« SEARCH 2014 Sea Ice Outlook: July report | Main | Poof, it's gone »


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


As reported earlier: I accidentally deleted the PIOMAS 2014 July blog post yesterday. I managed to retrieve it (with the help of TypePad support), restore it and copy the comments below the blog post. Post is now here.


Well done Neven, However as ironic coincidence goes, CT July 25-26 area dropped 160K km2! I done the same when I wrote my summer 2014 projection in April when I called for much fewer tornadoes throughout the coming season, the next day there was a serious outbreak of them. But all said and done, this year has a very low number of twisters so far. Sea ice may appear to melt slowly, but I appreciate Wipneus animation a great deal. What makes us better than most on this subject is that we have talented contributions capturing key moments. The ice over Beaufort is very fragile.

Another very important point, for a greatest melt to occur the thicker ice must be separated from the pack, otherwise it is much stronger in a consolidated formation.

Colorado Bob

SWERUS-C3: First observations of methane release from Arctic Ocean hydrates
Just a week into the sampling program and SWERUS-C3 scientists have discovered vast methane plumes escaping from the seafloor of the Laptev continental slope. These early glimpses of what may be in store for a warming Arctic Ocean could help scientists project the future releases of the strong greenhouse gas methane from the Arctic Ocean.


Colorado Bob


This is one of the largest Arctic expeditions ever mounted, a 100 day expedition with 80 scientists.

You can follow them here with a inter active map :


The best part, they are on the icebreaker "Oden"

The Norse gods are rolling in their graves.

Colorado Bob


Colorado Bob

Sorry -


What an amazing ship .

The Norse gods are rolling in their graves.

Jim Hunt

Bob - See also the SWERUS-C3 thread on the Arctic Sea Ice Forum:



Thanks Colorado Bob & Apocaplypse4Real for your posts on the SWERUS expedition to investigate Arctic methane. I put up a post about it over on DailyKos.


Heat continues to move into the Arctic ocean from the Atlantic along the edge of the continental shelf. The ice tells us what's happening near the surface but there's a lot more going on at depth.

Hans Gunnstaddar

D said: "Heat continues to move into the Arctic ocean from the Atlantic along the edge of the continental shelf."

The specifics of that warming from the SWERUS expedition article linked by CB:

"Örjan Gustafsson thinks that the mechanism behind the presence of methane seeps at these depths may have something to do with the ”tongue” of relatively warm Atlantic water, presumably intruding across the Arctic Ocean at 200-600 m depths.”

Hans Gunnstaddar


Global warming amplifier: Rising water vapor in upper troposphere to intensify climate change

Date:July 28, 2014

Source:University of Miami Rosenstiel School of Marine & Atmospheric Science

Summary: "A new study from scientists at the University of Miami Rosenstiel School of Marine and Atmospheric Science and colleagues confirms rising levels of water vapor in the upper troposphere -- a key amplifier of global warming -- will intensify climate change impacts over the next decades. The new study is the first to show that increased water vapor concentrations in the atmosphere are a direct result of human activities."

John Christensen

Great analysis, thanks Neven, and very much in agreement with your conclusions.

I rarely hear your opinion on the NAO, but as you see we are likely to keep hovering around near neutral NAO index, which IMO will keep intrusion of warmer, moist air from the Atlantic reduced. The Barents Sea is very cold (relatively), when compared to every year since 2007, and I see no other good reason than lack of Atlantic airborne heat transport since late spring.

2008, 2009, 2010, 2011, and 2012 all had significant negative NAO index during summer months:


2007 however, would still be a different story, as NAO and AO for the summer months of this year don't fully explain what happened, and lack of ice cover in the 2006/07 freezing season was probably an important precursor also.

Back to the current: On the bi-daily DMI 60N weather updates it seems like the night temperatures are starting to dip below freezing, as the 'western' and 'eastern' Arctic shows colder temps alternated by each update. I agree that given the forecast and consolidated central pack we should see reduced (though not stalled) melting in the coming week, unless the next low should cause significant water mixing in the periferal seas.

Jim Hunt

Times may be slow in the Arctic, but not as slow as the BBC's Andrew Neil claims. He reckons that "This summer's Arctic sea ice extent at 10 year record high". Needless to say I disagree!

"Andrew Neil Fails Simple Maths Test"


GFS forecasts high pressure returning to the polar area. There is a tendency towards 2 week oscillations in N polar highs & lows in the summer according to a paper I read last year but don't know where to find the cite. There has been a fairly steady push of ice and cold air from the pole towards the Barents sea this summer. That's why it's cooler this summer.

