The Arctic record

The Arctic sea ice record is one of the primary 'poster children' for demonstrating that the Earth's climate is undergoing warming, at least in the North Polar Region. Although the record exhibits considerable interannual and regional variability, the evidence for decreasing Arctic sea ice coverage is extremely strong (see Figure 4.3) and has been extensively reported and discussed. This decrease has been most frequently presented as a decrease in Arctic sea ice extent (see, for example, Parkinson and Cavalieri, 1989; Johannessen et al, 1995, 2004; Maslanik et al, 1996; Bj0rgo et al, 1997; Parkinson et al, 1999; Walsh and Chapman, 2001; Comiso, 2006), but the record also shows a decrease in the length of the sea ice season (Parkinson, 1992, 2000b), an increase in the length of the melt season (Smith, 1998), and, from in situ and submarine data, thinning of the sea ice cover (Rothrock et al, 1999; Wadhams and Davis, 2000; Yu et al, 2004).

12.5 —I—|—I—I—I—I—[— I I I I I —I—I—I—I—I—I— S—1—I—I—J—I—I—r

1979 1984 1989 1994 1999 2004

Figure 4.3 Annual average sea ice extent in the Northern Hemisphere, 1979-2004, as derived from data from the satellite-borne SMMR and SSMI instruments

1979 1984 1989 1994 1999 2004

Figure 4.3 Annual average sea ice extent in the Northern Hemisphere, 1979-2004, as derived from data from the satellite-borne SMMR and SSMI instruments

Note: Ice extent is the sum of all grid cells (each approximately 25 km x 25 km) with ice concentration of at least 15 per cent. The dashed line is the linear least-squares fit through the data points.

The Arctic sea ice decrease is not uniform temporally (see Figure 4.3) or regionally (see Figure 4.4). For instance, the ice cover in the Seas of Okhotsk and Japan largely decreased, on an annual average basis, from 1979 to 1997, then increased until 2001 and then decreased again, while the sea ice cover of the Bering Sea instead started in 1979 at one of its lowest annually averaged extents for the entire 26-year record and had its highest extent in 1999, although still showing a negative trend overall from 1979 through 2004 (see Figure 4.4). The differing sea ice changes in the Bering Sea and the Seas of Okhotsk and Japan have been linked to the North Pacific Oscillation and the positioning of the Aleutian Low atmospheric pressure system (Cavalieri and Parkinson, 1987), and other regional contrasts throughout the Arctic have been similarly linked to various atmospheric circulation patterns and large-scale oscillations in the climate system. Considerable attention has been given to possible connections of the Arctic sea ice with the North Atlantic Oscillation (see, for example, Hurrell and van Loon, 1997; Johannessen et al, 1999; Kwok and Rothrock, 1999; Deser et al, 2000; Kwok, 2000; Parkinson, 2000a; Vinje, 2001), although some attention has also been given to connections with the Arctic Oscillation (Deser et al, 2000; Wang and Ikeda, 2000), the Arctic Ocean Oscillation (Polyakov et al, 1999; Proshutinsky et al, 1999; Rigor et al, 2002), and an unnamed interdecadal Arctic climate cycle (Mysak et al, 1990; Mysak and Power, 1992).

Still, despite the non-uniformity of the sea ice decrease and the possible connection with various oscillatory patterns, the long-term decreasing trend for the Northern Hemisphere as a whole is quite apparent, both visually (see Figure 4.3) and statistically. The trend value for the yearly averaged ice extents in Figure 4.3 is -38,100 ± 4200 km2/yr (-3.1 ± 0.34 per cent per decade), which is statistically significant at the 99+ per cent confidence level. Furthermore, the negative trend is apparent, from the SMMR/SSMI data, in every month of the year. The strongest trend is a decrease of 51,300 ± 10,800 km2/yr (6.9 ± 1.5 per cent per decade) for the month of September, but even the weakest monthly trend (for February) is still significant at 27,500 ± 5800 km2/yr (1.8 ± 0.4 per cent per decade).

The September Arctic ice decrease has garnered particular attention in recent years (see, for example, Maslanik et al, 1996; Comiso, 2002; Stroeve et al, 2005), not only because it is the largest of the monthly decreases, but also because of speculation that the September decrease could result in a late-summer ice-free Arctic, perhaps by the middle to end of the 21st century. Although the lesser wintertime decreases have also been reported previously (see, for example, Parkinson et al, 1999), recently they have been further highlighted because of particularly low winter sea ice coverage in both 2005 and 2006 (Comiso, 2006).

Figure 4.4 Annual averaged sea ice extent for 1979-2004, as derived from data from the satellite-borne SMMR and SSMI instruments, for (a) the Seas of Okhotsk and Japan and (b) the Bering Sea

Note: The dashed lines are the linear least-squares fits through the respective data points.

Was this article helpful?

0 0

Post a comment