost 350 m higher than the previous ELA of 1500 m) the surface mass balance was estimated to be -74 cm w.e. (water equivalent). Relatively low winter snow accumulation at high elevation resulted in relatively low albedo, which, coupled with high air temperatures, compounded high melt rates after melt onset. Below 1500 m elevation, surface mass balance values decreased gradually to normal values near the ice margin (Fig. 5.14).

Figure 5.14 suggests that the mass balance along the transect in 2012 was the second lowest since measurements began in 1991. However, a weighted mass balance that includes the S10 site, which is above the former ELA of 1500 m, indicates that the 2011-2012 mass balance year was the most negative in 22 years.

Fig. 5.14. Surface mass balance as a function of elevation below 1500 m along the K-transect since 2008-2009. The 20-year (1990-2010) average is also shown. The 2012 record does not include the highest elevation site, S10, which was measured for the first time in summer 2012. Figure is updated from Box et al. (2011).

Atmospheric Circulation and Air Temperature as they Relate to Melting, Albedo and ELA

The large melt extent and high melt index, low albedo and negative mass balance in 2012 were a consequence of the atmospheric circulation and high air temperatures.

Summer 2012 was characterized by a negative North Atlantic Oscillation (NAO) index for the entire season; a -2.4 standard deviation anomaly relative to the NAO average for June-August during 1981-2010. Consequently, sea level pressure was anomalously high over the ice sheet (see Fig. 1.6 in the essay on Air Temperature, Atmospheric Circulation and Clouds) and atmospheric circulation was characterized by warm air advection from the south into western Greenland (Fig. 5.15). This same circulation pattern has occurred each summer since 2007 (Box et al., 2012). It is noteworthy that a very negative NAO in spring 2012 had a significant negative impact on circum-Arctic snow cover extent (see the Snow essay).

Fig. 5.15. Geopotential height (500 hPa) anomalies for June-August, 2012 (blue lines) referenced to the 1981-2010 mean (magenta lines). The arrow in the lower left quadrant shows that the prevailing upper air flow from the south advected warm air into Greenland. Data source: NCEP/NCAR Reanalysis version R1.

As a consequence of the atmospheric circulation pattern (Fig. 5.15), surface air temperatures at long-term meteorological stations in Greenland were characterized by record-setting warm summer months (not illustrated), particularly in the west and south of the island and at high elevations. For example, the Greenland Climate Network (GC-Net) automatic weather station at Summit (3199 m above sea level) measured hourly-mean air temperatures above the freezing point for the first time since measurements began in 1996 (Fig. 5.16).

Fig. 5.16. Hourly mean air temperature at Summit (elevation 3199 m above sea level) from 9 July through 2 August 2012 (red) and for 1996-2011 (grey). Unpublished data from K. Steffen.

Seasonally-averaged upper air temperature data available from twice-daily radiosonde observations show anomalous warmth throughout the troposphere in summer 2012 (Fig. 5.17). Similar upper air temperature profiles were observed in 2011 (Box et al., 2011). The overall warm pattern near the surface between 850 and 1000 hPa is consistent with a warming trend evident in the period of reliable records beginning in 1964, and most pronounced since the mid-1980s (Box and Cohen, 2006). This recent warming trend is seen in the long-term air temperature reconstruction for the ice sheet, which also shows that mean annual air temperatures in all seasons are now higher than they have been since 1840 (Fig. 5.18).

Fig. 5.17. Upper air temperature anomalies relative to the 1981-2010 reference period in winter, spring and summer of 2012 at four coastal locations in Greenland. The winter season includes data for December 2011. Data are from twice-daily radiosonde observations available at the Integrated Global Radiosonde Archive (http://www.ncdc.noaa.gov/oa/climate/igra/). WMO station numbers are in parentheses.

Fig. 5.18. Seasonally-averaged, near-surface air temperature reconstruction for the entire Greenland ice sheet, 1840 to August 2012 (after Box et al., 2009b).

Greenland Mass Changes from GRACE

GRACE satellite gravity solutions computed according to Velicogna and Wahr (2006) are used to estimate monthly changes in the total mass of the Greenland ice sheet (Fig. 5.19). The data show that the ice sheet continues to lose mass and has contributed +8.0 mm to globally-averaged sea level rise since 2002. The rate of mass loss has accelerated during the period of observation, the mass loss of 367 Gt/y between September 2008 and September 2012 being almost twice that for the period June 2002-July 2006 (193 Gt/y). GRACE data also show that significant mass loss has occurred from glaciers and ice caps in the Canadian Arctic (see the Glaciers and Ice Caps essay).

Monthly smoothed and unsmoothed values of the total mass of the Greenland ice sheet

Fig. 5.19. Monthly smoothed (purple) and unsmoothed (blue) values of the total mass (in Gigatons, Gt), of the Greenland ice sheet from GRACE March 2002-September 2012. The barystatic ("bary" refers to weight) effect on local sea level change is the volume of freshwater added or removed divided by the ocean surface area. It does not include the effects of water thermal expansion, salinity or the associated changes to the gravity field. Figure is after Velicogna and Wahr (2006).

