Cryosphere Today sea ice area: maximum reached

I finally feel confident enough to call the maximum for the Cryosphere Today sea ice area data set*:

graph comes from the Pogoda i Klimat blog

More than two weeks have passed since I wrote the following:

ECMWF is showing a big high taking over the Barents/Kara regions 5 from days from now, with a big low doing its thing in the Bering/Okhotsk region. If this forecast comes about I think we will have our CT SIA maximum around March 5th.

It came about, but didn't last, and so I decided to hold on to my horses:

I'm 75% sure this will be the maximum, but I'm seeing some things in the ECMWF weather forecast maps that could make for a later maximum. Too early to tell now.

Things were a bit uncertain for a couple of days. A high of 13,680,472 km2 had been reached on March 6th, followed by a few small downticks in a row, but those downticks could still easily be overruled by a big uptick, many of which we saw in the past weeks. The weather can be very capricious at this time of the year, and quickly add hundreds of thousands of square kilometers at the fringes of the Arctic sea ice pack.

But after downticks of 60K and 51K in the last two days, looking at various weather forecast maps, and considering the fact that most regions have an anomalously high area, I think it's safe to say the CT maximum SIA has been reached on March 6th. So my first guess was pretty close, although it can hardly be called rocket science, as most years in the period 2005-2012 reached their winter season maximum around this time (in million km2):

• March 6th 2005: 13.46
• March 11th 2006: 13.36
• Febr. 26th 2007: 13.32
• March 11th 2008: 13.89
• March 2nd 2009: 13.85
• March 7th 2010: 13.81
• March 8th 2011: 13.14
• March 6th 2012: 13.68

Here's a visual version of that, one of Larry Hamilton's excellent bar graphs:

This good guess of mine is totally eclipsed by my prediction in the 2012 Maximum Area Pool. I said CT SIA would end up between 13.0 and 13.2 million square km. I did indicate at the time it was a bit of a gamble, but also didn't think the maximum would go over 13.5 million square km. Well, ahem, it did. Big time. More on the poll in a later post (sneak preview: almost everybody was wrong, but some more wrong than others).

The freezing season is over.The fat lady is on her way to the concert hall, where she will take her time getting dressed, flexing her vocal muscles, and eat a basket of chicken wings, before giving a concert at the end of the melting season that will now slowly get underway. A new year, a new melting season. Tempus fugit.

---

* This data set currently gets more attention on the Arctic Sea Ice blog, since it's the only dataset that is updated every day and has easy access (view/download data here) that makes it possible to compare to previous years. It's the only dataset left that does this ever since the demise of AMSR-E and the IJIS SIE data that went with it.

Comments

[Neven]

Larry Hamilton also has a bar graph up for the max DMI extent.

[Arne Perschel]

Looks like we'll have to tone down next time Watts comes up with poll results from lala-land.
We were unanimously pessimistic (my guess was a cautious 13.4-13.6).
Over 96% of poll participants got the maximum area wrong on the small side.

Yes, it's thin ice, yes, it'll be gone in 5-15 years. But still, let's continue to be the true skeptics: skeptical with ourselves!

[crandles]

It can take quite a while to be absolutely sure.

For example, 2003 showed this:
2003.1973 -0.2748360 13.8126793
2003.2219 0.3708031 14.3390617
An increase of 526k in 9 days all subsequent to the last data date. It does seem fairly unlikely that there will be a 149k increase - only 2003 of last 10 years achieved that increase from the date of the last data. However, another 4 years did manage over 100k increases. So it does need some weather interpretation to get the probability down to well below a 10% possibility of a new max.

If a mew max was going to occur, we would likely see something dramatic in weather forecasts to cause large ice area increases. Instead I see nothing much until days 5 and 6 when a low could reduce Okhotsz ice area and KARA looks set so it is likely to become less than 50% covered.

So your calling the maximum looks fairly safe to me, though AFAICT it is hard to eliminate a couple of percent chance. What sort of probability would you put on it? 98%? 99% 99.5%? Also what probability do you need to call the maximum?

[Neven]

For me a 90% probability would do, but how do you quantify that? Right now, based on current weather forecasts, I say the chance of the 6th being the max is over 95%.

Like you say, the next 4-5 days we probably won't see an increase in CT SIA numbers. Let's say the difference with the current max stays the same, in 5 days it's March 20th. What are the odds of a 150K+ increase after that date? My gut feeling says fairly low odds.

[Neven]

Looks like we'll have to tone down next time Watts comes up with poll results from lala-land.

As long as we don't submit the results to SEARCH it won't become too embarrassing. :-P

But 'we' better be more conservative at the end of the melting season. Polls are after all a kind of wishful thinking magnets.

