What happened in 2005 to Arctic sea ice?

By Steve Goddard

I need help from the readers to determine if 2010 will finish ahead or below 2005 – the red line in the DMI graph.

2010 is currently tracking just below 2005, but note there was a downwards dip in mid-September, 2005. What caused this?

The PIPS video below shows what happened in September, 2005.

In mid-September, strong winds started blowing off the East Siberian and Laptev Seas, which compressed the ice towards the North Pole. This caused the dip seen in the DMI graph.

The images below show the current date in 2005 and 2010 respectively. Note that 2005 had a lot of thin/low concentration ice in the Laptev/Kara Seas which was vulnerable to being blown around by the wind in September. The ice is less extensive, but thicker in that region in 2010.

What do you think? Will 2010 beat 2005? Please explain your reasoning.


newest oldest most voted
Notify of

Where was the jetstream in sept. 2005, and how hot was moscow-lat russia?
With all that heat flowing around northern russia, prevailing winds there should drive toward the pole.
Also, qualitatively, I’ve been hearing a lot about how thick 2010 ice is, but these two images (comparing time for time) show broad expanses of thinner ice in 2010. I would look for those to shore up and thicken due to wind compaction on the russia side and for 2010 to clear 2005 slightly to the negative within 10 days.


Voting on what the weather is going to do eh? I suppose that’s a change from being asked to vote for politicians who claim they can improve the weather and the climate. I’m afraid I can’t vote, my piece of seaweed has gone missing.


Is this Based soley on the Danish Data?
If so, based on trends, I’d think 2010 would finish above.
If not, I’d say below.
Regardless, we all know whats upcoming 12 months from now….. 🙂 Lets see the warmists try to Hide that!

Someone added the words “Arctic Sea Ice” to the title.
Grammatically that is incorrect. Please change the title to “What Happened in 2005 to Arctic Sea Ice?”
[Fixed, thanks. ~dbs]

Where did you hear that the ice is thicker now than it was in 2005?

2010 is already “slightly to the negative ” of 2005.

R. Gates

By Steve Goddard
I need help from the readers to determine if 2010 will finish ahead or below 2005 – the red line in the DMI graph.
On the DMI graph, 2010 will finish below 2005 and likely below 2009 as well, but we won’t know for a week or so…

Perhaps hydrothermal vents were more active, or an undersea volcano erupted in the area late August/early September in 2005?

Cold Englishman

It will beat 2005.
Here in Englandland, August is our warmest month, a time for buckets and spades on the beach, and the occasional dip in very cold water.
So here we are in our warm month, and as I take my Labrador Millie for her morning walk across the fields of Shropshire, it is already getting cold, so much so that the fields are covered with very heavy dew, enough to go through the hiking boots.
My guess, and it is a guess, is that winter will be worse than last year, because it will start early, and a good bet will be that the Highway Authorities will not yet have stocked up on enough salt.
It would be nice if I were wrong, I could really do with a bit more of that GW stuff, my old friend Arthur Rightis plays me up when it gets cold.
Pity really that The Met have stopped giving us all a laugh with their seasonal forecasts. After all, by the law of averages, eventually they’d have got one right, and they could have basked in our adulation.


Right now my vote is for it to not reach the 2005 extent. Although I’ve been impressed by the quality of your predictions thus far this year, I’m still worried about it clearing 2009. There are just too many unpredictable factors to say for sure.
My reasoning for this is purely statistical and relatively poor. I mentioned it in a recent comment in Sea Ice News #17…I’m using the JAXA extent records to predict the final minimum. I know this correlates poorly early in the season, but for 08/10 the R^2 value for the correlation is 0.914. Using 08/10 data, this method predicts 5.08e6 +/- 4.74 e6 km^2 (+/- 95% confidence interval). Yes, tons of uncertainty still. I’d like to improve my fitting methods by combining the JAXA data with other data sets…can I get the data (not graphs) for those somewhere online? I’d really like the 30% extent data or the area data, as those would make good multiparameter fits.
While I’d love to see it top 2005, I keep reminding myself that it’s only 1 year.


