« Signs of Arctic climate change | Main | Sea ice loss 2012: what do the records mean? »


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

R. Gates

Thanks for posting that graph Neven, and enjoy the rest of your vacation. Again, Chris R. and I both agree that these data points are likely erroneous, but with each additional one that is clustered around these higher levels, greater attention should be paid.

I think the issue of a rapid rise in methane levels, not quite a huge "burp" of methane, but an accelerated rise is now certainly an issue in play based on the rather amazing decline in sea ice we've seen this summer. We know that Semiletov was shocked by what his research crew found last year in the E. Siberian sea related to the vast plumes of methane coming up from the sea floor. It is not time to be overly "alarmist" about this (calmness always wins the day anyway), but I think it is certainly time to consider the possibility that methane could begin to take some much larger jumps in concentration than might have been thought possible just even a year or two ago. This is all of course related to the declining sea ice.



Have fun with your grandfather and his wine. I suspect the melt record will still be in the future when you return.

My guess is September 21. It could be as much as a week before or after that.

As was noted earlier, **all** of the outlier methane data is high. None is low. This limits what sort of errors could be at fault. It seems unlikely they would be using methane anywhere near where the samples are taken, though sillier methodology errors have happened before. I suspect it is more likely that they are getting whiffs of the clathrate break in progress.

With the warming of the arctic and the breaking of the clathrate, as well as the tundra thawing and rotting, I suspect we crossed a major transition into a new state. The warming will continue, the methane release will increase. Already we have likely transitioned to a place where the systems growth in methane and CO2 release exceed our ability to cut emissions.

It seems likely that we are in a run away condition already and that CO2 will peak at over 2,500 ppm. Burp.


Measurement errors are routinely jumped on as signs of a catastrophe. Two or so years ago there was a similar reading at Zepplin station that also turned out to be just erroneous readings.

Peter Ellis

This whole "measurement error shouldn't be one way" is unmitigated bullshit of the type that's regularly trotted out at WUWT and other denialist sites to discredit various adjustments to the temperature record.

Simply put: it depends on how the measuring apparatus works, and what the likely failure mode is.

The methane measurements are taken by gas chromatography.

The sample is vaporised, and then injected along with an inert carrier gas stream into the instrument. If bits of this fail, will it read low or high? I don't know and neither do you, however the evidence from past measurements indicates that the likely failure mode is over-reading. Or at least, there have been many previous high readings which were ascribed to leaks and instrument problems by the people that actually use the thing, who should know.

One possibility I can see is that if there's a leak causing a low flow rate of the inert carrier gas, then the sample will presumably end up at higher concentration. If this is the most common failure mode, then all the errors will be high.


IARC-JAXA ticked UP (corrected final values) for the first time this summer:

9/6/12 - 3676406 km^2 (prelim 3614219)
9/7/12 - 3664531 km^2 (prelim 3601875)
9/8/12 - 3674844 km^2 (prelim 3595781)
9/9/12 - preliminary 3593750 km^2

It might not be the minimum for the summer, but it definitely shows the end is near. 2007 had a local minimum, went up a bit, then down for the final minimum a few days later.

Today's preliminary value is very slightly lower than yesterdays - 2,031 km^2 smaller.


CT area update: new low (-4.5k):

2012.6740 -2.3895028 2.3765626 4.7660656
2012.6768 -2.3920436 2.3635330 4.7555766
2012.6794 -2.3956523 2.3500910 4.7457433
2012.6823 -2.4353788 2.2982104 4.7335892
2012.6849 -2.4319782 2.2937613 4.7257395

How likely is an extent min before the area?


Semiletov went on a new expedition 3 days ago (from Murmansk, Northern Sea Route).
"According to the head of the expedition, Igor Semiletov, the focus will again be methane emissions in the Arctic seas, particularly in the Laptev Sea". So new results can be expected soon.

And one more news:

An international expedition of Russian, Japanese and South Korean scientists on research vessel Academician Lavrentiev went from Vladivostok on August 7 to explore new deposits of gas hydrates.

The expedition discovered gas hydrate deposits on the slope of the Kuril Basin in the southern part of the Okhotsk Sea.
On the same slope they discovered a powerful stream of bubbles of methane. It rises to the surface from a depth of 2.2 thousand meters, the first known stream rising from such a depth.
Large concentrations of gas hydrates were also discovered in the Sea of Japan on the west slope of Sakhalin in the Tatar Strait, with at least 43 plumes of methane bubbles rising from the sea-bed.

(sry for my poor English :)

Lord Soth

The date for the minimun will be up to the weather, but I believe we are probably only talking about a maximum of another 100K of sea ice lost in the worst case scenario.

I do however predict a long flat bottom, before recovery in late September. The Cryosphere anomaly will drop below three, forcing them to extend their graph.

If you look at the DMI north of 80 temperture graph, the average temperture is going down, but it is going down slowly and without any wild swings like the previous years.

Just like fall is delayed by warm seas in coastal areas and islands, The sea ice island north of 80, will encounter the same effect to a lesser extent.

The 2012 DMI north of 80 average temperture graph is only a couple of degrees below freezing. The curve for the past few weeks really stand out when comparing against previous years.

L. Hamilton

DMI and CT both down by tiny amounts this morning, so by tiny amounts set new records. Here's DMI past 10 days (total in millions, delta in thousands):

. list edate totaldmi delta0 in -10/l

| edate totaldmi delta0 |
2706. | 30aug2012 2.6618 -15.9 |
2707. | 31aug2012 2.6246 -37.2 |
2708. | 01sep2012 2.7068 82.2 |
2709. | 02sep2012 2.6252 -81.6 |
2710. | 03sep2012 2.6473 22.1 |
2711. | 04sep2012 2.6186 -28.7 |
2712. | 05sep2012 2.6178 -0.8 |
2713. | 06sep2012 2.5681 -49.7 |
2714. | 07sep2012 2.5694 1.3 |
2715. | 08sep2012 2.5658 -3.6 |

Matt Arkell

Lord Soth,

The 2012 DMI north of 80 average temperture graph is only a couple of degrees below freezing. The curve for the past few weeks really stand out when comparing against previous years.

I was looking at the DMI temps earlier and it does appear that the decline is slower than previously seen. The value at the moment is about the same as 2007, it will be interesting to watch over the next few months to see if it tells us anything about the progress of the freeze-up in that particular part of the world.

2007 did have a few more swings though, which suggests we have more uniform conditions this year to me. With less ice about, the energy should be a bit slower coming out, so i'd expect warmer conditions than we seen for some time persisting.

Jim Williams

The seas have been typically freezing up rapidly during October. I wonder if that is going to be true with the Beaufort Sea this year?

Also, earlier in August someone speculated about the possibility of the freeze-up starting on the coasts rather than with the ice pack this year. While I kind of doubt it, the WACC scenario could make this happen -- soon, if not this year.


>"With less ice about, the energy should be a bit slower coming out"

Don't follow that. It isn't heat above freezing point going into ice (that would melt ice). For extra ice to form heat has to be lost to atmosphere or space and ice acts as insulator slowing that down.

Without anything for ice to form on, the water could be supercooled below freezing point but I doubt that creates significant delays before ice forms. The water surely isn't all that pure is it?

However I think I agree with you about slow freeze up. I think GAC 2012 and other lows will be keeping a longer column of water mixed with more heat to be lost before freeze up starts. However weather could trump that, lots of highs causing clear skys could be more important.

Current highs bringing lots of warm air to weak parts of ice are a little more ambiguous but that won't last long.

So I am expecting minimum to be fairly soon but then have slow freeze up.

Kevin McKinney

Calmness in the face of challenge is indeed admirable, as this video aptly illustrates:


(Caution: Canadian humor alert!)

(Cross-post from RC.)


I was wondering how the climate models fare at predicting minima, and how much improvement CIMP5 shows over CIMP3. Haven't found anything conclusive, however...

The CIMP3 models, used as the basis for the IPCC AR4 (2007) report, were way off but in a surprising way. If you look at the month-by-month model values (Fig 1 in Stroeve et.al. 2012, http://www.agu.org/pubs/crossref/2012/2012GL052676.shtml),
their errors vs. observed mean increase markedly from January through April. So where reality wrong-footed the models was not so much the amount of peak-to-trough melt (by area/extent), but the lack of re-freeze afterwards.

The CIMP5 ensemble has a similar monthly pattern of errors - at least on data through 2011 - though to a lesser degree. This reinforces the idea that we need to figure out what caused the melt anomalies to shift earlier in the season.

Nightvid Cole

I doubt the minimum will be before the 13th (since the first uptick on IJIS is usually not "the real deal" and the air temperatures around the edges of the current ice pack are still fairly high) but I also doubt it will be later than the 19th since PIOMAS predicts refreezing beginning by about then. Time will tell, I expect it to be in the September 'teens .

Matt Arkell

Don't follow that. It isn't heat above freezing point going into ice (that would melt ice). For extra ice to form heat has to be lost to atmosphere or space and ice acts as insulator slowing that down.

Without anything for ice to form on, the water could be supercooled below freezing point but I doubt that creates significant delays before ice forms. The water surely isn't all that pure is it?

Your're correct about the water not becoming supercooled, those conditions simply wouldn't exist in the open ocean. I was thinking along the lines of additional heat being present in the water having to be lost to the atmosphere before freezing occurs. And a lot of >80N is still ice covered, so even if that was the mechanism, the effect should be small.

Once enough heat is lost, then yes, there would definitely be ice forming on the rest and acting as an insulator. I'm inclined to agree with minimum being soon, but that the additional heat energy hanging around being sufficient to slow down the freeze up. I can't help but wonder at what point the straw that breaks the camels back is applied and the winter volume gets damaged enough that summer becomes a real struggle.

Protege Cuajimalpa

Yesterday was my first time that I saw the NOAA page with the Global Monitoring Division and it is great to have worldwide gas measurements in just one page:


Even though, the perception that I had yesterday is that humanity isn’t properly making measures of methane, especially when we have incredible melts at the Sea Ice Arctic (like the one we are having now). Seems that most of the measurement places are concentrated at cities and none of them are at sensible permafrost places. It was unexpected to find that laboratories at Tiksi-Russia and Fairbanks-Alaska only measure ozone. In fact, this page shows that in Russia there is not a permanent place measuring carbon cycle gases.
In this context, I want to bring out the comment of R. Gates when he wrote:

I agree that the most likely explanation is erroneous readings, and in no way was suggesting some big methane "burp" had occurred. However, please keep in mind that Barrow is the only Arctic location to record daily average methane readings as those other sites I believe are all monthly averages. Thus, those other Arctic sites would not be comparable to the readings from Barrow.”

Also seem that we are relying in sporadic explorations. Again, R. Gates wrote:

“We know that Semiletov was shocked by what his research crew found last year in the E. Siberian sea related to the vast plumes of methane coming up from the sea floor. It is not time to be overly "alarmist" about this (calmness always wins the day anyway), but I think it is certainly time to consider the possibility that methane could begin to take some much larger jumps in concentration than might have been thought possible just even a year or two ago. This is all of course related to the declining sea ice.”

So, would you agree that there is the necessity of permanent places to measure carbon cycles gases at selected places at the permafrost? Or it will be better to measure the methane by other methods, like a specialize satellite?


I've heard yet another silly denialist meme twice since yesterday. The claim is that 450 more cubic kilometers of ice volume went away in 2010 (18,974 km^3) than have so far this year (18,523km^3), meaning--of course--that the ice is in "recovery".


