Arctic Sea Ice

Freezing season has started, or has it?

We tend to focus on the end of the melting season, also known as the minimum, because that's when the ice covers the least amount of water. That's all fine and dandy, but of course, it's just an arbitrary measuring point. For instance, this year's minimum for the JAXA sea ice extent data set was 6th lowest on record, but flash forward to the first week of October and this year's trend line is third lowest on record and could soon be second lowest on record:

This, as always, had to do with the weather, which has been pretty crazy the past two weeks. First, fierce winds pushed back against the ice edge in most of the Arctic Ocean, delaying the ice pack's expansion, while at the same time continuing to melt that last, stubborn sliver of ice extending towards the East Siberian Sea. The result has been a continuing decline of sea ice extent in the Arctic Basin, and to a lesser extent in the East Siberian Sea, where, normally, the trend line should have been going up already (image courtesy of Wipneus, ASIG Regional graphs):

Another factor that has come into play these past few days, is heat. It's mostly coming from the Pacific, where sea surface temperatures are running quite hot again, as can be seen on these DMI sea surface temperature anomaly maps, comparing this year's start of October with those of 2016 and 2017:

Bering goes extreme

The melting season hasn't started in earnest yet, but it seems the Bering Sea hasn't received the memo. For almost the entire winter, sea ice has been reluctant to form there, and now that the Sun has returned, the ice edge has started to retreat to record high latitudes, past the Bering Strait all the way up into the Chukchi Sea. Here's how that looks on Wipneus' regional graph:

To emphasize how truly exceptional this is, here's a comparison with the situation in all other years from the 2006-2018 period (images retrieved from the University Bremen sea ice concentration maps page on the ASIG, click for a larger version):

Now, if this was it, other years would maybe catch up in the next few weeks, but this isn't it. The coupled HYCOM-CICE model from the Naval Research Laboratory forecasts the sea ice to continue to drift northwards, until the end of the month at least:

Global sea ice records broken (yet again)

This is starting to become an annual tradition, here on the Arctic Sea Ice Blog. Just like last year and the year before that, the record for lowest global sea ice minimum has been broken. The lowest global sea ice extent minimum on record (NSIDC and JAXA) was reached two weeks ago. I wanted to wait a bit and see whether the record for NSIDC global sea ice area would be broken too, but it wasn't, although the minimum came in at a solid second place, practically on a par with last year.

Here's the graph, as provided by Wipneus:

And here you see the anomaly of all those trend lines in chronological order, having hovered between 2 and 4 σ (standard deviations) since the 2016 crash:

Not enough Antarctic sea ice melted to clinch the lowest Antarctic sea ice minimum record (coming in second lowest 'only'), and so the decisive nudge was given by the sea ice in the Arctic, refusing to freeze along some of the edges, most notably in the Bering Sea. Here's how the graph for that region looks as of today, with some of the open water even spilling into the adjacent Chukchi Sea, which is highly unusual for this time of year:

Beaufort under early pressure

A high-pressure area has been moving in over the Beaufort Sea lately. It's quite big and quite high in pressure. Environment Canada has the current pressure there at 1042 hPa:

I've coloured the land masses to make it easier to orient. I've also put in the red arrows to show which way the winds blow along the so-called isobars.

This is the second notable high of this year, following another big one in February that produced strong winds that pushed the ice away from the coast. This phenomenon called the Beaufort Gyre resulted in what we call a cracking event around here (see this blog post). Of course, the Arctic was plenty cold around that time and so the leads froze over again, but the marks are still visible on this AVHHR image (provided again by Environment Canada):

On the one hand those leads let heat escape from the ocean water, on the other hand the thin ice between the enormous floes will melt out easily as soon as sunlight and higher temperatures start taking over, possibly making the entire zone more vulnerable early in the melting season.

The reason I've decided to dedicate a blog post to this, has to do with the current ECMWF forecast for the coming 6 days:

Global sea ice area record minimum?

This post is about global sea ice area, the simple addition of Arctic and Antarctic sea ice area.
As a statistic it's somewhat interesting, but it doesn't convey all that much information about the individual health of both polar regions,
let alone their sub-regions.

Despite this fact, or probably because of it, it's often being touted by climate risk deniers as proof that global warming is a scam and all is well, because Arctic sea ice loss is compensated by a growth in Antarctic sea ice. Even if this were true - it isn't, as this Skeptical Science article explains - it's like saying there is no hunger in the world because there are so many obese people.