Plenty of Atlantic heat has warmed up the water off of Norway. The far north Atlantic is very warm. Moreover, a large amount of heat has blown north from the equatorial Atlantic over the past year and a half, cooling the main development region where hurricanes form. In a few more years that heat will work it's way towards the Arctic. Let's not think that a few years of recovery in Arctic sea ice are a trend. There's no evidence of that yet, while there's plenty of evidence in the record of short-term variability.


John, with the North of Equator Pacific sst's becoming one very warm zone, the clouds are bound to take over strongly soon, this means extra heat by advection and cloud cover. The cooling seen now is from clear skies, quite normal, but the influence of heat to clear skies is waning in its place there will be a lot of cloudy days.

I have noted for weeks a great disparity between Jaxa and Cryosphere Today data. Often when Jaxa extent drops CT Area goes up! Using July 27-28 as an example I think that CT has a problem with clouds and the Europeans not so much. So given the near certitude of expansive cloudiness to come hold on to your hats for a little rocky roller coaster ride. But of course cloudy warmer weather implies a later onset of minima day.


If you consider there is about 20,000 km of sea ice coastline next to open water at this moment, if 1 km of it thaws (a reasonable melt). There should be about 20,000 km2 of ice melt a day, of course this reasonable estimate is contrary to a 36,000 km2 increase in area by CT between July 27-28. So the numbers given must be taken with a wide grain of salt. They seem to be an approximation but I think better surveying can be achieved.


OT & without apology!
Greetings to other lurkers: if like me you are not up to making any useful contribution to all the wonderful info & discussion here, then we all still can via the Support button up on top right of this page!
Thanks to Neven & all you contributers here & over at the Forum.
Clare in NZ


Thanks, Clare. I hope to do more next year, once our house is finished. Of course, it will never be finished. I mean, once we live in it. :-)

I rarely hear your opinion on the NAO

That's true. I used to watch the AO index a lot, but I rarely do now (just watch SLP forecasts instead). Maybe I should, but first I have to improve my understanding of both AO and NAO (they are somewhat linked, of course).

Anyway, commenter Friv on the ASIF mentioned the NAO a lot this year, saying what you say too.

John Christensen

Thanks for the feedback Neven!

The impact of AO on arctic ice mass changes is less obvious, and I prefer also tracking forecasts to get more details on where the lows and highs are positioned.

However, the NAO appear to impact both Greenland and the eastern/central Arctic, which e.g. DMI has described in this update for July 2013:


And from this update:

Greenland's seasonal weather patterns are strongly influenced by the North Atlantic Oscillation (NAO). This is a regional scale pattern of high and low air pressure systems that can alternately cause cold air from the north to dominate the island (the positive mode) or allows warmer air from the south to move over the ice sheet (in the negative mode).The pattern of a cooler early July and warmer late July with more extensive melt correlates very clearly with a switch from a positive mode to a negative mode in the North Atlantic Oscillation index which occurred 15th – 18th July. The negative mode of the NAO persisted for much of the record setting 2012 season, promoting the delivery of warmer air and clear skies but has been less prominent this year.

This summer the long stable period of hot sunny weather experienced in Denmark and northern Europe has in part been a consequence of the positive mode with a high pressure system centred over Scotland and to the south west of the British Isles. In contrast, this North Atlantic see-saw means that Greenland had rather cooler weather until the later part of the month when the change occurred. In late July (25th – 28th) the melt area expanded to such a large extent that the model results were confirmed by satellite observations from the MODIS sensor which showed melting in the Saddle region. This is unusual though not unprecedented and coincided with positive daily mean temperatures (up to 1.5 oC) recorded at at an elevation of 1850 m at the PROMICE KAN-U weather station for 3 days in a row (Figure 2). A further interesting feature of the melt area this year is the relatively high melt seen in Northern Greenland.

John Christensen

Hi wayne,

Extent and area will under certain conditions move in the same direction, but under other conditions move in opposite direction:

General freezing conditions:
- Both area and extent increase

General melting conditions, little wind/current:
- Both area and extent decrease

Melting conditions, strong wind/depression:
- Can temporarily cause extent to increase due to fragmentation, and can also cause area to increase due to reduction in melt ponds

Melting conditions, medium wind/ice compaction:
- Can cause area to drop slightly or even to increase if slight fragmentation takes place and melt ponds are reduced, but extent to drop significantly due to melt and compaction at edges

The lack of drop in area this year and last year during the last week of July, I would mainly see as consequence of reduction in melt ponds, which outweighs level of melting at the edges of the pack.


Thanks John, there is something amiss when extent goes down and area rises. It could very well be that melt ponds are freezing. That would be a suitable explanation, but it happened at warmer surface temperatures, note buoy 2013f, there is a picture showing snow floating on water, I doubt they measure things to the meter square, but snowfall on water does this to CT sensors, I proved it a couple of years ago. If extent drops it means the ice has melted or compacted but certainly not scattered.

For area to increase simply means more ice. It can't be because its more compacted, there is a small chance that scattering may increase its value, but I find it a difficult concept, it would mean that more than 15% ice moved into wide open water, however the catch is that extent would increase when area does so due to winds or currents. Back to 2013f at 16z today the surface (2 meter ASL) temperature is -.5 C while the very first thermistor measures above 0 C so it isn't ice on ponds. At least for known measurements.

Hans Gunnstaddar


More mysterious craters found in Russia's remote Siberia region

Two more craters of unknown origin have been spotted in Russia's Siberia region, weeks after a similar-looking hole was found in the isolated northernmost area, a local paper reported.

The Siberian Times, an English-language newspaper, published pictures of two new giant holes discovered by reindeer herders, one located in the Yamal and the other in the Taymyr peninsula, both above the Arctic circle.

The paper said that theories of their origin range from meteorites or stray missiles to aliens or an underground gas explosion. The report could not be confirmed independently.

My best guess is due to the warming tundra, the ground is softening coupled with expanding methane pockets that at some pressure build up blast an opening. Is this yet another new sign of global warming? How many pockets will there eventually open up?

John Christensen

Hi wayne,

The argument for area going up during melting conditions would be that movement and fragmentation of the ice pack causes cracks and leads to appear, which drains melt ponds from the ice surface.

It would be plausible that melt ponds prior to 2007 could remain for longer periods - resting on a stable MYI pack - and that in recent years we are seeing increasingly that the many, fine fragmentations of the thinner, primarily FYI pack causes melt ponds to be drained, not freezing up.

Melt ponds trick the sensors providing us area numbers into believing ice is replaced by water, so when a melt pond is drained, the sensor will see the ice area increasing. If draining of melt ponds outweigh actual ice area reduction caused by melting, the area number will go up. This is why we still like the extent number also, to review both in comparison, since the extent number does not have the 'melt pond flaw'.

Jim Hunt


See also the "Siberian permafrost hole/blowout" thread over on the forum:


Hans Gunnstaddar

Thanks for the link, Jim.

On another note, could the 2014 Arctic melt season be a sleeper? The reason I ask this is because of a military, political and even computer science strategy of divide and conquer. Of course only natural forces are at play here, however I direct your attention to the top left graph, (the false color sea ice map) that once again this season indicates the possibility of open water from the Beaufort possibly linking up at some point with open water from the Laptev. This could occur in August and would in effect divide the Artic ice cap into two areas; one smaller in the ESS and the other in the CAB. The conquer step would come from more peripheral surface area in contact with open water with the possibility of a storm breaking up the smaller of the two. If so, this season could be a sleeper that surprises many. Below is a link and two explanations of divide and conquer.


A divide and conquer strategy, also known as “divide and rule strategy” is often applied in the arenas of politics and sociology. In this strategy, one power breaks another power into smaller, more manageable pieces, and then takes control of those pieces one by one.

Divide-and-conquer eigenvalue algorithm: The basic concept behind these algorithms is the divide-and-conquer approach from computer science. An eigenvalue problem is divided into two problems of roughly half the size, each of these are solved recursively, and the eigenvalues of the original problem are computed from the results of these smaller problems.

Pete Williamson

It doesn't seem too implausible to think come mid-Sept that we might be in pre-2007 territory WRT CT area although lower does seem more likely.

General conditions in the NH midlatitudes more resemble the early 2000's than post-2007 conditions. Tropospheric temps are unchanged for ~15 years. Atlantic Ocean surface temperature, heat content and the heat being transported by the AMOC north all seemed to show peaks in the mid to late 2000's with present conditions a little cooler. Given these are the sources of heat for the Arctic it seems about right that arctic ice should start to return to pre-2007 condition unless things at mid-latitude change. Of course there are internal feedbacks, these may or may not have worked their way through the system for the earlier warmer period, so there may be more melting from that. All things being equal though it seems a small increase in ice, say back to 2003-2006 conditions wouldn't be outside expectations.


Pete Williamson: General conditions in the NH midlatitudes more resemble the early 2000's than post-2007 conditions.

Citation please?

Pete Williamson:Tropospheric temps are unchanged for ~15 years.

Again, Citation?

I'm afraid here, a number of folks will disagree with your conclusion:


Pete Williamson: Atlantic Ocean surface temperature, heat content and the heat being transported by the AMOC north all seemed to show peaks in the mid to late 2000's with present conditions a little cooler.

I presume you're drawing this conclusion based on something you posted earlier elsewhere:


I'll point out this from that article...

The temperature and salinity in 2013 were low compared to the previous 10 years, suggesting that the North Atlantic subpolar gyre has increased in circulation and expanded, bring cooler, fresher water into the eastern regions.

This does not suggest less heat being transported to the arctic; quite the contrary, it implies there is more circulation between the Arctic and North Atlantic, which means more rather than less heat.

I'm afraid your conclusion needs better supporting evidence; I see no return to pre-2007 conditions indicated here.

Colorado Bob

Bad News: Scientists Have Measured 16-Foot Waves In The Arctic Ocean

For the first time, waves as tall as 16 feet have been recorded in Arctic waters. If these waves are speeding the breakup of the region's remaining ice, as oceanographers suspect, they could signal the birth of a feedback mechanism that will hasten the Arctic's march toward an ice-free summer.
During a September 2012 storm, University of Washington Researcher Jim Thomson detected wind-generated waves as high as 5-meters tall. While researchers have known about Arctic melting for decades, Thomson says in a statement what we're seeing with waves of this size "is potentially a new process, a mechanical process, in which the waves can push and pull and crash to break up the ice." As he notes, in the latest issue of Geophysical Research Letters (emphasis added):

It is possible that the increased wave activity will be the feedback mechanism which drives the Arctic system toward an ice-free summer. This would be a remarkable departure from historical conditions in the Arctic, which potentially wide-ranging implications for the air-water-ice system and the humans attempting to operate there. The increasing wave climate will also have implications for the coasts, which are already eroding rapidly during summer months as a result of climate change and subsequent loss of permafrost.


Jim Hunt

Bob - Regarding big swells in the Arctic see also http://forum.arctic-sea-ice.net/index.php/topic,624.msg27178.html#msg27178 et seq. on the Arctic Sea Ice Forum.

Jim Hunt

Over on the forum once again Jim Pettit recently quoted Thomas Jefferson as follows:

Ridicule is the only weapon which can be used against unintelligible propositions

Very much in that spirit Great White Con are proud to present a fight to the death. In the red corner "Snow White". In the blue corner "Steve Goddard". At stake the title of "Greatest Arctic Sea Ice Prophet on the Planet":

Shock News! Breathtakingly Ignorant Deranged Blogger Batting 1,000 This Summer

Be sure to check out the archive footage and the comments. Be even surer not to inadvertently end up on (un)Real Science.

Colorado Bob

Over on the forum once again Jim Pettit recently quoted Thomas Jefferson as follows:

"Ridicule is the only weapon which can be used against unintelligible propositions"

It’s not the distance, it’s how fast we are crossing it. The largest city in Brazil has less than 100 days of water. They have driven the price LNG on the world markets , because water doesn’t flow in their damn's turbines any more.

19 million people running this test tonight.

Another 45 million in Calf. , Az., Nev., Ut., N.M. ,Co., Wy., right behind them.

Hans Gunnstaddar

Below is a link to the end of the movie about James Balog's retreating glaciers research in the movie, 'Chasing ice', to the song by J. Ralph, 'Before my time'.


Pete Williamson


I have to be brief and write in note form

i dont have citations was using KNMI climate explorer to look at data sets e.g. For the NH midlatitudes I took Hadcrut4 (20N to 60N) and get this

OHC for the Atlantic is NODC I get this

For AMOC, yep I like the Elliott line work, I think there is more insight to be gained from it than just the quote you highlight. But as well as that work there is also a study called US AMOC/UK RAPID which is shorter but shows a similar situation. You need to hunt for poster and publications based on that study

(Id give you a fuller answer if I was on a more user friendly device sorry)

Jim Hunt

From the ridiculous to the sublime.

Arctic sea ice is NOT at a ten year record high extent, so on to round 2:

Mr. "Goddard" needs to take a natural sciences course, or simple 101 physics might be enough. on Facebook

"Snow White" suffers a most unfortunate wardrobe malfunction on Twitter


Arguing with imbeciles is imbecilic. Goddard doesn't have a clue. Nothing you can do will change that.

Jim Hunt

Thanks for your concern David, and I do realise that!

Have you explored any of my links carefully?

If so can you determine any method in my apparent madness as yet?

John Christensen

Hi Pete W,

I would agree with your point, if you are trying to say that the AMO is important, and that there is a nice correlation between the current warm phase of the AMO and the Hadcrut temps for 20-60N, with the graph that you mentioned.

Global Warmning fanatics will probably claim that the AMO is rather irrelevant, while climate change deniers will claim that changes in the AMO is driving the temperature change and CO2 is irrelevant, or a miniscule factor.

The truth evidently is between the two extremes:

- There is a strong correlation between the phase of the AMO and changes in global surface temperatures
- BUT, the global temperature in the negative AMO phase of early 1960's to 1995 did not quite return to the level of the last negative AMO phase, early 1900's - 1930
- And yes, the change from negative to positive AMO phase seems to explain part of the global temperature increase reaching the peak in 98' combined with the strong El Nino, and also is part of the explanation why the global temperature did not continue increasing as fast after 98'

Then consider PIOMAS ice volume data:

January 1, 1980: 25.039 km^3
January 1, 1994: 22.321 km^3
January 1, 2008: 16.648 km^3

It dropped 10.9% between 1980 and 1994, and then 25.4% between 1994 and 2008.

So, yes, the move from negative to positive AMO had a significant impact, but it only accelerated, what was already happening.

There is therefore no return to prior state of ice e.g. 10 years ago, even if we have impressive ice area and extent numbers these days. If ice volume would reach the level 10 years ago, we can talk..

Jim Hunt

David - This blast from the past seems to have gone down well on the forum, so I'll repost it here for more general consumption.

It may be of some assistance to you as you do your due diligence on my disturbed psychological condition:

The Great White Con - Update 3 from Jim L. Hunt on Vimeo.

P.S. Or at least it would if I could work out how to embed it! Is there a Neven in the house?

Chris Reynolds

Pete Williamson,

It will take many more years for ice state to return to pre-2007 conditions. 2007 removed a lot of thick ice. Concentrating on extent and area at the expense of volume isn't using all the available information.

John Christensen,

The AMO might have a role in ice retreat, especially in the Atlantic sector. However the AMO index proper has the signature of AGW driven ocean warming removed. Once that is done one has to wonder whether the the AMO has a role or if it is AGW driven ocean warming that has had a role.

Using data from this site:
I have plotted the 'detrended' AMO and the raw Kaplan SSTs.

The AMO did cross zero on the upslope around 1995, and has crossed zero on the down slope this year, so could be taken as being the start and end of a limited period of volume loss. However Kaplan SST suggests that if SSTs are the issue then they cannot explain the last two years of sea ice as a phenomenon of the declining AMO - they remain historically high!

Furthermore the start of the recent period of volume loss may have nothing whatsoever to do with the AMO. I have blogged earlier this year about what caused the decline in volume in the PIOMAS model.

Taking the interannual difference between successive september NSIDC extents and plotting together with the cumulative sum of the differences shows that the period where declines in extent outweighed recovery started in the mid 1990s.

However Lindsay & Zhang put forward a theory that this period of decline after the mid 1990s was due to a three stage process.

With the trigger, after decades of early thinning, being the PDO and AO indices, and ice albedo feedback taking hold thereafter.

I've argued recently that we're at an inflection in winter volume because first year ice growth is taking over from multi-year ice volume loss. If I am correct the reultant steadying of winter and summer volume loss is bound to be misdiagnosed as being due to the AMO going negative. Ho hum. :)

Steve Bloom

"I've argued recently that we're at an inflection in winter volume because first year ice growth is taking over from multi-year ice volume loss."

There's been a long-standing argument that some degree of inflection in loss is to be expected simply because ice at higher latitudes is easier to form and harder to melt. Given that the mobility of sea ice points to the inevitability of big multi-year losses once substantial ice-free areas encroached on the central basin, your formulation seems to be a different way of saying the same thing.

Chris Reynolds


Basically it's because first year ice can reform quickly in the winter, whereas it takes years for MYI (multi-year ice) to 'bounce back' from losses.

Recent PIOMAS maximum volumes have been about the volume one would expect for a pack with an average thickness of around 2m. I think the previous volume loss trend, which was due to loss of MYI volume, has come to an end. We now have a pack with a residual of MYI that is mainly first year ice. Winter volume losses will level out, summer volume loss, and hence extent/area losses will level out too, I suspect.

Kevin O'Neill

Steve, we can interpret Chris' rationale in terms of the 'Tietsche Effect'. If we removed all of the ice on September 1st, winter temperatures would cause almost the entire Arctic to still be covered with ice come spring. In most places this ice would have a thickness near 2 meters since winter temps are cold enough to cause that much growth over the winter.

Until winter temps rise enough to limit growth, we are going to start each spring with an equivalent amount of ice. Meanwhile, most of the energy accumulated during summer is lost to space during fall/winter.

If the average winter growth was 3 meters we would have seen this happen sooner; if it was one meter we'd still be approaching it.

Kevin McKinney

Kevin's comment sounds like a pretty good insight to me.

We'll see how well it plays out.

Jai Mitchell

Kevin's comment neglects to include the feedback mechanism of increased humidity during the late September early November period after an ice-free event.

this effect will help to retain heat and was the primary mechanism for plant survival during previous houthouse climates.

Jim Hunt

As a self styled world expert on Arctic surfing I was compelled to give the Daily Telegraph a very stern talking to yesterday.

After a serious disturbance in the force and a DMCA takedown the waves are now subsiding on an ebbing tide.

"Break out the bubbly?"


Jim, the remainder sea ice over the Basin can be subtly deceiving, it is quite interesting to read buoy ice profile temperatures which reveal the extremely fragile existence of first year ice at the same temperature of the sea, particularly over North Beaufort area. Once surrounded by open water the sea warms and therefore the ice can vanish extremely quickly. So sudden to astound as we can witness now. And therefore there should be many more days of significant vanishings. As I wrote before, each year has its own thing going, and ice shores surrounded by 20,000 km of much warmer sea water may trigger some significant surprises.

It is a matter of judgement or lack thereof to say that this sea ice has been recovering, if it was it would be like 3 meters thick rarer sea ice similar to 2014D, amazingly resilient and readily colder than the air above and water below by several degrees centigrade. To write that first year ice albeit apparent from space for now, but having the same temperature of the sea "has recovered" because the extent and area (again for now) is wider, illustrates the commenters lacking crucial skills, I salute you for your efforts in pointing this out especially for those dedicated to learn.

Hans Gunnstaddar


Global Warming Kicks Up Record Pacific Trade Winds

Rapid warming of the Atlantic Ocean is "turbocharging" Pacific equatorial trade winds, according to new research. These are the strongest trade winds since recording began in the 1860s, according to scientists from the University of New South Wales and the University of Hawaii.

"The increase in these winds has ... amplified the Californian drought, accelerated sea level rise three times faster than the global average in the Western Pacific and has slowed the rise of global average surface temperatures since 2001," the study's authors report.

This pressure difference between the two ocean basins isn't expected to last. And as previous research reported, when it does end, a sudden acceleration of average temperature around the globe would likely occur.

"It will be difficult to predict when the Pacific cooling trend and its contribution to the global hiatus in surface temperatures will come to an end," co-author Matthew England said. "However, a large El Niño event is one candidate that has the potential to drive the system back to a more synchronized Atlantic/Pacific warming situation."

Colorado Bob

Why an 1879 Voyage Is a Time Machine for Climate Change
A doomed Navy expedition kept exacting records that show the rapid weakening of the polar ice cap.
—Mr. Sides is the author, most recently, of “In the Kingdom of Ice: The Grand and Terrible Polar Voyage of the USS Jeannette,” which will be published Aug. 5 by Doubleday.

Over the past year, an international team of climatologists and historians, working with the National Archives, has dug back into those historic logbooks and started digitizing and analyzing De Long’s work. “The data De Long gathered is quite valuable and amazingly thorough,” says Kevin Wood, a scientist affiliated with the National Oceanic and Atmospheric Administration. “The Jeannette was well-equipped for science, and it was the first vessel ever to go through that part of the Arctic.”

The climate-change story that De Long’s logbooks tell is a sobering one: The once impenetrable polar ice cap, at least in that 1,000-mile swath of the High Arctic north of Siberia, has shrunk, weakened and thinned far faster—and far more dramatically—than anyone had realized. Dr. Wood has taken dozens of research trips to the waters north of the Bering Strait and has closely compared recent ice conditions with those described by De Long. “If the Jeannette were embarking today in the same season,” says Dr. Wood, “she probably wouldn’t find any sea ice to get stuck in.”

The extent to which the ice in the northern Chukchi Sea has shrunk since De Long’s ordeal is astonishing. “For the most recent decade, satellite data show that summer ice concentrations in the Chukchi Sea south of the 80th parallel have declined by as much as 70%,” Dr. Wood says.

As the extent of the ice has shrunk, its quality has weakened. De Long’s journal speaks of enormous pressure ridges and towering bergs of multiyear ice unleashing a din of shrieks, groans and shuddering explosions. “The kind of ice that De Long describes seems almost fantastical to me,” says Dr. Wood. “In his journal, he repeatedly describes ridges and grounded floebergs that are as tall as 60 feet that have run aground in 150 feet of water. You just don’t see this kind of ice in that part of the Arctic anymore.”


Jim Hunt

Wayne - Thanks very much for your kind words, and I can assure you that it is my pleasure!

Regarding ice mass balance buoys in the Beaufort Sea, are you familiar with the work currently being undertaken by the British Antarctic Survey?


