Given another uptick and the current weather forecast, I'm ready to call the minimum for IJIS SIE V1 on September 12th at 5,000,313 km2. Apparently the high was too big and the pressure gradient too low to prolong things (see below).
This blog post should perhaps have been written last week, but better late than never, right? Although it could even be too late, as the minimum might have occured already in some data sets. But if there's one thing I did right on this blog so far, it's calling the IJIS sea ice extent minimum! So let's forget all the rest and keep doing what we're good at.
Pinpointing the minimum is a slightly tricky business, especially this year with all the thin ice that didn't melt out. On the one hand this causes a lot of compaction (and some last melting) potential, but this might be offset by faster freeze-up due to all the open water between floes. Either way, it's the atmospheric situation that determines when the minimum occurs in September.
I personally re-invented this wheel back in 2010, the first melting season that was covered on the Arctic Sea Ice Blog, when comparing the end of the melting season with previous ones in a series called End Zone. In the third instalment I covered the influence of atmospheric pressure on the timing of season's end. As I wrote at the time:
Atmospheric pressures determine where low-pressure areas (cyclones) and high-pressure areas (anti-cyclones) are situated. Along the edges of such systems winds are either blowing in an anti-clockwise fashion (cyclones) or a clockwise fashion (anti-cyclones). The magnitude of the pressure determines how strong these winds are.
The way these pressure areas are distributed over the Arctic are a big factor in the annual extent decrease. Especially now, at the end of the melting season. The Beaufort Gyre, the clockwise movement of the ice pack from the Pacific side of the Arctic towards the Atlantic side, is dependent on a high over the Beaufort Sea. Combined with a low over the Siberian coast, preferably between the Kara and Laptev Seas, the Transpolar Drift Stream is activated and transports a lot of ice through Fram Strait. The stable positioning of these two poles opposite each other is known as the Arctic Dipole Anomaly.
This is what you get when that happens:
The sea ice gets compacted towards the coasts of Greenland, Ellesmere Island and the rest of the Canadian Archipelago, and it gets transported towards Fram Strait where it melts out in warmer southern waters. A double whammy, as the American expression goes.
Globally speaking you could say that a high over the Beaufort Sea, combined with low SLP on the Siberian side of the Arctic, makes for compaction and prolongs the melting season (though the rate of decrease has slowed to a trickle), despite re-freezing already occurring in different regions of the ice pack. It's the pressure gradient between those highs and lows that counts.
To help comparing with previous years I have added yet another segment on the Arctic Sea Ice Graphs page (which has now reached perfection) under SLP Patterns. This segment offers an overview of sea level pressure patterns from May to September in 6-day averages, retrieved from the Daily Mean Composites page, as provided by NOAA's Earth System Research Laboratory, Physical Sciences Division. I'm hoping this will come in handy for myself and others in the coming melting seasons. I'll try and update regularly, but drop me a line when I haven't.
If we look at the SLP patterns in September, we quickly see what kind of set-up prolongs the melting season:This is for September 13-18 2007 and it shows a perfect set-up, with an intense high over the Beaufort and low pressure elsewhere. The melting season in 2007 ended on September 24th, the latest date in the 2005-2013 period.
The same goes for 2010. Here's the 6-day period covering September 13-18:
2010 was also a tad on the late side, with the minimum occurring on September 18th, getting it within reach of 2008's minimum, which was relatively early (September 9th) because of this set-up:The high and low-pressure areas are on the wrong side, causing a lack of compaction, and low air temperatures on the Siberian side of the Arctic where all the open water is. The same goes for 2011 which also had its minimum on September 9th:By the way, I'm basing myself on IJIS' version 1 of SIE data, which was revised a couple of days ago. I simply haven't had time to update my spreadsheet yet, but I will as soon as version 1 is terminated on September 30th, 2013. Version 2 is based on calibrated and validated AMSR2 data, in other words higher resolution.
So, let's have a look at what the weather has in store for the Arctic in the coming 6 days (weather forecast by ECMWF, click for a larger version):
There's a high over the Beaufort Sea alright, but perhaps it is too big, reaching all the way to the Siberian coast. High-pressure systems make for clear skies, and with temperatures dropping, this causes sea water to release its heat so that it can freeze up. If sea water in the Siberian seas freezes up, sea ice extent won't be going any lower. The pressure gradient over the ice pack is also lower, which means less wind to compact the ice pack.
This makes me tend to say that the minimum could be reached by Saturday, September 14th, with the caveat that a big low coming in from the Atlantic might invigorate the pressure gradient, causing more compaction where the weakest part of the ice pack is situated. Perhaps the hole we discussed previously might be blown even closer to the North Pole, making enough room for the US Navy's entire fleet to park there. This animation posted by Jim Hunt on the ASIF shows progression during September so far:
We'll have to wait and see how things develop, and how the ice pack reacts.
Some tragic news from the Arctic: Marc Thibault, commanding officer of the Canadian icebreaker Amundsen, helicopter pilot Daniel Dube and Klaus Hochheim, a veteran University of Manitoba Arctic scientist, died in a helicopter crash. May they rest in peace.