But anyway, we're approaching that time of year when global sea ice area as calculated by and presented at Cryosphere Today is going to hit its minimum, or lowest amount of sea ice cover. And currently the number is quite low, as can be seen on this graph from Piotr Djaków's Pogoda i Klimat website:

According to the data 2016 is already 4th lowest at 14.73 million km2, just behind 2007's minimum, and almost 350K behind 2011 and record holder 2006 (14.39 million km2, the grey trend line just below the 2011 green trend line). Below I'll discuss the factors that will determine whether a new record is in the books.

An absolute prerequisite for a new record is for Antarctic sea ice area to go low. Whereas it's freezing in the Arctic right now, it's summer in the Southern Hemisphere and so the Antarctic melting season is slowly moving towards its apogee. If we look at the NSIDC sea ice extent graph for March, for instance, we can see some fairly large swings, with record years 2006 and 2011 standing out with their distinct dips:

Early record, late record

10 days ago I posted a blog post called Mad max. The title - referring to a very early maximum sea ice extent - ended with a question mark, because it was far from sure whether the preliminary max reached on February 15th would remain standing. This is because of the oscillatory nature of the final phase of the freezing season, where weather conditions can cause the ice pack to expand very rapidly in one or more of the fringe regions.

Here's how things currently look on the JAXA sea ice extent graph:Edit March 18th - A commenter sent me this updated graph today, with a correction (2010 had the latest max on record, not 2014):

Since posting that blog post, JAXA SIE dropped up to 318 thousand square kilometres below the preliminary maximum, but has since crept upwards again. The current difference is 186 thousand square kilometres, which still is quite significant. In fact, a new max becomes less and less probable as time wears on, with the Sun creeping higher in the sky, shining on larger parts of the ice pack every day.

At the same time, with extent relatively low, there's plenty of freezing potential on the fringes. And we're talking about the Arctic here. It has a reputation of crazy swings to uphold. Here's a map with the names of the regions that have a final say in where the maximum ends up, with the orange line showing the average ice pack edge for the 2000's:

And here are the graphs for each of these regions, produced by the prolific Wipneus, that show what has been happening regionally in the past weeks, with the purple line showing the JAXA SIE data for 2015.

First the regions with freezing potential:

ASI 2014 update 1: melt pond May

During the melting season I'm writing (bi-)weekly updates on the current situation with regards to Arctic sea ice (ASI). Central to these updates are the daily Cryosphere Today sea ice area (SIA) and IJIS sea ice extent (SIE) numbers, which I compare to data from the 2005-2013 period (NSIDC has a good explanation of sea ice extent and area in their FAQ). I also look at other things like regional sea ice area, compactness, temperature and weather forecasts, anything of particular interest.

The animation on the right consists of NSIDC
sea ice concentration maps, one for each ASI update.

Check out the Arctic Sea Ice Graphs website (ASIG)
for daily updated graphs, maps, live webcam images and
the Arctic Sea Ice Forum (ASIF) for detailed discussions.

May 19th 2014

Welcome to the first update of the 2014 melting season.

There are two things that determine the outcome of a melting season: initial ice state and subsequent weather conditions. The last two melting seasons were so diametrically opposed that we actually learned a lot in this respect. Or, at least I did.

In 2012 the initial ice condition and first half of the melting season was so conducive to massive sea ice melt, that it didn't matter all that much what the weather did during the second half of the season. Even during cloudy, colder periods sea ice extent and area kept dropping. In 2013 it was the opposite: initial ice state (probably influenced by the cracking event of March that year) and a cold, cloudy first half of the melting season were slowing the sea ice melt so much that even spells of clear, warm weather couldn't change the melting season's outcome.

Of course, we don't know what the weather will be like weeks and months from now, but we can somewhat estimate the initial ice state (see my 2013/2014 Winter Analysis) and check the sea ice pack's pulse in these crucial first weeks of the melting season. One very important influence I've been obsessing over every now and then, is the amount of melt ponds on the sea ice. These small lakes on ice floes can greatly accelerate melting as they decrease albedo, meaning that more sunlight gets absorbed by the ice.

