« More on ice thickness from AWI | Main | Daily NSIDC SIE data and graph »


Feed You can follow this conversation by subscribing to the comment feed for this post.


The unthinkable has become the mundane is just 4 years.

If this continues we will have front row seats to one of the most significant geophysical events of the past 12 000 years.

If we do lose permenant ice cover on the Arctic Ocean it will be something people will be taught about at schools hundreds of years after names like Reagan, Churchill, Hitler and Mao have been confined to very niche graduate history classes.


Is there a concise post somewhere that explains the different methodologies that Bremen, IJIS and NSIDC use?

Are these the only 3 groups tracking arctic ice? What does NCEP/NOAA do?

I think a post explaining all the different agencies and their basic methods would be useful.

Chris Biscan


I have recently done extensive work on this.

They are all the same relatively speaking.

Jaxa this year has all of it's flaws on display.

Follow that blog. Some Skeptics called me insane, then I shut them down something fierce.

The bias everywhere is sick.

Once I posted the historic stats between the two AMSR-R algorithms it was clear they were equally good.

The lack of knowledge and arrogence from the posters there making general ignorant assumptions on the Bremen map was pure bias.

Bremen is an accepted industry standard and clearly does better in 2011s situation.

I posted that stuff at 3 am. you think those folks haven't had time to retort?

No they read what I wrote and don't want to feel like even more jackass.

instead of saying once again Friv you did work while we insulted you, then we shut up. Will wait for you to have a bad prediction, then come insult you more and ignore your work.

they will do it and say nothing.

Read back about 10-15 pages and see how horribly bias folks are.


Looks like spring of 2012 is going to start out with a S**T-TON of first year ice...will be interesting next year for sure...

Janne Tuukkanen

Tropics are providing more low pressure systems to North Atlantic. Post-Tropical Katia is heading to Norway, and a week from now the remnants of now-tropical storm Maria might end up straight in the Denmark Strait.



That big low in the middle of the Arctic Basin is really spoiling the party. If things don't change, I think the melting season will end somewhere this week.

IJIS has a drop of just 6K.

michael sweet

The Canada Ice Service has been posting charts going all the way to near the north Pole recently to support the Healy. They show a lot of thin ice north of Canada. Ice Canada shows up to 50% ice where Bremen shows no ice. These areas can melt quickly if the weather changes or last until freeze-up. When Ice Canada starts to show new ice in the Chukchi Sea that means the fat lady is singing (new ice is designated X in the egg shell that describes the ice state). Last year it was only a few days after they first showed new ice when the sea ice extent started to grow. Today there is no new ice.


Given what you wrote about the weather forecast having changed again, I am a little surprised there is not more difference between


Forecast still looks bad for ice toward East Siberian Sea. Is that low going to go there?


I just had a look and the next couple of days that low is forecasted to move over the Canadian Archipelago.

I still haven't got used to the ACNFS ice displacement forecast. It doesn't really seem to match the ECMWF weather forecast maps well.

Ned Ward

That big low in the middle of the Arctic Basin is really spoiling the party. If things don't change, I think the melting season will end somewhere this week.

IJIS has a drop of just 6K.

Now revised to an increase of 0.938 K.

Seke Rob

MASIE opposed to JAXA has not gone in parking:

Date.......Total....Change sqkm
08-Sep-2011 4498286 +27066
09-Sep-2011 4490119 -8166
10-Sep-2011 4379016 -111104

Lucia (The Blackboard)

Seke Rob--
I haven't checked, every year, but I'm pretty sure JAXA extent has wobbled up and down before hitting the minimum every year it reported. If not, it's frequent enough. That's one of the reasons I picked a week average for betting. When the 7-day average turns up, we can be pretty confident the minimum is reached. But we're likely to see 2 or three wobble-ups and downs before the final JAXA minimum.

michael sweet

In Baffin Bay and the Canada Basin the Canada Ice service shows new ice forming today. That is the first new ice they have shown this year. It is starting to get cold again in the Arctic.

Chris Biscan

That ice came from the Arctic Basin.


It is from thin ice floes because of winds.


Temps with that high of Salinity need to be about -2C to form ice and not only that it needs to be about 100 meters deep as well.

buoy profiles in that area show a 0-100 meter mean temps of -0.7C

So I am not trying to be a jerk but the ice didn't form there.

I Ballantinegray1

The point is Chris that at those temps the ice may well melt there? There are a few peripheral areas that are currently 'ice free' and have temps that would cause melt if ice were to drift there. Not only an odd season for melt but an odd end of season too?


There are some fairly strong winds towards the Canadian archipelago. Anyone has a guess what that will do to the ice there?




It would push some multiyear ice between those channels.

I hope we get at least one clear satellite image before the melting season is done, so we can ascertain if next year 50% of the channels between the Queen Elizabeth Islands are filled with first year ice or not.

Paul Klemencic

Seke Rob, I agree, the MASIE has certainly not gone into parking. The MASIE extent shows three century breaks in the last six days reported up to September 11 !!! Either we are having an extraordinary September melt, or .... well, everyone knows my theory.

To be truthful, if the MASIE is being dated six days after the measurement period, then the latest century break in the MASIE came on September 4th. And that is a day earlier than I predicted, and a day earlier than the Bremen loss. So it doesn't appear a simple mistaken dating system error can explain this entirely. But we haven't seen three big declines like the MASIE data show, if the dates are correct.

I haven't heard back from NSIDC yet regarding this issue.

Peter Ellis

My theory, which tallies with the Northwest Passage stuff I posted a few days ago, is that MASIE is better at picking up very thin ice. That means it's been reading higher than other metrics to date, but is gradually catching up with them at the end of the season. Bremen seems to do the reverse, which may be why it reads lower than IJIS. This is already known to be one of the main differences between the SSM/I and AMSR-E sensors, explaining why NSIDC reads lower than IJIS.

We will be able to test this once the season turns. If you're right, then MASIE should stay lagging Bremen. If I'm right, then MASIE should lag during the melt and lead during the re-freeze.

L. Hamilton

Bremen dropped -46k for 9/12, now stands at 4.25m. That's slightly higher than the minimum of 4.24m recorded on 9/8.

Seke Rob

[OT] SLR drop explained from extreme deluges over past few years [rain and snow] (Visit SkS). With 1% of global water in soil and aquifers [massive depletions around the globe], who's surprised?

Ned Ward

My theory is that the MASIE data are very noisy. Yes, there are three "century breaks" in the past week. There are also two "anti-century-breaks" (extent increases > 100000). I don't think those noisy ups and downs are "real" (in the sense of reflecting real changes in the ice).

The individual days' maps from UB are also very noisy. Thus, I'd be cautious about applying Peter's test. With both data sets registering more or less random 100K ups-and-downs during a couple of weeks around the minimum, it may not be easy to infer the true date of the minimum, or to say that A reached its minimum before B.

Kevin O'Neill

University of California economist Brad Delong writes:

If the past decade has taught me anything, it has taught me that mistakes are avoided if you follow two rules:
1.Remember that Paul Krugman is right.
2.If your analysis leads you to conclude that Paul Krugman is wrong, refer to rule #1.

Today Paul Krugman writes:

So what’s an economist to do? Well, what I do is use both. Typically, I work things out first in terms of IS-LM, but then try to write down a stripped-down NK model with similar results; if I can’t make that work, it’s time for some hard thinking about where the difference lies. And sometimes this leads to insights that I don’t think would have come from either approach alone, like the crucial role of monetary credibility in making a liquidity trap possible.

The point is always to realize that the map is not the territory, but also to realize that IS-LM is a pretty good map, and that you can often fill in the gaps with a fancier—but not better, just different – NK model.

I hope the parallels between our different sources (UB, IJIS, NSIDC, MASIE) and different metrics (extent, area, volume) are obvious.

Paul Klemencic

Ned, I think you meant to say that MASIE has had two extent increase days of over 10000 sq km, not 100000 sq km. I agree that MASIE is volatile.

But "real" or "unreal", regarding this I don't agree in the main. It does appear the ice can be swept up and compacted, leading to a real (?) decrease in ice extent, and when it spreads back out when the wind shifts, we see a corresponding real increase. Or the pack can move over 20 km away from the Canadian Archipelago, showing open water, then can shift right back in the following few days. The movement is real, and it leads to volatility. Some more volatility is measurement errors, but in general, when extent gains, we can generally see what caused the gain by 'blink' comparisons of the Bremen maps.

So I believe that most of the volatility is real, and I expect to see it. If we don't see volatility, then this is a pretty good indication of some sort of "smoothing", generally longer term temporal averaging

Peter Ellis

How are you assessing that the movement is real? Is the floating buoy data somewhere accessible?

Ned Ward

Ned, I think you meant to say that MASIE has had two extent increase days of over 10000 sq km, not 100000 sq km. I agree that MASIE is volatile.

Actually, I was looking at the wrong column of my spreadsheet. My colleague's extent calculations from the daily Bremen maps show three 100K drops in the past week, and two 100K increases. But I was mistakenly reading that as MASIE. Sorry for the confusion.

Ned Ward

We've semi-automated the calculation of extent data from the daily maps in the Bremen archive. I had Mr J run it back through last December, corresponding to the same period covered by crandles's spreadsheet. So for that period I now have daily data for IJIS and Bremen (every day) and MASIE (most days, minus a few gaps).

First, here are the correlations between MASIE and Bremen:

No lag: 0.233
1 day lag: 0.278
2 days: 0.279
3 days: 0.278
4 days: 0.234
5 days: 0.202
6 days: 0.172

Over the course of the past nine months, MASIE extents have been higher than Bremen extents all the time, except for a couple of days in September (so far). For most of the year, MASIE was 0.4 to 1.2 million km2 above Bremen.

As I mentioned before, the numbers I'm citing as "Bremen" are unofficial, and are based on measuring the 15% extent in the daily maps posted in their archive. The totals differ slightly from the occasionally reported official ones.

Ned Ward

Paul writes: in general, when extent gains, we can generally see what caused the gain by 'blink' comparisons of the Bremen maps.

Doesn't that just show you where the gain occurred, rather than what caused it? It could still be either a real increase in ice extent, or an error due to over/underestimation of ice on one of the two dates.

Ned Ward

Here's a graph of the IJIS, MASIE, and Bremen extents from Dec. 2010 through yesterday:


Paul Klemencic

Ned, I can usually see what causes it. For example, the big extent drop MASIE reported for day 243 showed an unusual 32.8k drop in the Canadian Archipelago (CA). Reviewing the Bremen maps, you can't see any significant ice loss in the CA that day. On the other hand, six days earlier on August 25, you can see the ice moving away from the archipelago and opening 20-50 km of open water. Then over the next two days, the wind shifted, and the archipelago filled back in, and the MASIE extent for the CA region rose back within 4.6k of the ice extent reported for day 242.

I see this kind of action over and over again, so much so, that I expect to be able to see where the loss occurred. That's why the IJIS reports have proved so frustrating.

The comments to this entry are closed.