That's about as witless an argument as one will ever hear. The fact is, there's less ice to melt every year, so of course there'll be less total melt. Extending that silly argument into the future, then, one supposes that the ice will be in full recovery a short number of years from now when "only" a few thousand cubic kilometers of ice form during winter, meaning that "only" a few thousand cubic kilometers of ice are available to melt during the following summer.

"Recovery" for the win!!!


Anyway, should you run across any WattsBots™ attempting to avoid reality by using this particularly lame argument, just show them this graph (click for larger image):

Here's your ice 'recovery'.

The total amount of melt is less important than the total percentage--and that percentage is rocketing upward, as anyone not blinded by either disconnected ideology, or sycophantic loyalty to Watts can see.

(FWIW, I also created also a version of that chart based on CT SIA. It tells pretty much the same story.)


Peter Ellis

This whole "measurement error shouldn't be one way" is unmitigated bullshit of the type that's regularly trotted out at WUWT and other denialist sites to discredit various adjustments to the temperature record.

Is it your assumption that the green crosses are measurement error rather than wiffs of methane?

Jim Williams

Jim Pettit, would you please add your volume graphs to the same page as the area and extent graphs (reachable from Nevens' Arctic sea ice graphs page) so I can find the most recent versions when I want them?

Nightvid Cole

If we get another big storm before month's end or so to stir up salty water from the deep, next spring most of the Arctic Basin will be dominated by extremely salty first year ice which given the current pattern we've been seeing in May and June snow cover and albedo feedback, is very unlikely to survive the summer. The question I have is, how intense does the storm need to be, and how likely is this to happen? If it does, I wouldn't be surprised if in 2013 or 2014 we have an ice edge pulling back somewhere to over 88 degrees North. Takers, anyone?

Jim Williams

Protege Cuajimalpa, I spent some time with that site yesterday and couldn't find any other CH4 readings later than early August. The Barrow readings look like the latest by a good two weeks. I was going to ask Chris to explain his assertion, but decided to let it slide.

Seke Rob

Talking ESRL and world gas measurements on one page, the AGGI (Annual Greenhouse Gas Index) silently update as "Summer 2012".


A little snip relating to CH4 burping:

Radiative forcing from CH4 increased from 2007 to 2011 after remaining nearly constant from 1999 to 2006.

We're nearing 3 Watts added since 1979. The 2011 addition was 1.2% over 2010. We're doing good [born in the USA] thanks to shale gas, is the bold claim.

ODGI (Ozone Depleting Gas Index) has not update yet.


Wili has posted a reference to Methane emission from the Arctic Ocean? - Satellite data - A letter on RealClimate and quotes

The main conclusion of this paper is a qualitative detection of high and over the years increasing methane mixing ratios in areas coinciding with predicted locations of methane hydrates.

I'm to whacked to read it properly. Comments would be useful - I should be talking to someone at the Met Office tomorrow.

Jim Williams [September 09, 2012 at 16:07]: Jim Pettit, would you please add your volume graphs to the same page as the area and extent graphs (reachable from Nevens' Arctic sea ice graphs page) so I can find the most recent versions when I want them?


I realize my graphs site, particularly the home page, is starting to become an overloaded mess. After the melt season is over, I plan rename and rearrange the various graphs into a more logical structural order. I hesitate to do it now, as people don't like things not being where they're used to them being... ;-)

Artful Dodger

I agree, Geoff. Methane is SAMPLED. Some samples are high. It is the average that matters.

But it should be well known by now that at 1.95 ppm CH4, and a 120x C02e for CH4, the effect of Arctic methane is to add about 235 ppm CO2e to the GHG load in the Arctic.

This is especially important as the polar night descends, increasing the thickness of the insulating blanket over the warmed seas.

Aaron Lewis

Our montoring net work is sparse! Compare hourly ch4 concenration with wind direction,and you will some small spikes land sources, and larger spikes from some vectors from ESS. At the hourly level, adjusted for wind direction the spikes look like real data because the next hour when the wind is from a different direction, the level goes down.

Getting good results from a sparse air sampling network requires very smart air modeling-- that we are not getting. In fact, intellegent placement of sampling stations requires good air modeling, which was not done because they did not know where the sources would be!

r w Langford

Here is a link to an article from Daily Kos. It describes a possible link between increased arctic methane and pearlescent blue cloud formation at high altitude.

Artful Dodger

Has anyone else noticed that Neven has titled the post "Minimum open thread", and not "2012 Minimum open thread"?

Yeah that's right, this is HISTORY in the making, and not just since 1979. We know with certainty this is the lowest sea ice in more than a century (Neven likes centuries), and we're also highly confident this ice retreat is unprecedented in over 1,500 years.

Now, what shall we talk about?



If I am not mistaken the way to read the upper graph is that the Blue points are verified data, the Green are points that failed validation and the Orange point are waiting verification. So there are no data points after Dec 31, 2010 that are validated yet. Is this mistaken?

Another point. Lodger stated that the methane load in the arctic amounts to about a 235 ppm CO2e addition to the CHG load in the arctic. I think this may not be accurate. If one goes here


they find that the 2009 global level of GHG give a CO2e of 439 ppm. I quickly looked at 2009 levels of CO2 and CH4 and the change since then does not come close to the number one would get from the 235 addition. The CH4 levels in the arctic are about 10% higher than the global average so it would seem more accurate to say that heating potential in the arctic due to CHGs is about 110% of the global average. I think.

Am I misunderstanding something?

J. Dunlap

Jim Williams

Thanks Jim.



I meant to say that since CH4 in the arctic is about 110% of its global level then its 'contribution' to GHG warming of the arctic should be about 10% greater than its contribution globally. Not that CHG warming of the arctic is overall 10% greater than it is globally.

J. Dunlap

Chris Reynolds

Protege Cuajimalpa, Geoff Beacon,

I suggested that the emissions from the East Siberian Shelf were visible on AIRS. Which is surprising because AIRS retrievals are from the 400mb level, up above mid troposphere.
Dr Yurganov had some strong objections which I answered here.

So I still think it's possible to see the ESS emissions in AIRS. But Dr Yurganov has greatly improved on my amateur fumblings in his latest paper.

He's used a different system, IASI, with a better resolution of IR spectrometer. This has enabled detection of methane close to the boundary layer. From that he's shown how emissions from the ESS appear in th data.

The main figure is the penultimate one. The right most light blue bar is the location of the ESS methane deposits, and there is a substantial peak of methane detected in autumn months.

So rather than a massive and costly network of ground stations, satellites can be used to spot anomalies and direct ground work to those areas.

Jim Williams,

Protege Cuajimalpa, I spent some time with that site yesterday and couldn't find any other CH4 readings later than early August. The Barrow readings look like the latest by a good two weeks. I was going to ask Chris to explain his assertion, but decided to let it slide.

The bottom line is this - a few data points from one site are not enough. A large emission of methane would produce a large regional pattern. Once the month had ended you'd be able to check AIRS, but you might still have to wait for the next month. The correct course of action when faced with a few data points from one site and no further corroboration is to sit back and await further data.

But if you lot want to run around flapping your wings, fine, I'm not going to spoil your fun in future.

Artful Dodger

Hi J. Dunlap,

People often confuse the 100-yr time horizon for the effect of CH3 with it's instantaneous effect.

Over 100 years, a single gram of CH3 will have the same Global warming potential as about 20 grams of C02. That's because of the half-life of CH3 in the atmosphere as it slowly degrades into CO2.

However, that is not the situation in the Arctic. CH3 levels are either steady or rising, so it is appropriate to use the 1-yr time frame to compare the effect of CH3 to CO2, which is about 120x CO2e.

So, 400 ppm C02, plus 230 ppm C02e from CH3, gives 630 ppm C02e for just these two trace gases in the Arctic. This Winter. While the sea is trying to cool, and the ice should be thickening...

This topics has been discussed here on the ASI blog frequently. Hope this helps.




R.Gates, who is "waiting to hear more about this process from several sources directly involved", writes on the previous board

BTW, the question was asked was the green plus marks mean. This from the ESRL site:

"+ Symbols are thought to be not indicative of background conditions, and represent poorly mixed air masses influenced by local or regional anthropogenic sources or strong local biospheric sources or sinks."

Thanks again, Chris,

I've reread those posts. While I'm not discounting the imminent possibility "some sort of methane driven apocalypse" that you downplay, I was more wondering if extra "whiffs of methane" from whatever source have been increasing. Does this harbinger an addition to climate change? If so how big?

Your answer to Jim Williams suggests you might say sit back and await further data. I'm due to quiz someone from the Met Office and it's coming up to the autumn political round. Do I tell them just wait and see?

Seke Rob

Lodger, if it exists, CH3? (Thought we're talking CH4)

Espen Olsen

I am mushroom enthusiast, but I am disappointed of the recent 2 seasons, the collection of Porcini (Boletus edulis) and Chanterelle (Cantharellus cibarius) are record low, and we are talking record low (Scandinavia)!!
I believe there is a relation between the 2 last winters and the Arctic situation.
We are probably surviving without the different mushrooms, but what else is showing up in the horizon?


It is highly unlikely, but if 2012 is followed by two years where melt increases by the same amount as we have on 2011's minimum then we'll be more or less ice free in summer 2014.

This is unlikely as a BIG melt increase, as we have seen this year, last came about 5 years ago in 2007. However, I propose that the frequency of BIG melts will increase or even become the norm (a true 'death spiral' to no ice)

We know that due to the large amount of melting over the last 5 years the ice is thinner, has far less coverage and seas are warmer. This is topped off by additional agitation of the sea by the increased frequency and magnitude of storms, which leads to increased salinity in the Arctic Ocean. I agree with Nightvid Cole that the large amount of first year 'salty' ice next season will be arguably weaker, quicker to melt and thus make the chances of another record melt next year more likely.

Neven is constantly reminding us that the weather no longer seems to be an overriding factor due to the weakness of the remaining ice, so could we really be ice free by 2014?

If we do get to an ice free state, sorry when we get to an ice free state in summer; I believe it will have an immediate and devastating effect on the winter recovery that year. In fact this devastation for winter ice may start sooner (keep your eye on Winter 2012-13). I believe one of the key months will be June (the month of maximum insolation for the high Arctic); as once it is more or less ice free in this month the seas will be getting so warm I don't think a re-freeze will be possible to any significant extent.

So interesting, yet so worrying!


Sorry for the double post, but just doing my evening rounds of the data, graphs satellite images etc. Take a look at tile r03c03 on the NASA 1km satellite images. The remaining pack has massive leads all over it. It must be so weak a big storm would have a really bad effect on it right now. For those without the link:



Thanks Chris and R Gates for the papers and references to methane release, Barrow data and Yurganov's research into satellitte based detection of CH4. I am particularly indebted to Chris and Dr. Yurganov for what follows.

One project I have worked on but held off on publicizing was mapping CH4 in the Arctic. Dr. Yurganov has graciously given his permission to include his IASI mapping in this new website.

Here is what is postede:

AIRS/Giovanni data at 359hPa:

The imagery is the methane/CH4 PPBv from the 2012 AIRS/Aqua data, 45-90N latitude, as a 10 day average at 359 hPa ascending, created using Giovanni.

The methane mapped is that tracked in the middle troposphere at approx 26,000 feet or 7,900 meters.

The IASI data:

Maps the IASI CH4 methane PPBv data as a 10 day average from 970-600 mb or from 1,201.6 feet or 366.3 meters to 13,794.9 feet or 4,204.7 meters - the lower troposhere.

For the science, IR processing and platform description, see this paper:

The actual alghorithm is explained in:

Xiong X, Barnet C, Maddy E, Wei J, Liu X, and Pagano T S (2010) Seven Years’ Observation of Mid-Upper Tropospheric Methane from Atmospheric Infrared Sounder. Remote Sens 2: 2509-2530, doi:10.3390/rs2112509

All IASI data are downloaded from http://www.nsof.class.noaa.gov/saa/products/welcome

The AIRS algorithm by Xiong et al was applied to IASI without significant changes.

The A4R methane website pages are:

1) AIRS 2012 vs 2011 CH4 359 hPa


2) IASI 2010-2011-2012 970-600 mb CH4 Comparison


3) IASI 2008-2009-2010 970-600 mb CH4 Comparison


4) 2012 Arctic CH4 AIRS 359 hPa and IASI 970-600 mb


5) 2011 Arctic CH4 AIRS 359 hPa and IASI 970-600 mb


6) 2009 Arctic CH4 AIRS 359 hPa and IASI 970-600 mb


I hope this imagery recorded in 3 10-day increments per month, will aid our discussion and set the stage for methane observation and assessment.

I will keep this imagery updated, and will add more 2010 imagery later.

After looking at this, I conclude there have been significant incremental changes in atmospheric CH4 in the last five years.

What we will see this fall, after unprecedented melt and heating in the Arctic, remains to be seen.

Artful Dodger


I find one must occasionally introduce a small error, in order that people may accept the larger truth. ;^)


Dan P.

Going back to DMI temps above 80°, I agree with Matt Arkell. During the refreeze I would expect open water to hover at freezing point, while air over ice can drop via radiative cooling. There is much more open water above 80° than in previous years, so the average temperature will be held high compared to previous years by the unusually large fraction of air stuck at freezing while water below it slowly freezes.

I would also assume that ice would have less heat exchange with the air above it than open water. That would make the ice-covered fraction more prone to temperature swings, both up and down, due to weather.

Kevin McKinney

Insightful, IMO, Dan P!

Seke Rob

Today it's 120 times, last week saw 100x quoted, at CH4 emission time... it's not really that important in the larger truth finding. Adding Wadham's 20 years CO2e emission effect from the recent sea ice loss. We're needing very big screwdrivers to close the lid on this, but that ain't going to happen... too many believing in miracles, led by hordes of Tonies, stuck in their BAU ingrained habits... budget deficits have to be recouped. Bill said it right "Arithmetic" and the grand failing in that department.

Seke Rob

Dan.P, few days ago was going through the different years of DMI >80N temp graphs... this years curve looks different... too smooth, only 2007 looks a little like it and 1998 maybe... well above the base line for this period: http://ocean.dmi.dk/arctic/meant80n.uk.php


The air being warmed is part of the process of water losing its heat. So if the air temps are higher that means water is losing its heat faster. So does that argument amount to:

The water loses its heat more slowly because the water is losing it heat faster?

Which is it?

Dan P.

Seke Rob, I noticed the same thing. I think that smoothness is a lot more likely before all the area above 80° is frozen. Basically we're seeing an island climate surrounded by moderating water, with the special feature that there is a ratchet pulling temperatures down but holding a portion at freezing until ice forms. Once the whole area is filled with ice you could expect more continental weather, with a much lower temperature floor but occasional incursions of warm air from outside to spike temperatures up.

Still that reasoning seems kind of flimsy, and I wouldn't be surprised if the right weather patterns could mix up the temperatures dramatically even with the open water moderating.

Dan P.

crandles, what I'm talking about is not the rate of heat transfer but the actual amount of energy stored as latent heat in the unfrozen water. The water and air above it are certainly losing heat faster than the ice, because they are warmer as well as having less insulation. But the *temperature* drop is slower because you have the pause while the ice refreezes.

Dan P.

It would be interesting to compare actual daily air temperatures in the Arctic this year versus past years to get a sense about spatial patterns & variation. Today's surface temperatures are at http://ocean.dmi.dk/arctic/weather/arcticweather.uk.php

and show the pattern you would expect: temperatures hovering near freezing over open water nearest the pole, and coldest where the thick ice is pressed against the Canadian archipelago. But it would be cool to look at several-day animations from past years to see how temperatures evolved during re-freeze. Anyone know where to find archived ECMWF data like this?

R. Gates


Thanks for all those charts. Lots of very interesting comparison work. Of particular interest in a direct comparison of between the later weeks of August 2008 and August 2012. It would be interesting to do a blink animation between these two months. Based on a quick scan, Greenland for example shows a huge change.

Overall though, Chris R. is absolutely correct. We need to wait for a bit more data to come in and see how some of these outliers play out. Certainly, at the very least just based on the charts we're seeing for comparison year to year, methane levels across the Arctic have increased quite markedly over the past 4 years.

Mikkel Fishman

I know nothing and thus have no opinion about the Barrow methane readings.

That said these papers are frightening:


Basically they suppose localized permafrost decomposition can create pockets of methane that cause a subsurface chain reaction when the pocket pressure is enough to break through its container. This can then occur fractally and the total potential is a function of the rate of temperature increase.

The second paper likens it to an action potential in which there is a quiescent period where build up is undetected but a critical rate point in which it is all simultaneously released.

Maybe you guys are already all over this theory, but I haven't seen it explicitly talked about much around the internet. The common perception seems to be that methane could be a huge problem and we may pass a "tipping point" in which we can't turn back the climate so the methane release will inexorably increase; but if these theories are right then we would see little increase followed by a mind boggling explosion.