I've always known this was important, but have now become convinced that it could be of paramount importance, in the sense that it can perhaps not make, but definitely break a potential record melting season. And all of it happens in May. Melt pond May. And probably the first half of June too.

I'll be focusing on this a lot in weeks to come.

Sea ice area (SIA)

This is the time of year when the trend lines on various sea ice area and extent charts start to converge with differences becoming more prominent again in June. It can still convey some information, as we saw last year on the Cryosphere Today SIA graph. The 2013 trend line started the month out low, but was highest at the end of it. This was a sign (in hindsight, I didn't see it as such at the time) that cold and cloudy conditions were already having a big influence on the final outcome of the melting season. The slow SIA decline meant that there were less melt ponds than in years before.

Mind you, I'm not saying that SIA or SIE numbers in May are a proxy for melt pond cover fraction, but last year really jumps out in this respect. This year the trend line is more or less in the middle of the pack, as you can see on this graph based on the latest data:

Stronghold

I think we all remember what the conclusion was after the last freezing season (see 2011/2012 Winter analysis):
sea ice on the Atlantic/Siberian side of the Arctic looks vulnerable, sea ice on the Pacific/North American side should be thicker. So far there has been a very clear confirmation of the first half of this conclusion, in that the ice on the Atlantic side of the Arctic, along the Siberian coast, has retreated very fast, perhaps even faster than last year.
This comparison I made a couple of days back (for the Fringe Fries 2 blog post) shows it well. For a more recent comparison I refer you to the Concentration Maps page on the ASI Graphs website.

I now want to take a look at how the ice on the other side has been doing so far. The Pacific/North American side basically consists of three regions: the Canadian Archipelago, the Beaufort Sea and the Chukchi Sea. In last month's blog post concerning the Northwest Passage it became clear that the Canadian Archipelago was mostly filled with first-year ice, but because of a relatively cold freezing season the ice was probably a bit thicker. The ice in the Northwest Passage and the rest of the Archipelago has been holding up pretty well, better than in previous years (as confirmed by the MASIE extent graph), but due to high temperatures and quite a bit of insolation in the last couple of weeks, the NWP now seems to be breaking up from the inside out. As the fast ice in Nares Strait broke up 10 days ago and ice transport from the Arctic Basin is about to start, I don't think the ice in the NWP is going to hold out much longer.

Because the Chukchi and Beaufort Seas are neighbours, I have made this comparison of University of Bremen sea ice concentration maps on July 7th in the 2004-2012 period, showing both regions (click for a larger version):

Ocean heat flux

There are many factors involved in the current long-term decline of Arctic sea ice. From relatively small ones like river discharge, precipitation and soot, to bigger factors like atmospheric patterns and air temperatures.
The possibly biggest factor of all is also the
factor that is the most unknown and uncertain: ocean heat flux. How much heat are ocean currents bringing into the Arctic?

To quote the first paragraph of this 2011 article by Ron Kwok and Norbert Untersteiner: "The surplus heat needed to explain the loss of Arctic sea ice during the past few decades is on the order of 1 W/m2. Observing, attributing, and predicting such a small amount of energy remain daunting problems."

Ocean heat flux is what separates the aggressive and famous forecast by Wieslaw Maslowski (Arctic essentially ice-free by September 2016, ± 3 years) from more conservative estimates that predict the thing to happen between 2050 and the end of this century. As explained in this BBC article from 2007:

Fringe fries

With things looking so serious in the Arctic right now, it was high time for another corny pun!

Instead of doing three different posts, I'm combining three different comparisons of fringe regions for today's date. Of course, I'll be using the excellent University of Bremen sea ice concentration maps. If you prefer the holistic view, you can go to the concentration maps page on the Arctic Sea Ice Graphs site. A reminder from the UB webpage: "As we are using SSMIS data instead of AMSR-E data, the quality of the resulting maps has yet to be assessed." There are probably some differences between this year's SIC map and those of previous years - probably in colouration -, but IMO they're small enough to discard.

Let's start with Bering and Chukchi (click on the image for a larger version):

2012 seems to have made up the deficit and looks very similar to all other years except 2011. Which is not surprising as the Bering Sea is almost devoid of ice. Yes, all that ice that was used to mislead a couple of extra people here and there in some quarters, is now gone. On the MASIE chart we can see 2012 right in there with the other years, except for 2011 diving down: