During the melting season I'm writing (bi-)weekly updates on the current situation with regards to Arctic sea ice (ASI). Because of issues with data based on the SSMIS sensor aboard DMSP satellites, I mainly focus on higher-resolution AMSR2 data from the Japan Aerospace Exploration Agency (JAXA), as reported on the Arctic Data archive System website. I also look at other things like regional sea ice area, compactness, temperature and weather forecasts, anything of particular interest.
June 11th 2016
This update could have easily borne the same title as update 1: Both sides. Two weeks ago the ice pack was shrinking on both sides of the Arctic, causing sea ice extent to drop to unprecedented daily lows. In the meantime, weather patterns have switched and the ice pack is now expanding on both sides!
Of course, a dispersing ice pack will cause extent loss to slow down, or even stall completely for a while. Especially if extent was extremely low to begin with. And so that enormous gap between this year and 2012 is getting smaller every day, fast. 2012 had perfect weather for melting during the first two weeks of June, and there was a large amount of thin ice that kept extent artificially high.
More importantly, these weather conditions set the tone for the rest of the 2012 melting season, as a large amount of melt ponds preconditioned the ice for big melts during July and August. This is where 2016 is falling behind now, even though it has been and still is lowest in many respects (extent, volume, snow cover).
2012 is punching hard in this round. Will it be enough to knock 2016 out?
The JAXA SIE graph perfectly shows what's going on:During the first 10 days of June, 2016 has only lost 24K on average every day. That's the lowest average in the 2007-2016 record, even 10K per day slower than last year, whereas 2012 lost a record 91K per day. But that was because of an impressive 7-day series of century breaks, right around this time.
With three days left to go in that series, and the gap with 2016 now being 377K (it was more than 1 million at the start of the month!), it will be exciting to see whether 2016 can hold onto that number 1 position. And if so, how long.
Over on the Arctic Sea Ice Forum, there's still three days left to vote on the JAXA daily minimum poll for June!
Cryosphere Today area per JAXA extent (CAJAX)
Cryosphere Today is still publishing faulty data from the DSMP F17 satellite that has been experiencing sensor problems since April, as they're probably waiting for the calibration process of data from the replacing F18 satellite to end. But, the NSIDC has been publishing new F18-based data (with the caveat that calibration is still ongoing) and as commenter Wipneus knows how to turn this data into CT SIA numbers, I've decided to use them to calculate the CAJAX percentage.
CAJAX - formerly known as CAPIE - is a crude compactness calculation using Cryosphere Today sea ice area data divided by JAXA sea ice extent data (which used to be known as IJIS, hence the name change). In fact, Wipneus told me I shouldn't call it compactness as the data comes from different sources. He publishes the real deal here, comparing Uni Hamburg (3.125 km resolution), JAXA (10 km resolution) and NSIDC (25 km resolution) compactness numbers.
Compactness ratios tell us something about how compact the ice pack is. The lower the percentage, the more water there is within the ice pack, either from leads or melt ponds. I'll explain this more thoroughly in an upcoming post.
2016 is among the lowest years.
Regional SIE and SIA
Map showing the names of all the different regions can be found here.
Wipneus also uses the highest resolution AMSR-2 data (3.125 km grid) provided by the University of Hamburg to create sea ice concentration maps of the Arctic (see here), and he is so kind as to send me maps showing the changes since the last update in one particular region or other. This time it's the Kara and Barentsz Seas (red means there's less ice now than there was two weeks ago, blue the opposite):
Ice extent has been going down fast in the Kara Sea, but things have levelled off in the Barentsz Sea (after going down really early, really fast), as can be seen on Wipneus' regional graphs as well:
It doesn't look like the entire North American coast will be free of ice before mid-June, as I was expecting three weeks ago.
What kind of weather conditions can cause the ice pack to disperse in such a way that it expands on both sides of the Arctic? The answer lies in this animation of Danish Meteorological Institute SLP images showing the distribution of atmospheric pressure during the past two weeks:
The high-pressure area that kept the Beaufort Gyre spinning was pushed back by cyclones causing increased winds and thus transport of ice towards the Atlantic. But at the same time another series of cyclone pushed the ice in the Beaufort Sea back towards the coast. That's how you get both sides to expand.
This ice drift map (Naval Research Laboratory) gives an idea of how that looks, with arrows pointing towards both the Pacific and Atlantic sides of the Arctic:
Another effect of low pressure taking over is cloudiness, and at this crucial stage of the melting season clouds blocking solar radiation dampen the preconditioning of ice through melt ponds. This is where 2012 laid the foundation for its record low minimum. 2016 was doing above average during May, but June is more important in this respect, as freezing temperatures vanish entirely from the Arctic.
Cyclones are going to continue to dominate the Central Arctic, pushing away the high pressure over the East Siberian Sea and Canadian Arctic Archipelago, where open skies and high temperatures caused massive melt ponding on the thicker, fast ice in the past 10 days. At the same time high-pressure areas are forming over the Kara-Laptev region, and the open waters in the Beaufort and Kara Seas could be receiving some more sunshine.
Either way, these weather conditions are the best possible scenario for sea ice retention, so we can expect the extent decrease to remain slow, maybe even to the point that 2016 is no longer in first position when the next update is due.
Here's an animation showing the GFS weather model actual temperature forecast maps for the coming week, as provided by Climate Reanalyzer:
Unlike the temperature forecast in the previous update there are no persistent heat waves to be seen that spill out from the coasts over the ice (darker green). In fact, some patches of blue - indicating freezing temperatures - keep popping up here and there. 2016 has been running very warm so far, but things look more average now.
To my delight, the Danish Meteorological Institute has put back its sea surface temperature anomaly map that I've been using for many years now. This will enable us to compare current maps with those from previous years (as I've done here on the Forum). Here's the latest from DMI:
Things are running quite hot in the North Pacific, hotter than in any other recent year, but the question is how much of this heat is going to enter the Arctic, and also whether it will influence weather conditions (for instance, by causing persistent cyclones to dominate). Over on the Atlantic, things are looking pretty toasty too, which explains why that expanding ice edge probably won't be reaching Svalbard, still ice-free, unlike any other year in the 2005-2016 period (except for 2006 at this point).
2016 has finally stopped being unprecedented. Changed weather conditions and less easy ice to melt have caused the rapid decline to take a breather, and it's uncertain whether that pace will be reached again, as this is a crucial period of preconditioning. Early melt ponds soak up a lot of extra solar energy, thinning the ice both from above and below. The effect becomes noticeable later on during the melting season, when melt rates are sustained despite adverse weather conditions. 2012 is a prime example of this melting momentum that perhaps didn't entirely cause, but certainly helped forge the record low minimum.
Now, there has been quite a bit of preconditioning over parts of the ice pack in the past week, notably over the thick ice in the East Siberian Sea, where open skies and a heat wave caused all the fast ice to turn blue and break up. CAJAX is also relatively low, indicating there's quite a bit of open water within the ice pack (leads and melt ponds). How much melt ponding there has been exactly so far, is unknown, but it seems there isn't going to be a lot of it in the coming week, given the temperatures and atmospheric pressure that weather models are forecasting.
There will still be 10 days left in June if things turn around, and as we saw last year, a freakishly warm July can compensate a lack of melting momentum. But as I said in the introduction: 2012 has the upper hand for the time being, and it will be difficult to catch it come September, if things stay the way they are for too long. Which is good for the ice, of course, and it wouldn't be much fun if a new record was a done deal already.
I'll have more on melting momentum after the weekend.