As ice behavior in the Arctic is dependent on weather conditions, attempting to predict which way the wind will be blowing at various points around the Arctic is a task best left to Madam Sofia up the street.
I believe what one must do is look at probabilities and not attempt to put too fine a point on a prediction. At this point in time, as far as I would be willing to go is to say that I believe that 2010 will finish with more ice than 2009 and indications are good that it will be a 2005-ish year.
But a change in the weather can change all that in only a few days time.

John S.

Arctic air temperatures for 2010 are running well below 2005 temperatures. Therefore I predict that 2010 sea ice extent will exceed that of 2005.

Pamela Gray

Since 2007, that is some bodacious death spiral right thar! Not. But then the incredibly ignorant folks here at WUWT must be reading the graphs wrong. Now where did I put my AGW-colored glasses, and now that I think about it, it is time for my daily dose of red koolaid with a lemming chaser.


Always have your Kashmir sweater available when we have an Indian summer.


Hmm, looks like the JAXA record was modified overnight and for the 08/10/10 value. With the new value the correlation method is predicting 5.07e6 km^2.


All the heat’s in Russia so it’s lacking at the pole. Jetstreams farther south this year and all… 2010 will have more arctic sea ice at the minimum than 2005. Simplistic logic i know but me no weatherman, just listening to some of them…


My guess would be close or over 2005, based on lower solar activity. You can clearly see the rising trend since 2007. But could you ask Piers Corbyn, he follows solar activity, ocean currents and high-atmosphere turbulence, resulting in pretty accurate weather-predictions.


2005 was a monster in terms of hurricane activity and looking at the track info
shows that a lot of tropical heat went north that summer/fall. 2010 is analog for 2005 and 1998 so we might expect a similar result (other things being considered) if the season ramps up as quickly as Dr. Phil, Joe Bastardi and Mark Sudduth have explained.
What does the ice situation look like for those other years where hurricane activity was high in number and strength? Just proposing a piece of the puzzle.

Earle Williams

I’m beginning to think that predicting the fate of Arctic sea ice is akin to predicting where the roulette ball will fall. Just looking at the DMI graph, you have 2010 tracking 2005 for the last few weeks. But for all of June 2010 showed the lowest extent while for the month of April 2010 it showed the highest extent. The within-year variability of the sea ice extent metric makes it extremely problematic to forecats based on which year it is currently tracking.
Looking at the PIPS images for 2005 and 2010 it seems to me that there is a significant difference in the arctic sea ice extent portrayed. Yet the extent as shown by DMI has 2005 and 2010 on par. Accordingly I would have a hard time interpreting anything from PIPS 2005 that could be applied to the 2010 DMI situation.
I do find the Arctic sea ice conditions an entertaining and compelling melodrama. There seem to be a broad range of expert and amateur predictions for the sea ice minimum this year, but I have my doubts that we will learn anything from the 2010 results. Most predictions will be wrong, and one or more will be close enough to the mark to be deemed prescient. No matter who gets that brass ring, our understanding of the dynamics of Arctic ice conditions will be no richer.
Regardless of the outcome, the CAGW alarmists will continue to refer to “dwindling Arctic sea ice”. The 2007 minimum will give years of dividends in ensuring a linear fit that matches the narrative. So 2010 will be spun as confirming the consensus. Accordingly the public understanding of Arctic sea ice conditions will continue to be an order of magnitude below our feeble scientific understanding. *sigh* /rantoff

Douglas DC

N.Pole is cold the wind should go to the lowest pressure-where it is hot.
BTW-Check out Joe Bastardi’s comments.
things are about to change…


Forecast! Why? Leave that to IPCC to get wrong. It is the area under the curve that matters. That will give a comparative measure of the amount of energy locked up by the latent heat of freezing. So far, looks like a record anyway.

