NSIDC calls maximum extent

The National Snow and Ice Data Center in Boulder, Colorado, has called it:

On March 18, 2012 Arctic sea ice likely reached its maximum extent for the year, at 15.24 million square kilometers (5.88 million square miles). The maximum extent was 614,000 square kilometers (237,000 square miles) below the 1979 to 2000 average of 15.86 million square kilometers (6.12 million square miles). The maximum occurred this year 12 days later than the 1979 to 2000 average date of March 6.

This year’s maximum ice extent was the ninth lowest in the satellite record, slightly higher than the 2008 maximum (15.24 million square kilometers or 5.88 million square miles) Last year, 2011, was the lowest maximum on record, 14.64 million square kilometers (5.65 million square miles). Including this year, the nine years from 2004 to 2012 are the nine lowest maximums in the satellite record.

Read the whole thing here. The NSIDC Icelights blog also has an entry on the maximum and what it means.

Update: The NASA Earth Observatory website has an article about the maximum. I'm copying the following quote by Walt Meier:

On average, Arctic sea ice has historically peaked around March 6, but the maximum extent has tended to occur later in the month in recent years. The cause for the later peak is unknown, but NSIDC’s Walt Meier suspects it might be related to the minimum sea ice extents that occur each September.

“There are constraints on how long Arctic sea ice can keep growing in late March,” Meier says, citing springtime sunlight and rising temperatures. “But since Arctic sea ice has melted so much in the summers, it could be that the ice has more room to grow at the end of the season.”

Comments

[Account Deleted]

It's non-sense to get interested in ice extent this much. I am calling it for the enstitues researchers not for the blog.

We have try to understand the charecteristic behaviour of the Arctic sea ice. It's weakining at the core. losing thichness. And i think it could not support ice extent this much. It ll step back and try to set another foothold . It wouldnt be surprising to see lowest record this summer.

[Hans Verbeek]

"the maximum extent has tended to occur later in the month in recent years. The cause for the later peak is unknown, "
Interesting. Looks like a negative feedback.

[crandles]

Could it be due to later minimums or rather later before peripheral seasonally ice covered ocean has given up heat and started to freeze? So less time for ice to grow so it is still growing by the time the maximum was previously reached.

Consider following as graph of ice thickness against time with progressively later start of freeze season:

http://farm8.staticflickr.com/7064/7020589229_3dc291cda1.jpg

Perhaps, this needs the vertical line to be at a slight angle.

[DrTskoul]

Hans, it is not necessarily a negative feedback.

The arctic night conditions are such that a certain ice extent will be created. It also seems to me that there is a quantity that at maximum area extent it reaches zero, if that happens earlier in the winter, we reach the maximum earlier. In recent years due to the late freeze-up we reach "zero" in that quantity later and the maximum appears later - no later than last week of march, since the sun comes up and the warming up of the arctic spring starts.

I understand that this is lots of handwaving, however it seems to me that the september - december conditions will dictate the end - freeze thickness and extent of the arctic ocean. The earlier the freeze starts the thicker the ice and the larger the extent. We will not see a big change in the maximum state until much later.

This from someone novice in atmospheric/oceanic physics and climate science....

[crandles]

Last graph was apparently confusing, try again:

[Kevin McKinney]

"...since Arctic sea ice has melted so much in the summers, it could be that the ice has more room to grow at the end of the season."

I think that will likely turn out to be the explanation. After all, maximum extents are declining, too.

If this turns out to be the case, it will mean that the shape of the annual cycle shifts, skewing 'to the right.'

[Neven]

Interesting. Looks like a negative feedback.

Well, like it says in this NewScientist article:

Up until 2007, sea ice systematically fluctuated between extensive cover in winter and lower cover in summer. But since then, says Lenton, the difference between winter and summer ice cover has been a million square kilometres greater than it was before, as a result of unprecedented summer melting. These observations are in contrast to what models predict should have happened.

So, if this melting season the sea ice area goes low again, we might see the biggest drop from max to min in one single season (although 2008 will be hard to beat). I believe that in systems theory wild swings often are evidence of a system transitioning from on state to another.

[Rlkittiwake]

I would guess that perhaps one of the factors driving a later peak is that the heating of the sea water during our dramatic summer melts is creating a heat pool that is taking more time to dissipate, which in turn is pushing the peak forward.

However, this wouldn't explain why in previous years the power of the sun was able to begin the melt earlier in the year. It seems like more heat in the system would promote earlier melting through the increased water vapor and subsequent heat-trapping ability.

[crandles]

There may be many possible explanations.

One simple one might be the ice edge retreat northwards. This means that the sun's power is weaker at the more northerly edge so area is still being gained until a few days later when the sun rises higher (and is sufficiently high for longer) when the area losses balance out the area gains.

[Kevin McKinney]

That's good, crandles. Don't know if it's right or not, but definitely a good thought.

I haven't been paying close attention to the geographical distribution of sea ice, other than the general Kara-clear, Okhotsk-not generalization. Does the late-season growth this year follow a north-south gradient discernibly? (Of course, one year's pattern isn't likely to tell the whole tale anyway, I suppose.)