[Account Deleted]

I guess the issue is how quickly that last few hundred thosand m^2 gained in the last week or so will disappear.
I suspect fairly quickly and we will be back to the situation where we have larger than average amounts of the arctic ocean receiving sunlight for a longer period.

[Account Deleted]

Sorry some of the keys on my keyboard are sticking. That should have been km^2

[Apocalypse4Real]

I'd suggest we have a 50% chance of bouncing 100 km2 higher, given the forecasted increase in positive Arctic Oscillation over the next 7-10 days. The last SIA increase seems to correlate with the change in Oscillation.

Time will tell - although I hope Neven's right.

[Jim_pettit]

That 2003 rebound up to day 82 was definitely flukish. In fact, only eight of the 33 seasons from 1979 to 2011 had an ice maximum on or after Day 75 (which this year is today, March 15), so, probability-wise, I'm about 85% confident that 2012 is beyond the peak.

[L. Hamilton]

DMI extent for 3/14 shows a 200k decline since 3/11, or 250k below the high point on 3/5.

[Philiponfire]

any yet the Kara sea is icing over again.

[Bob Wallace]

The regional map shows the Kara topping out on March 2 and falling off a bit after that.

[Rlkittiwake]

I'm always curious this time of year about how many of these supposed upticks are actually the earliest stages of fragmentation and export.

[Phil263]

But still, let's continue to be the true skeptics: skeptical with ourselves!

Arne, totally in agreement with you on this. Rigour and true skepticism is what should make us different from the Wattsup crowd and those who would not hesitate to manufacture evidence or manipulate data to push through their own agenda.
We 're all passionate about these issues but must keep a " cool" head :-)

[crandles]

>"That 2003 rebound up to day 82 was definitely flukish."

Of 34 years here are the top 10 increases from date YYYY.2 (.2=latest data at time of posting)

0.4986028 1985
0.4723453 2003
0.3902941 1997
0.2576638 1984
0.2289477 2001
0.19979 1999
0.1987982 2009
0.1971932 2004
0.1740398 1991
0.1696053 1992

Currently 235k below max so far, so 4 in 34 years would get a higher max. That is slightly higher than the 1 in 10 years so I think the 1 in 10 wasn't far off or was too low rather than dismissing one occurance as flukish.

The weather forecasts may allow us to push the probability of having already passed the maximum to over 90%. However I am thinking there is still possibility of Kara icing over again (rather than thinking likely to become half uncovered). So I am thinking it is still be somewhere around 95% despite the further decrease in latest data.

[Neven]

The weather could even be causing another uptick. I don't see any change in Baffin, unless the Sun comes into play there. Like you say, crandles, that open water in the Kara Sea is getting an itsybitsy layer of ice again. Bering/Okhotsk could be under threat of strong compacting winds though in about 3-4 days.

[idunno]

Hi Neven, Crandles et al,

76k lost today - now 240k below maximum.

[Phil263]

76k lost today - now 240k below maximum.

Yes but big uptick on DMI ! I find that DMI is a good predictor. Expect an uptick in SIA on CT tomorrow...

[L. Hamilton]

Yes, DMI jumped almost +300k, twice the size of the second-largest March increase in their data (and incidentally setting a new high for this year). Has anybody seen +300 on the maps?

[Neven]

Perhaps if you count the nilas in the western Kara Sea. No spectacular expansion elsewhere IMO.

[Chris Biscan]

There is at least 500K to 1 million square kilometers of deceivingly thin ice from the Kara to Laptev, greenland sea and barents.

this is pretty on extent graphs but will melt fast regardless.

[FrankD]

Agreed, Chris - the conditions in Kara & Laptev will give comforting numbers on extent for now, but which probably won't hold up in the long term.

But there is a flip side to that coin. It appears that a decent ice arch has formed at the bottom end of Nares Strait. If that arch holds for any length of time, Baffin extent will start to drop quickly as ice below that arch is cleared from the Bay. While that looks bad on extent numbers in the short term, it's actually good news overall, if it blocks ice from being exported from the central basin.

A certain fixation with nurmerical data is understandable, but the quantitative always needs to be viewed through a qualitative prism.

[Jim Pettit]

"That is slightly higher than the 1 in 10 years so I think the 1 in 10 wasn't far off or was too low rather than dismissing one occurance as flukish."

I see what you're saying. But my point--which I failed to put across clearly and without adequate substantiation--was that the large growth of ice up to day 82 in 2003 was flukish due to the amount of overall ice that was added. Using your table of the top ten SIA increases after day 75 (YYYY.2), I've added a third column to show what percentage of each year's maximum area was added after day 75. (The asterisks indicate years in which the increase occurred after maximum had been reached. That is, the post-day 75 increase failed to result in a new maximum.) As you can see, the amount of overall ice that was added after day 75 was abnormally large (>3%) in both 1985 and 2003--thus, my use of the term "flukish".

0.4986028 1985 3.42%
0.4723453 2003 3.29%
0.3902941 1997 2.72% *
0.2576638 1984 1.81%
0.2289477 2001 1.58% *
0.1987982 2009 1.49% *
0.1971932 2004 1.43% *
0.1997900 1999 1.4%
0.1696053 1992 1.18% *
0.1740398 1991 1.15%
-----------------------------
0.2358441 2012 1.72%

I added the 2012 line at the bottom to show how much SIA would have to increase in the next week or so in order to surpass the YTD high on day 66. As you, Neven, and others have said, it certainly seems possible that we could add enough this year to set a new maximum--though we'd have to see another 450,000 km2 added this year to achieve the same percentage added after day 75 in the "flukish" year of 2003, and I really don't see that happening. ;-)

[Kris]

Neven wrote:

No spectacular expansion elsewhere.

Right!

Meanwhile, do have a look at the 22 March ecmwf Europe chart.

If this forecast would be confirmed, subtropical air masses will be squeezed directly to the Nordkapp and the Barents Sea region again, just as in the 7 °C record day at the 8th of February.

[Neven]

Kris, I followed your advice and had a look at the ECMWF forecast maps. I'm also seeing that in the next 4-5 days Okhotsk is going to get a huge beating. I expect SIA to drop very fast there the coming week. Probably a bit in Bering as well.

[Neven]

Crandles wrote in the March open thread:

Now only 51k above max so far and 4 of last 10 (or 17 of 33) increases from YYYY.2028 to subsequent max were more than 51k.

A new max looks quite likely from the figures only. OTOH Okhotsk looks like it is going to get some strong warm winds from south that will melt and compact ice.

Wow, increases of 112K and 68K. This is going to be close. I don't know how much of that Barentsz/Kara ice growth is still in the pipeline, but Okhotsk (and maybe Bering) should start to go down from today onwards.

[crandles]

I wrote "only 51k above max so far"
Oops!

[crandles]

Also, it was increase from YYYY.2083.

[Neven]

I'm seeing increases in practically all regions on the Uni Bremen SIC maps from the 18th to the 19th. If CT sees it the same way, this could very well make for a higher maximum.

[Darren Wood]

Hi how accurate are the "realtime" images from http://www.fourmilab.ch/cgi-bin/Earth ?

If they are acurate there seems to be a big difference between what's seen on the images here and the sea ice area from cryosphere today. appreciate your patience as I'm a complete layman on this a most interesting topic.

[Neven]

Hi Darren, welcome.

Those images of the Arctic regions do not look accurate at all. I have no idea from which year it is, but it is definitely a couple of months later than now.

[Peter Ellis]

It's not updated at all. That site is a case of "Show me what bit of the Earth the Space Station is currently looking down on" - the location of the satellite viewport is updated in real time, but the picture of Earth is not, it's simply a static model.

[idunno]

Hi Darren,

Clicking the "Daily Graphs" tag above gets you to direct links to two sites - "Arctic Mosaic" and "arcticio" with near-real-time satellite images.

[Darren Wood]

Thanks guys much appreciated.

Now I know the context I will refer to the daily graphs in future.

Cheers :)

[Apocalypse4Real]

Neven,

Given the jump in AO, my earlier assumption that we might get a new SIA max, still looks a good possibility. Today through the 22nd is my guess for th enew max date.

Thanks again for all your work on this site.

[Neven]

A4R, I knew there would be an uptick, but I'm surprised at the magnitude. So even if the old max holds, I feel like I was wrong to call it so soon. Well, you know what they say about assumption.

[Phil263]

CT SIA seems to be stuck on xxxx2083 (Day 76= March 16th) showing an area of 13.629. Given the recent movemements on DMI, I guess we should now have passed the old maximum of 13.6804.

[Jim_pettit]

Much to my astonishment--though I shouldn't be astonished, I suppose--a new CT SIA maximum has been reached.

On day 79--Monday--SIA dropped by 146K km2. But on Tuesday, Day 80--it gained 217K km2 to reach a new peak of 13,700,674 km2. (42K was lost yesterday.)

SIA is now 883K above where it was on this date last year, and 735K above where it was in 2007.

[Bob Wallace]

The 2010 freeze season gave us an even later 'quick freeze'. And it melted off fairly quickly, returning the extent/area lines to the recent years pack.

Just wondering if anyone has tied this weather-blip freeze to the incredible 'summer in the spring' temperatures that are hitting the middle of North America?