Frederick Michael

Right now, the lack of ice going trough the Fram strait is remarkable. If this continues, significant recovery will result.
More Arctic sea ice freezes every winter than melts every summer. The loss through the Fram strait is what keeps it from building massively (plus a bit more through the Nares strait.) We didn’t lose much this year and that means significant future buildup.
Also, the Arctic has been quite cloudy recently and this is slowing the melting.
Does this mean we’ll beat 2005 this year? No; I’d be surprised if that happened.
But we should beat it next year.

Tom in Texas

…note there was a downwards dip in mid-September, 2005. What caused this?
It appears to be a dip because of the convex shape of the curve in the 1st half of Sept.
Compare the shape of those 2 weeks, to the 2 weeks in the other 4 years.

Ed Caryl

The mid-September 2005 was a one-time event. Temperatures in Siberia south of the Kara and Laptev Seas are below normal and the ice is relatively thick north of there. Current air temperatures on the ice are at or below freezing. Nothing bad is going to happen to that ice. I expect the 2010 ice to track right along the 2005 line until mid-September, then cross over to positive territory.


I predict 2005 finishing ahead of 2010 by 5 years =)
But seriously, all years finish about the same. I don’t know why this would be any different.

The NSIDC newsletter includes this graph, which seems to hint at a near record low in 2010.
How would you reconcile your ideas with the NSIDC graph ;^)


Tom in Texas says:
August 11, 2010 at 12:21 pm
It appears to be a dip because of the convex shape of the curve in the 1st half of Sept.
Agree, I don’t see a dip at all.


I am thinking it will dip below 2005. Possibly even lower, and match 2009 give or take a bit. Why you ask? Well, first off, Joe Bastardi said it would likely be lower this year than last, and the good folks governing us will see that it is at least reported as being so low as to cause a death spiral of overheated polar bears and water chest deep in the Sierra’s.
Actually, my bet is it will dip below zero much earlier this year, thus slowing the melting, but probably be much windier so who knows… I’ll go with Bastardi.

Rhoda R

I’ll put my vote in for exceeding the 2005 extent based on that whanking big ice slab that blocking the outflow of ice.

Michael Schaefer

I don’t know what it’s going to be THERE.
But I know that HERE in Germany, all signs are indicating an early, COLD, long, harsh winter.
Mark my words.
I shall return.


WordPress seems to dislike when I use Joe Bastardi’s name twice!


Tom in Texas,
HAHAHA…Good one, wish I had said that! 🙂

Pamela Gray

I don’t know about beating 2005 but I am of the opinion that current weather parameter oscillations predict continued ice behavior that points to a possible recovery. So my guess is less melt than last year.

Pamela Gray

Question, I’ve noticed that my comments are being delayed for moderation by nearly an hour. Is there something wrong?
[Moderator shortage today. Sorry.]

It will beat 2005.
The trajectory is coming from further back and is rising towards 2005. By late August it will be past 2005 in extent by a small margin.
“It’s the jet stream stupid”!!! 🙂

Sean Peake

My inspired guess would be between ’05 and ’06. Rationale: weak solar activity for three years, La Nina evolving, reversal of the PDO, possible continued negative NAO, and low daily mean temperatures north of 80 degrees (now below freezing).


I think we might be missing the big old ice munching anomaly (+5deg) in the Bering Straits of 2005.
I appears to tamed itself to Sep-Oct 2005 levels (two months early).


Can I vote for a tie?

Josh Grella

I believe 2010 will come in just a hair above 2005. How far depends on a couple of factors. The main factor will be wind direction. In September 2005 it appears the winds began to do two things. First, it looks to me that a lot of ice was pushed out to sea off the east coast of Greenland. That loss of ice allowed for added compaction which was caused by those same winds. Right now the winds are very much different than they were this date in 2005. If that continues, there should be considerably less compaction for the remainder of the melt season and less of a push out to sea; Sea ice extent and area should not take the dip seen in 2005. Second, the temperature is below normal. I don’t know what the temperature was for this date in 2005, but I do know that it is (from what we can tell) below normal in the arctic right now. If that trend continues, there should be less melt and an earlier start to the refreeze. Again, sea ice extent and area should not take the dip seen in 2005.
All that said, if the winds take a sudden shift (something similar to what Bastardi seems to indicate in his updated hurricane prediction), the thinness of the ice could cause a dip to 2009 levels or lower.
On another note, I was surprised by the different reactions I had to the graphics. In the DMI graph, sea ice extent looks very much in step with 2005. However, when I looked at the PIPS images, it appears that there is a lot less ice than in 2005. Sure, there are a few small areas where the ice is thicker this year, but overall, the ice appears to have been .5 to 1 meter or more thicker in a large chunk of the image. Also, by yet another simple eyeball guesstimate, that slightly thicker ice appears to have covered a larger area. I know you can’t measure extent with the PIPS map, but it looks a little different than what I would have guessed.


Daily totals are still well above 50sk loss a day. Until I see a slowdown of the melt, it projects as beating only 2007.


Looking at the DMI Graph, most of the historic lows occurred in the first 2 weeks of September. Except 2005, when the historic low was in the last 2 weeks of September.
So to answer your query, ” determine if 2010 will finish ahead or below 2005 “. My prediction is: 2010 will be below 2005 in the first 2 weeks of September, but above 2005 in the second two weeks. But the low-low will be about the same, just at different times.

donald penman

the dmi arctic temperatures are going slightly further below freezing tonight and are much lower than they were in 2005 at this time.I would be pleased if 2010 finished above 2009 in arctic ice extent but we will have to wait until the ice melt is over before we can be certain of that.


No whiskey added on september?

R. Gates

Frederick Michael said:
“Also, the Arctic has been quite cloudy recently and this is slowing the melting.”
This time of year, the cloudiness is not necessarily a factor in keeping temperatures down. Certain type of cloud cover actually keep temperatures warmer by increasing downwelling LW radiation. Clear skies can allow more heat to escape.


I think it will be higher than 2005.
This is based on a gut feeling… sometimes it’s all we have in this damed science.


simpleseekeraftertruth says:
August 11, 2010 at 12:14 pm It is the area under the curve that matters. That will give a comparative measure of the amount of energy locked up by the latent heat of freezing.

That’s good physics! Knowing the equations of motion for the problem at hand is required to understand! It is amusing to observe the complete lack of understanding by most “experts”, IPCC, New Scientists and New York Times:
Arctic Melt Unnerves the Experts
Scientists are unnerved by this summer’s massive polar ice melt, its implications and their ability to predict it.


My vote is that it will beat 2005 because of an ealier freeze up.
Comparing the temp north of 80th parallel of 2005 and 2010, the temp for 2010 has already crossed the freezing line. 2005 follows the average curve, and does not cross the freezing line for another several days.
IF this trend in temp continues, I’d speculate that your gonig to see the extent and area lines begin to flatten out, and maybe even begin grow a tad, making the assumption that lower tems and below freezing temps leads to ice formation, and by default, a halt in ice melt.


Quick question – when it comes to the end of the melt season, which source will be used to determine who was closest to the ‘minimum ice extent.’ NSIDC, JAXA, DMI? I feel there needs to be some sort of standard metric for this, else people will claim to use the one that most closely matches their prediction (I fear its already too late to do this… should have been done long ago).

R. Gates
In order for clouds to produce downwelling LW radiation, they had to have been first warmed by upwelling LW radiation. Not much of that over the ice, is there?

Wes M

Based on the historical data for the last several years, I’d have to posit the theory that 2010 will end up slightly above 2005. The warmest months are coming to an end, and by the end of December I could see a small uptick just enough to nudge it ahead of 2005. The 2007 line shows a major reduction during August to October, moreso than any other line on the graph. November and December are big gain months for sea ice, and based on the similarities to 2005 I’d say 2010 will end up slightly to very slightly above once the final tally is finished, based on the recovery of the 2007 line during those months.