A huge, long-lived Arctic ocean storm has decimated the sea ice area which was melting out at a record rate before the high waves and winds shattered the Siberian side of the ice cap.
NOAA Pacific Surface map, 18GMT (6PM Greenwich Mean Time) August 6, 2012.
The record minimum ice area will almost certainly be broken well before ice starts to rebuild in mid to late September because weather patterns remain favorable for continued melting. Before the extraordinary melt season of 2007 summer storms kept the sea ice cool and cloudy, preserving the ice cap. In 2007 most of the thick multi-year sea ice was lost to melting and steady winds that blew large amounts of ice out of the Arctic. Now the thin ice is vulnerable to storms which break up the ice and mix it with water that's been heated by the summer sun in ice free areas. Large areas of open water this summer set the scene for the catastrophic effects of this exceptional storm. Because sea ice is so much more reflective than open water, progressive loss of sea ice establishes a feed back loop of more warming and ice loss. In previous posts I have called this feedback loop a death spiral for Arctic sea ice because it is rapidly leading the Arctic ocean to ice free Septembers.
The sea ice area for August 9 was lower this year than in previous years as determined by Cryosphere Today of the University of Illinois. The rate of area loss is the highest since measurements began in 1979.
Because large storms blow sea ice away from the storm's center, like a centrifuge, the sea ice extent, which includes all areas with greater than 15% ice, is slower to respond to storms than sea ice area. In 2007 ice was compacted by high pressure, minimizing the sea ice extent. This year is the opposite of 2007 in terms of storminess, but both years have seen rapid melting. The slushy, thin ice blown out from this storm is likely to melt out in the area offshore of Alaska and Siberia, likely leading to a drop in sea ice area over the next few weeks. Sea ice extent will probably reach a record minimum this year a week or two later than the date the record minimum sea ice area is broken.
Disturbingly, a feedback loop no one expected to see any time in the near future may have been started by this storm. The Arctic ocean is protected from melting from below by a surface layer of cold fresh water that flowed in from Siberian rivers. A layer of relatively warm, salty water that originated in the north Atlantic lies below the fresh cold surface water. Preliminary, uninterpreted data may show that that stratification has been broken by the intense storm. Hurricanes and large winter storms are well known to cause upwelling of cool water from deeper layers. That is normal. However, if warm salty water is upwelled in the Arctic ocean, it could have a large positive effect on the heat balance. It could be another destructive feedback loop for sea ice. It could significantly amplify the rate of Arctic warming and global climate change. We don't yet know that this feed back loop has been triggered, but an observation on Neven's blog needs to be followed up by researchers to determine what has just taken place in the Arctic. When I examined the buoy profiles he linked to, I saw possible evidence of upwelling, too. This is a problem I didn't think we would need to worry about until I saw the size and duration of this storm.
I understand Neven's position that we should wait and see what the impact of the storm will really be after the clouds are gone, but let me tell you that I have a very bad feeling about this one.
Last year, with the "flash melt" during the November storm, I was confident that the "flash melt" was mostly caused by sea water flushing over the ice, which only temporarily (and artificially) confuses the satellite sensors in believing that large swats of ice have turned to water.
At that time, the ocean flux data from ITP buoys revealed that ocean water was stirred up down to 25 meter or so, which caused some salty water to bubble up to under the ice, with the potential to 'flash' melt out about 10 cm from the bottom of the ice pack.
This storm however, is completely different.
For starters, freezing has not yet started seriously, so any flushed-over ice will take longer to re-appear on the SSMI satellite instruments. Second, the ice pack in the Western Arctic was already heavily fragmented even before this storm even started. There were plenty of polynia and areas where ice concentration was no more than some 50 %. This means that the "flash melt" area starts to behave much more like "open ocean" than an area with a semi-solid ice pack.
But even worse, this storm seems to have stirred up not just the upper halocline at 20-75 meters, but even the lower halocline at a 200-500 meter depth. Two separate ITP buoys, separated by a few hundred km, record this disturbance so it is not just a local event :
Preliminary, automated, uninterpreted buoy data from the Arctic ocean area affected by the huge storm.
Several studies show that the upper halocline can be disturbed down to some 50 meters, and these are great reads :
but I've not seen anything like this disturbance down to 500 meter over a wide area ever before in the ITP records.
What does this mean ? Well, the stratification layers in the West Arctic in the "flash melt" zone seem to be completely eliminated for the moment. This brings up 'warm' and more importantly, 'salty' water to the surface (and cool, fresh water downward). The saline anomaly at the surface is disturbing : it increased from 25 psu to about 31.5 psu over the past couple of days. That means that the melting temp of sea ice just reduced by 0.5 C, and this is even without counting the increase in water temperature towards the surface that the stirring down to 500 meters causes.
It's hard to estimate how much ice melt this will cause, but since the storm still is causing significant water movement along the ice (ITP53 reports rock-and-rolls of 0.8 m/sec) as well as up/down welling, and because the crumbled ice above has a large surface area to absorb the heat, melting may resemble sugar in a stirred cup of tea right now.
The real important question is, how long will this de-stratified state persist ? The short answer is that we simply don't know. Ocean models (even without ice cover) are simply not yet capable of simulating deep ocean turbulence effects after a storm. Here is a good read :
So after all is said and done, it seems that Neven was right and we need to wait out this storm to assess what it's long-term effects really will be. But at the same time, the de-stratification this storm has caused down to 500 meter depth and the mixing of salinity and heat content throughout that column over the area of this storm (about 1 million km^2) will most certainly have caused very significant damage to the already fragmented ice pack, to the point that it become hard to believe
that the entire area or even part of it will somehow magically re-appear on the satellite record.
I think the "flash melt" area is knocked out for the season, and on top of that this storm will leave a lot of salty water behind at the surface, which will eliminate small pockets of ice remaining and make it harder for the area to freeze up in fall.
What do you guys think ?
Posted by: Rob Dekker | August 10, 2012 at 10:12