Has anyone mentioned the

SEARCH Late August update


range 3.3 to 4.3

I am not sure when the deadline was for that one. Value on 30 Aug was 3.8 but presumably had to be submitted before that.



On Nevens graphs page there is a link to

Allows average of any number of consecutive days. Just have to fetch graphs for different periods and compare. There is even a screenshot for inputs as getting polar sterographic requires custom and certain entries to be completed.

Jim Williams

Chris Reynolds, we are here to watch the data and flap our wings. Feel free to sit back quietly and watch.


Mikkel Fishman

cause a subsurface chain reaction when the pocket pressure is enough to break through its container. This can then occur fractally

Who knew that nature would join the natural gas fracking initiative?


I noticed Watts did not submit for the latest SEARCH estimate. I wonder if he took on board my comment about him dropping his estimate by 1M SQKM between 2007 and 2012 whilst trumpteing "recovery". It would have been impossible for him to have submitted a high enough estimate to avoid that criticism again due to the melt.

I notice the current weather pattern seems to be kicking loss off again but very slowly. I'd bet on about Septmeber 15th at the latest before we see the end of melt.

Re-freeze is another thing. I've been wondering about the mpemba effect for years now. Perhaps we will eventually see if this phenomena actually has an impact in sea ice.

It's going to blow all the anomaly records in October though. Without a doubt.

Dan P.

Thanks crandles! That's a great resource. Somehow I missed it on the graphs page.

Jim Williams

Looks like Leslie will be saying hello to Norway this coming Thursday evening. No idea how much energy she'll be bringing with her.

Mikkel Fishman

Djprice537: I call dibs on producing There Will Be Blood 2


The Godiva2/MET Arctic Sea Ice Concentration and Thickness maps are posted for Sept 6, 7 & 8.


The kmz's for 090812 are posted as well.

One thing to be noted, the winds from the recent low have created significant mostion through the Fram and in the Canadian Arctic. However some spots seem to be thickening already - prehaps due to wind pushing the ice into northern Greenland.



Hi Jim,

According to PolarMet's charts, Leslie will not make it to Norway by Thursday. There is another strong SLP in the Norwegian Sea, with pressure at 971 mb, that will drift northward through Thursday with pressures est. average about 970-975 mb through Thursday.

Those SLP's are equivalent of a Category 2 hurricane.

It has an evil twin off the southwest coast of Greenland over the next several days as well.


Steve C

On that topic of exceptional weather, this forecast is still 10 days away, and thus quite uncertain. The "bottom line" seems to be that the weather up there is increasingly stormy. The ice may not be done with destruction this season. Will we call this one GAC-2012B ?

Chris Reynolds

Jim Williams,

"Chris Reynolds, we are here to watch the data and flap our wings. Feel free to sit back quietly and watch."

No. I'll continue to do what I've always been doing - understand the process that's unfolding.

Geoff Beacon,

I'd tell them to keep a close eye on the research and actively help the researchers where they can. But they're probably already doing that.


Team from Hamburg university is now sure about the NSIDC September minimum : 3.5 +/- 0.0 Mm2


Michael Tabony

The scenario, the summer of 2012 with its great ice loss and ocean warming is half over. All of NOAA's ESRL 2012 methane measurements taken at Barrow, AK, have ranged between approximately 1865 and 1935 ppb (preliminary measurements).

For about a week after 8/4/12 the Great Arctic Cyclone of 2012 destroyed much more sea ice and, very importantly, mixed up the waters of the Arctic Ocean.


The first ESRL methane reading at Barrow after the storm (about 8/14/12) showed a 10%+ increase in atmospheric methane to about 2085 ppb (preliminary measurement). The next methane measurement taken a week later (about 8/21/12) showed an additional 2%+ increase to about 2135 ppb (preliminary measurement).

Is there instrument errors or is there a cause and effect linkage between the storm and the additional atmospheric methane? Did the storm drive enough warmer surface water to depth to disturb the methane in stable compounds there? It sure looks suspicious to this observer.

The real questions now are:

1)how far geographically the atmospheric change will be measurable,

2)will the readings return to pre-storm levels in the coming weeks and months or will they continue to climb,

3)will the additional but smaller Arctic storms of the 2012 season further disturb the methane deposits, and

4)what will this additional atmospheric methane will do to the climate of the Arctic and the globe?


Michael Tabony
"Is there instrument errors "

These kind of readings are not unusual. They are routine and corrected.

Even if it is in the actual atmosphere there are many potential sources.

No need to jump through hoops to imagine some methane release from the oceans.

Al Rodger

"Ocean" temperatures for August may be worthy of comment.
The Ocean Lower Troposphere Temperatures (UAH) were somewhat less than 2007 & 2009-11 (0.4 deg C compared with roughly 0.8 deg C).
Graphing NOAA OI SST above the Arctic Circle shows SST were not quite as warm as 2007. One significant place where SST were warmer was above 80 deg N where they exceeded all previous Augusts & were warmer than all months except September 2007. We await Sept 2012.

Seke Rob

"Minimum Open Thread", saw news flshing by that Shell has started the drilling campaign the Arctic... with a included comment from their side: "A historic event". Mark that in your diary. Start the clocks on when the first cover up spill comes to light.


Michael Tahoney,

As Chris and others have noted, the anamoly is likely instrument error and will be corrected. If you are interested in the satelitte data, which coersa a lerger area, see my previous post above.

Chris Reynolds,

I agree that the AIRS and IASI mapping gies us insight into CH4 release in the Arctic, from the 10 day charts - they may reflect high levels of CH4 at different points in time, so a 3D rendering is going to be the next step in the process of measuring CH4 and other gases in the Arctic, to determine sources and impacts.

I think someone recently posted a series of links to 3D atmospheric CH4 or CO2 modeling (was it you?). It is something I hope to learn more about.



I only find forecasts and charts like others. If hurricanes start listening to me.... Yikes!

Ar Vb

Looks like we passed the minimum extent for 2012
On IJIS on sept 7 with 3664531


I did a rerun of the AIRS/Giovanni data for 08-14-08/21 and found no major CH4 anomaly at 359 hPA over Barrow.

Chris, are you going to have a site with all the sea ice thickness data that you mentioned here previously - and is there a link?



Refreezing is beginning in some parts of the high Arctic. It is -10 C in Alert CA this morning, and the last few days have been mostly below 0 C.


Of course any Cyclone can do further damage to the remaining pack ice, because they are warm! They carry heat from the south as they always do. The mixing of the ice pack already under severe stress is secondary to the heat that melts it. Past summers melts have had several such cyclones, but there was more ice. When the Arctic Ocean ice pack becomes much smaller any storm may do more mixing. But the ice was melted relentlessly during summer over many years to a point where it has become much thinner than ever, hence much more fragile.

Lord Soth

NSIDC is still going down, after a small spike. I wonder if the IJIS will do a double dip this year also.

Anybody noticed that the Cryosphere Today Central Arctic Sea Ice Anomaly has gone off the bottom of the chart at -1.5M km^2


Jim Williams

Lord Soth: "Anybody noticed that the Cryosphere Today Central Arctic Sea Ice Anomaly has gone off the bottom of the chart at -1.5M km^2"?

I suspect most of us have. There are comments on one of the threads about it.


My deepest apologies - I hadn't updated the code which grabs the AARI's image hourly to display it on the Graphs page. Fixed now.
(btw Neven, you should also fix the link to http://www.aari.ru/odata/_d0015.php?lang=1&mod=0&yy=2012)

Espen Olsen


They will have hard times finding a floe 2 meter thick!

Jim Williams

Well, if we are not at bottom we are really close. Any guesses on how flat bottom is going to be?

Given chrome://newtabhttp//ocean.dmi.dk/arctic/satellite/index.uk.php my expectation is a fairly long flat bottom while the excess heat is consumed, followed by a bit of a slowish curve upward at the beginning of the freeze as so much ocean is being covered. I think the real freeze up will be a bit late, making this years graph of anomalous area neatly surround all previous years. The final freeze is expected to be slightly late, but not too much. Maxima will probably be in line with recent years. It's the next Winter where I expect the freeze to begin to fail, not the coming one.

Let's be clear here. This is a guess based upon watching all the conflicting data, both historical and projective. It is my claim that this guess is just as good, if not better, than that guess provided by the most widely quoted models.

Jim Williams

Oops. I know that link is going to be bad. try: http://ocean.dmi.dk/arctic/satellite/index.uk.php

(Why does Chrome provide bad links if you don't open in a new tab????)

Chris Reynolds


Try googling 'inversion model arctic methane' without the quotes.

Inversion modelling is where data such as atmospheric concentration or C13 ratio in CH4 is 'worked back from the location of the observations, using daily wind fields, to try to work out what the sources were.

I can't remember what papers I've linked to here. A good place to start will be my posts on methane at Dosbat, even if you disagree with my conclusions. Those posts have lists of references at the bottom and IIRC I think I managed to get paywall free papers for most of them.

I'd recommend Fisher et al, 2011, Arctic methane sources: Isotopic evidence for atmospheric inputs. I think this is one I may have linked to - it's in my 'recent' folder. That uses a lagrangian model - basically moving atmospheric parcels around using wind data to drive movement of parcels.

To get a paywall free copy of Fisher et al open this page:
And click on the Full Text link, mid right of screen.

A large emission of CH4 from the East Siberian Shelf will betray itself not only in the flask data, and in AIRS*, but also in ratio of carbon 13 to carbon 12. Table 1 and the first part of Fisher is an excellent summary of the ratios expected from different sources. These are measured in 'per mil'. The symbol is like a percentage but with two zeros, o/oo - that's because percent is per 100, but per mil is per thousand so has an extra zero.

Many Arctic sources of methane are depleted in C13, however combining that with large concentrations over the ESS would be a reasonably strong case for calling alarm about the start of a methane blow out. I don't think think early warning is feasible, but detecting the initial stages would be the next best option.

*As a result of the recent paper Dr Yurganov may move over to the European satellite system that allows detection of CH4 below 400mb height. Until then be careful about jumping to conclusions about one month's AIRS data. I still think I;ve shown that there is a persistent anomaly of 400mb CH4 in AIRS over the areas Shakhova and Semiletov have found sea bed emissions. But this is a pattern that keeps recurring. Odd months can show large patches of high methane in that region, these then go. These are probably mixed i from other regions e.g. due to long periods of prevailing winds from the permafrost areas.

It's because I've already been over all this stuff in detail that I get so annoyed with people placing too much emphasis on too little data.

Dan P.

Jim - I agree there is a good chance we will neatly enclose the other years, at least in ice area, given the heat in the water. Extent will as usual depend more on how well the weather compacts the ice over the next couple weeks.

I am still at least partly convinced by an argument made here earlier (by Chris Reynolds?) that the actual minimum in area will come early this year, because the ice edge is so far north that a thin refreeze will win out early. This argument is not inconsistent with the area graph staying in record territory.

Still, that's not really what happened in 2007. I'm still more convinced by reasoning like yours - that there's been enough warm water incursion far enough north that we will have a long plateau near minimum, just like 2007 but at a lower level. It will be a fascinating fall!

Chris Reynolds


The thickness plots are available here:

No plans for a site I'm afraid. I really wouldn't know how to set one up.

When the group Crandles approached got together there was talk of all sorts of ways to transate the original data for public use, but nothing has happened on that front yet. I've lost touch of whether anything is moving forward on that front.

Basically Crandles cracked the basic method I used to draw maps in Excel, and as I was itching to get going on the data because of my obsession with the 2010 volume loss - I just went ahead on my own. One of the group had offered server space and had sorted out a message board for us as the emails were out of hand - so I took up his kind offer and dumped the stuff I'd made for my own use onto that server.

If people want me to kick myself up the backside and do posts on how to access the data in Excel - just ask. I'll get rouund to it, but am busy with other things right now. If I know people want to get the data into Excel themselves then I will post. Beware though. The files are huge (43200 data points). I've found that Excel becomes unuseable with so much data. So what I have are various macros to get data into arrays for processing - after that the results of the processing go into excel.

The problem with me and programming, is I hate programming, I find it taxing and frustrating - I just love the results!

FredT34 is that you?

Chris Reynolds

Dan P.

Please note that what I suggested has not happened as it did in 2007 and 2011.

I'm expecting a more rounded inflection this year. It's very hard to say when it will turn the corner and whether we'll see a 2007 style massively delayed freeze up.

That the CT area loss is continuing makes me think the central polar area is not safe, and earlier/faster rate of melt is all we need now for a seasonally sea ice free state.

Colorado Bob

Famed Austrian peak nearly ice-free

The summit cross on the 3,660-meter Grossvenediger in Austria recently threatened to topple over, as warming temperatures have melted the permanent snow and ice that held the monument in place for decades. Photo courtesy Bergrettung Prägraten.

The permanent snow and ice that has covered the mountain’s peak for at least a century has just vanished within the past few weeks, according to Friedl Steiner, head of the local rescue group, who attributed the melting to climate change.



>"That the CT area loss is continuing makes me think the central polar area is not safe, and earlier/faster rate of melt is all we need now for a seasonally sea ice free state."

I have been thinking my criteria would be a minimum ice volume at maximum of under 20000 km^3 and a fairly early start to the melt.

The volume anomaly graph is making me wonder if I need to rethink that. Why does the volume anomaly start going back up after about day 175? Could this be due to the length of ice edge shortening as we get less ice?

Certainly there is a faster rate of ice melt near edges where waves can break up the ice and water can absorb solar energy before flowwing under the pack.

It is possible we will start getting lots of holes so that the length of ice edge stays fairly high as ice retreats further. But it also appears possible length of ice edge reduces more earlier in the season and if this is a major factor, it could reduce the acceleration towards a seasonally ice free state?


Richaburton wrote:
The remaining pack has massive leads all over it. It must be so weak a big storm would have a really bad effect on it right now. For those without the link:


The leads do look dramatic, and they probably do weaken the ice, but they are not all that unusual. An early spring image from May 2000 is equally dramatic,
and even in March and April, the pack is criss-crossed with leads, as many polar expeditions have learned!


Hey guys. Just to let you know I've contacted the Barrow site to get clarification on the recent flask methane measurements. I had to leave a message with a greenhouse gas specialist. However, if/when I receive a response, I will post the information here immediately.

Best to all and hope Neven's enjoying his time with the fam!



Chris R

"That the CT area loss is continuing makes me think the central polar area is not safe, and earlier/faster rate of melt is all we need now for a seasonally sea ice free state."

I agree with you. I posted yesterday saying, "I believe one of the key months will be June (the month of maximum insolation for the high Arctic); as once it is more or less ice free in this month the seas will be getting so warm I don't think a re-freeze will be possible to any significant extent."

Much like you I don't think we need to be looking at the minimum but how early the melting season starts/how precipitous the fall is. Should ice thin enough and get salty enough due to storms then it will melt faster. This will mean the pack will melt out at a faster rate and crucially expose the sea to the most intense sunlight in June. Once this happens i think it will be "Hasta la vista" winter ice.

Dan P.

Chris R -

I completely agree that this year's melt shows the central polar area isn't safe, so the early minimum plateau is not a pattern to expect going forward. I guess I was thinking of a slightly different argument - that refreeze might proceed briskly (though superficially) in the coming years, due to the high latitudes that are available for freeze as insolation drops. This theory would mean an increase in area that was not delayed far into fall, which given this year's trajectory would mean a pretty rounded minimum.

If instead the refreeze continues to be delayed, like 2007, it will be evidence that warm surface waters have penetrated further north during the melt season so as to delay the freeze.

Now, you can make the argument that once the new ice forms it does a better job of trapping heat in the deeper waters, so that a quick thin freeze is actually worse for overall ice thickness through the winter. But if we do see a pattern of later and later refreeze (which I'm expecting), it would be hard for me to take much comfort in the negative feedback of greater heat loss during the fall. Instead I would take it as a sign that the balance of all feedbacks over the year was doing nothing to halt increasing ice loss.


I'm with you Dan P., but do you think a rounded minimum or a flat bottomed prolonged minimum at this point is partly down to wind/weather patterns still?

I realise this season has been the one to buck the trend in terms of being affected by weather. However, at this late stage when the temperatures are just below freezing incursions of warm air, due to weather patterns, are going to be the deciding factor.

Just a thought.



If a slow freeze-up occurs in the Beaufort Sea, similar to last year's Kara Sea freeze, I might be inclined to think of it as a result of mixed stratification due to the storm. With little fresh water input that hasn't been mixed by subsequent storms it may be that we've got a much deeper column of water to chill before ice can form.

We should have some answers shortly.



Okay! I got a response from Andy Crotwell a NOAA scientist specializing in greenhouse gas emissions. He notes that the wind on the day the Barrow samples were taken was likely from a developed area where methane readings would have been higher. So he's pretty sure these are not representative of the Arctic environment near Barrow and will probably be listed in green along with the other outliers.

Jim Williams

In future years I'm expecting warm water from lower latitudes to be more of a factor. Even this year I think this can be blamed for the ice melting good weather or bad.

I'd like to have a better handle on how mixed the water became this year too, as I think this is what will determine how much ice there is to melt next Spring.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment