By Steven Goddard,
NSIDC published their sea ice news yesterday, and this one is definitely worth a read. Yesterday I pointed out that the graph below seems to be inconsistent with other data, including NSIDC maps.
The problem is that the 2010 curve appears too close to 2007. Other data sources have the spread much larger, and NSIDC’s own maps show a larger spread. The area of green below represents regions of ice present in 2010, but not present in 2007. As of today, NSIDC maps show 10% more ice in 2010 than the same date in 2007.
Walt Meier from NSIDC responded with this remark :
4. Our sea ice maps are not an equal area projection. Thus one cannot compare extents by counting grid cells – this is probably the reason for the 7.5% vs. 3% discrepancy. Steve has been alerted to this issue in the past, but seems to have forgotten it.
What Dr. Meier seems to have forgotten is that pixels further from the pole in a polar map projection represent larger areas. Thus a correction would slightly increase the discrepancy, not decrease it. Sadly, DMI stopped updating their graphs two days ago – so I am no longer able top do comparisons between DMI 30% concentration and NSIDC 15% concentration. Their most recent graph shows 2010 well above 2007, and close to 2006.
Another data source – JAXA. The gap between 2010 and 2007 has been decreasing in NSIDC 15% concentration data, but has been increasing in JAXA 15% concentration data.
The next item which caught my attention is the discussion of multi-year ice.
Older, thicker ice melting in the southern Beaufort Sea
This past winter’s negative phase of the Arctic Oscillation transported old ice (four, five, and more years old) from an area north of the Canadian Archipelago. The ice was flushed southwards and westward into the Beaufort and Chukchi seas, as noted in our April post. Ice age data show that back in the 1970s and 1980s, old ice drifting into the Beaufort Sea would generally survive the summer melt season. However, the old, thick ice that moved into this region is now beginning to melt out, which could further deplete the Arctic’s remaining store of old, thick ice. The loss of thick ice has been implicated as a major cause of the very low September sea ice minima observed in recent years.
The multi-year ice has largely survived the summer so far. Pixel counts show that ice greater than two years old has dropped by 11%, and ice between one and two years old has dropped by 4%. (These numbers are slightly low because of the distortion described above.) Most of the ice lost has probably been transported out the Fram Straight near Greenland, rather than melted in situ. The ice in the Beaufort Sea has split and moved north and west.
What about the future? The remaining multi-year ice in the Beaufort Sea is largely contained in areas which have dropped below freezing, and are forecast to remain below freezing for the next two weeks. The image below blinks between multi-year ice and current temperatures. Blue indicates below freezing temperatures.
The NCEP forecast below shows freezing temperatures over the ice for most of the remainder of the Arctic summer.
It appears that the vast majority of the multi-year ice will survive this summer – just as it did in the 1970s and 1980s. The language in the NSIDC article seems to indicate that something fundamental has changed. I don’t see much evidence of that. In fact, given the large amount of 1-2 year old ice, we should see an increase in the amount of MYI next year.
Ice age data show that back in the 1970s and 1980s, old ice drifting into the Beaufort Sea would generally survive the summer melt season. However, the old, thick ice that moved into this region is now beginning to melt out, which could further deplete the Arctic’s remaining store of old, thick ice. The loss of thick ice has been implicated as a major cause of the very low September sea ice minima observed in recent years.
And no mention of the record high ice extent in Antarctica.
I have alerted Dr. Meier to most of these issues by E-mail.