[Chris Reynolds]

Neven,

Alternatively - the eratic behaviour could have nothing to do with approaching a tipping point, it may be epiphenomenal of the thinness of the sea-ice. The lack of thick multi-year ice that stabiises the sea-ice by reducing the open water formation efficiency OWFE. OWFE is a metric of the ease by which a given volume loss can create open water; thick ice has to lose much more volume before it exposes open water, thinner ice can expose open water with a lower volume loss. So a weather event that causes open water now would not have caused open water in the past.

This issue of the later peak of sea ice is interesting. FWIW I think Crandles explanation above is the most persuasive. This is a issue worth pondering some more though.

[Chris Reynolds]

Regarding the Tim Lenton article on New Scientist - anyone got more info - preprints of paper / powerpoint or pdfs of presentation?

I can't find anything substantive at the Planet under Presure website.

TIA

[Rlkittiwake]

Crandles, I like this idea that it's all latitude that's moving the peak, but it seems too simple. Since it's simple, it should be easy to verify if this is the case.

Under your model, more southerly areas like the Bering Sea and Hudson Bay should generally start melting first while areas like the Barents Sea and the Greenland Sea are still gaining ice, but is this really how it happens? I took a scowl at the CT sea ice graphs, but I couldn't discern anything from those.

Also, which areas used to have sea ice at the maximum that no longer have as much sea ice? Based on the last two years of data from CT, the Bering Sea has hit at or above normal and Hudson Bay has hit normal even though last year was the worst year for maximum sea ice in recorded history. Meanwhile, last year the Greenland Sea hit normal while the Barents Sea was not doing well at all. So the seas that went further south had more ice and the sea that was furthest north had less ice.

(These questions aren't rhetorical; I am genuinely curious what someone with a better knowledge base and/or data set can make of it.)

Even though I am certainly not a sea ice scholar and I am lacking enough data to support my questions in anything approaching a scientific manner, I don't think that the supposed absence of low-latitude sea ice is telling the whole story. It's not just the sun "fighting" against latitude, but also the influences of land temperatures, sea temperatures, and local weather patterns.

[crandles]

I don't think my latitude explanation requires the ice to form a perfect circle around the pole. Heat flows are also relevant so the furthest north ice edge is where warm atlantic water approaches arctic. Land cools down faster than water so areas like Baffin Bering & Okhotsk that have lots of ice around them have ice edges that are further south.

It seems possible that ice edge in Barents reaches further south after the maximum extent with area gained there after maximum being more than offset by losses further south in Baffin, Bering & Okhotsk. I am not sure if we have appropriate information to test that as it wouldn't happen like that every year due to weather. Even if there does seem to be an effect like that, it could be due to seasonal variation in the other heat flows like volume and temperature of Atlantic water rather than due to the latitude effect being dominant.

So there could well be other explanations and testing and deciding what the dominant effect is does not look easy to me.

[Werther]

The Arctic isn’t a homogenous environment…
While analysing this years’ progress, you have to keep in mind the above. IMO the dominant factor last season was a lack of winter north of 80 degrees, combined with strong Atlantic input. The anomalously low Baffin SST’s relate to the lagging effect of the Atlantic Multidecadal Oscillation, having first peaked there 2010-2011, peaking now in the Barentsz-Laptev region. It could well be over in the next few months (having done it’s damage...).
An extremely negative PDO is contributing to anomalously cold SST’s in the Bering-Ochotsk region. Together, the weather-aspect of this is a thin but large SIE.
The configuration might alter considerably while the AO/ENSO coupling trades places/changes phase.
That will express itself in new, exciting or frightening weather features over the Northern Hemisphere (assuming that HEM is much more prone to change than the buffering Antarctic/southern HEM).
Through these weather-effects, I see no direct reason to expect an anomalous low SIE/SIA season 2012. The trend will present another close tie with 2007/2011, accent on the Atlantic side. And a spectacular reel down through spring.
The real alarming part could be methane/permafrost... if continued low SIE isn’t interesting enough.

[Andrew Xnn]

Walt Meier is stating that the ice is only 10 to 30 cm thick at most (4 to 12 inches)!

http://earthobservatory.nasa.gov/IOTD/view.php?id=77513

[Neven]

Walt Meier is stating that the ice is only 10 to 30 cm thick at most (4 to 12 inches)!

On the edges, that is. As this is being misinterpreted here and there already, I'm predicting Mr Watts will have a nice 'don't look there' addition for his next pre-season sea ice news article.:-)

[adelady]

On the edges, that is.

I read it as being all of the late growth in extent. I suppose that is pretty well the edges, but a couple of areas filled in a bit more that look to be pretty vulnerable if he's right about all of it.

[A Facebook User]

While I agree with most of the opinions seen in this thread, I do think calling the current level”the 9th lowest’ leaves supporters of climate change open to allegations of bias. After all if we express results in terms of league tables the 9th lowest can also be expressed as the highest in a given set of 9 years. The ice continues to melt, league tables distract from that fact.

[idunno]

Big rebound today to 13.636 - just 0.037 below the maximum.

Satellite images are now showing the whole Arctic.

[idunno]

Sorry - I meant 0.063 below maximum.