Marine-terminating Glacier Area Changes

Marine-terminating glaciers are the outlets through which the inland ice can flow most rapidly and in the largest quantities to the ocean. Iceberg calving and retreat of the glaciers leads to flow acceleration and inland ice sheet mass loss, which contributes to sea level rise.

Daily surveys using cloud-free MODIS visible imagery (Box and Decker, 2011; http://bprc.osu.edu/MODIS/) indicate that in the year prior to end of the 2012 melt season the marine-terminating glaciers collectively lost an area of 297 km². This is 174 km² greater than the average annual loss rate of the previous 11 years (132 km² yr^-1) (Fig. 5.20) and also greater than losses in the 1980s and 1990s (Howat and Eddy, 2011).

Fig. 5.20. Cumulative net annual area change at the 40 widest marine-terminating glaciers of the Greenland Ice Sheet (after Box and Decker, 2011). The dashed line is a least-squares regression fit with a slope of 131.5 km² of area loss per year since 2000.

Since 2000, the net area change of the forty widest marine-terminating glaciers is -1775 km² (Fig. 5.20), ten times the area of Washington, DC. Glaciers in northernmost Greenland contributed to half of the net area change. In 2012, the six glaciers with the largest net area loss were Petermann (-141 km²), 79 glacier (-27 km²), Zachariae (-26 km²), Steenstrup (-19 km²), Steensby (-16 km², the greatest retreat since observations began in 2000) and Jakobshavn (-13 km²). While the total area change was negative in 2012, four of forty glaciers did grow in area relative to the end of the 2011 melt season.

References

Box, J. E. and A. E. Cohen. 2006. Upper-air temperatures around Greenland: 1964-2005. Geophys. Res. Lett., 33, L12706, doi:10.1029/2006GL025723.

Box, J. E., R. J., Benson, D. Decker and D. H. Bromwich. 2009a. Greenland ice sheet snow line variations 2000-2008, Association of American Geographers Annual Meeting, Las Vegas, NV, 22-27 March, 2009.

Box, J. E., L. Yang, D.H. Browmich and L-S. Bai. 2009b. Greenland ice sheet surface air temperature variability: 1840-2007. J. Climate, 22(14), 4029-4049, doi:10.1175/2009jcli2816.1.

Box, J. E. and D. T. Decker. 2011. Greenland marine-terminating glacier area changes: 2000-2010. Ann. Glaciol., 52(59) 91-98.

Box, J. E. and fifteen others. 2011. Greenland Ice Sheet. In Arctic Report Card 2011, http://www.arctic.noaa.gov/reportcard.

Box, J. E., X. Fettweis, J. C. Stroeve, M. Tedesco, D. K. Hall and K. Steffen. Greenland ice sheet albedo feedback: thermodynamics and atmospheric drivers. The Cryosphere, 6, 821-839, doi:10.5194/tc-6-821-2012, 2012.

Howat, I. M. and A. Eddy. 2011. Multidecadal retreat of Greenland's marine-terminating glaciers. J. Glaciol., 57(203), 389-396.

Mote, T. L. and M. R. Anderson. 1995. Variations in melt on the Greenland ice sheet based on passive microwave measurements. J. Glaciol., 41, 51-60.

Nghiem, S. V., D. K. Hall, T. L. Mote, M. Tedesco, M. R. Albert, K. Keegan, C. A. Shuman, N. E. DiGirolamo and G. Neuman. 2012. The extreme melt across the Greenland Ice Sheet in 2012. Geophys. Res. Lett., 39, L20502, doi:10.1029/2012GL053611.

Tedesco, M. 2007. Snowmelt detection over the Greenland ice sheet from SSM/I brightness temperature daily variations. Geophys. Res. Lett., 34, L02504, doi:10.1029/2006GL028466.

Tedesco, M. 2009. Assessment and development of snowmelt retrieval algorithms over Antarctica from K-band spaceborne brightness temperature (1979-2008). Remote Sens. Environ., 113(2009), 979-997.

Velicogna, I. and J. Wahr. 2006. Acceleration of Greenland ice mass loss in spring 2004. Nature, 443(7109), 329-331. doi:10.1038/Nature05168.

van de Wal, R. S. W., W. Greuell, M. R. van den Broeke, C.H. Reijmer and J. Oerlemans. 2005. Surface mass-balance observations and automatic weather station data along a transect near Kangerlussuaq, West Greenland. Ann. Glaciol., 42, 311-316.

van de Wal, R. S. W., W. Boot, C. J. P. P. Smeets, H. Snellen, M. R. van den Broeke and J. Oerlemans. 2012. Twenty-one years of mass balance observations along the K-transect, West Greenland. Earth Sys. Sci. Data, 5, 351-363, doi:10.5194/essdd-5-351-2012.