At least one of their buoys does indeed look to be on its last legs. I give you BAS IMB 4:

That one looks like it will be gone long before IMB 2014B, which is currently speeding in the direction of the Bering Strait!

Crozet Dutchie

Book review: “In the Kingdom of Ice,” polar voyage of USS Jeannette, by Hampton Sides: An additional very interesting book review in the Washington Post:


Jim, it will be good to capture buoy data when the ice obliterates or vanishes. This will serve as an example , especially day before and after look, as epiphyte's work in Neven's latest example. On 2014b, the sea water and ice temperatures are totally undistinguishable, meaning any rise of sea water temperature will obliterate the last standing ice. Given that it will be sunny for a while, this ice will likely disintegrate during evening. If cloudier when the sun elevation may be important, the process is almost the same but slower.

Pete Williamson

I thought my original position was consistent with AGW theory. without extra heat in an accessible place in the NH then there is no reason for the Arctic to keep warming. Of course internal feedbacks may still be in play but at some point a new equilibrium is going to be reached until warming recommences.I dont have any firm idea how long (or how much) feedbacks continue to amplify warming beyond the initial external forcing but I do think nobody is suggesting its a runaway process.


Chris, Kevin:

You're correct that an Arctic Ocean composed predominantly of first-year ice is significantly more stable than one composed of older ice, because winter ice growth tends to be much larger where the ice is thin than where it is thick. I think you're overstating the degree of this stability, however. Verbal descriptions of mathematics being as imprecise as they are, probably we're agreeing on the mathematics and disagreeing only the words which are appropriate to describe it.

On the ice edge margin, the area covered by winter ice is different in different years, particularly in the Barents and Kara sectors. This does cause a large heat flux for a not terribly huge area change.

For our purposes, it is sufficiently accurate to neglect the skin transfer insulation between the surface of the water and the air, and also the heat capacity of the ice. If we also neglect the heat transfer to the ocean, the square of the ice thickness then changes directly proportionally with the integrated difference between the air temperature and freezing.

Eyeballing the DMI 80 degree temperature graph, during the last year's record warm winter, this difference was about 5/6 of the long-term normal, which would result in ice about 1/12 less thick. If the normal first-year thickness is 2m, this exceptionally warm year would have a thickness of 1.83m. Ice also only increases from the start of freezing, so a delayed freezing would decrease the thickness as well, but unless the open water lasts late in the season, the thickness change due to this is also similarly small.

If the initial thickness was 1m instead of 0, the final thickness would increase from 2m to sqrt(1*1+2*2)= 2.23m. If it were 2m, the thickness would increase to sqrt(8)=2.83 m.

So, yes, the ice thickness does converge reasonably strongly on the value of everything covered in first year ice at the normal climatological formation rate. The problem is that with the very low ice volumes which we have had recently, small volume losses lead to large albedo decreases. By September 2012, there was only 16% of the PIOMAS ice volume there had been in April. We wouldn't have needed much more heat in the right places to get rid of the ice entirely.

So long as we continue to not allow a change in ocean heat flux, it seems very unlikely to me that we pass a climate tipping point. But what if, say, we have extensive open water in the summer, waves cause the mixed layer drops to around 100m like much of the rest of the world's oceans, and we store 2C of heat there. This would be enough to melt around 2.5m of ice. Much of this heat would come back up in the winter, when it would melt the ice from beneath (or reduce bottom freezing), when it was insulated by a thick layer of ice. Spring would start with much less ice, and much of summer might be virtually ice free. The earlier albedo change multiplied by incoming insolation would be much larger than we have seen to date. This could possibly store enough energy to repeat the cycle the next year. It's not clear that there's a tipping point here, but it's certainly not clear that there isn't. Warming winter surface temperatures is not required, only deep ocean storage of summer heat with some of the heat returning as bottom melt in winter and spring.

So long as there is still surface ice to melt, added heat is trapped as surface melt, and mostly dissipates as extra surface heating in late fall and early winter. It's when the added ocean heat is substantially above that required for late summer melt of all the ice that we see carryover of the added heat into significantly lower April ice thickness. The maximum of albedo feedback occurs much later in the melt transition than most people here seem to suspect.

Chris Reynolds


Yes, numerical argument should take precedent over verbal.

However given this, and given that models are the most comprehensive numerical argument available, is it not significant that in models the growth thickness feedback overcomes volume loss from MYI resulting in a long tail? This is seen in 'forecasts' using PIOMAS, and Overland and Wang have twice shown that models (GCMs) that best repicate the seasonal cycle predict ice free by around the 2030s and no rapid crash in the 2010s or 2020s.

Largescale overturning of the Arctic Ocean allowing Atlantic Water to more actively engage in ice loss has often been posited as a mechanism by which a rapid final loss of ice may occur. But this mechanism does not appear in GCMs (as far as I've read), and if it were to appear it might be expected to result in a rapid transition. However it is possible it does occur in some models, only for the heat to be vented in the subsequent autumn and winter (i.e. Tietsche et al's removal of all ice from 1 July).

I had thought that small volumes would lead to strong albedo feedback and open water formation due to thinner ice.

Consider the following graphic.
The blue line shows percentage open water (%OW) formed as a function of April thickness. It can be seen that, going from thickest ice to thinnest, from the right the %OW formation for thick ice is low. But around 2m the %OW increases dramatically.

The red line shows average volume for each 10cm April thickness band during the 1980s. It can be seen that in the 1980s the peak volume was well away from the non-linear change in %OW formation around 2m.

However now consider the situation since 2007.
Here the peak volume has moved down to the region of non linear change in %OW formation. This has resulted in greater opwn water at the end of the melt season, and greater volatility in response to weather due to interaction between the thickness distribution in April and the steepness of the %OW curve.

The problem is, if during each winter ice re-forms to roughly the same thickness the peak volume doesn't get the chance to move down taking even more of the ice into the high %OW formation region.

Opposing the possibility of heat building by surviving the winter is the enormous amount of heat that can be vented in autumn after sunset in the Arctic. Consider the behaviour of PIOMAS thickness of ice in the Arctic Ocean.

Year Arctic Ocean Volume (k km^3)
1998 25.4
1999 24.5
2000 23.5
2001 24.0
2002 23.7
2003 23.7
2004 22.5
2005 22.7
2006 21.7
2007 20.3
2008 21.1
2009 21.2
2010 20.8
2011 19.3
2012 19.4
2013 19.6
2014 19.3

Now think about the summers of 2009 to 2013, all very, very different. The energy gains of these summers will have been very different, yet all those summers were followed by very similar April volumes.

The series is very short (2011 to 2014), but bear with that for a moment: This behaviour suggests to me a very powerful negative feedback with ample capacity to dump energy into the atmosphere, grow ice volume over the winter, and damp the trend of volume loss. Because despite those very different end summer amounts of energy accrued there is no indication of reduction of winter volume growth (the reverse is true).

Jim Hunt

Chris - Have you been following the forum discussion about increasingly destructive swells in the Arctic basin?


According to Jim Thomson:

It is possible that the increased wave activity will be the feedback mechanism which drives the Arctic system toward an ice-free summer. This would be a remarkable departure from historical conditions in the Arctic, with potentially wide-ranging implications for the air-water-ice system and the humans attempting to operate there.

I'm currently researching an "in depth" article on this very topic. Unfortunately the US Navy tell me "I don't think that there is much the Navy would want to contribute, at this point" :(


Yes, Tietsche 2011 does in fact simulate the exact situation I was referring and concludes that disturbance disappearance is rapid. If the water column is completely mixed to the bottom of the depth of heat penetration, cooling makes the column slightly unstable, any added heat comes out almost immediately, and you have the situation described in the paper, where there is minimal heat memory over the winter. If, on the other hand, the water column at the end of summer is slightly stable, surface ice can form relatively quickly, while the surface motion of the ice will create mixing which brings up heat from the ocean, and more of the heat from the previous summer is retained as thinner ice at the end of the winter.

It is encouraging that this one simulation finds that the former result is correct, but there is poor agreement amongst different simulations about near-surface Arctic Ocean temperature and salinity profiles, as well as poor agreement with measurements. The possibility of significant methane hydrate melting along the Siberian shelf was for a long time dismissed because models showed that the heat could not reach the sea bottom. Modelers generally dismiss the notion of significant heat flux from deep water at the southernmost part of the Eurasian Basin along 90 East, on this forum generally known as the "Laptev Bite", but somehow polynyas that they are unable to explain still often form there. I consider the issue of the existence of significant inter-annual ocean heat storage under sea ice still open.

Here is an interesting paper about Beaufort winter ice melt. It shows a significant part of summer stored heat coming up during winter storms, but the total heat storage here is small, and the heat flux ends fairly early in the winter, generally attributed to convergent Eckmann pumping in the Beaufort making the water column too stable for a significant heat flux by midwinter. Unfortunately, most of the literature seems to focus on the area which barely melts out at the end of summer, where the end-summer heat content is fairly low.

The thick ice region in the northern Canadian Archipelago and Greenland from east of McClure Strait to west of Fram Strait currently has strongly convergent ice motion, and this will only get stronger as the ice norther of this blockage thins. All of the channels will eventually open up for months during summer and fall, leaving only the thick ice region east of Sverdup Channel. As predicted by models, I expect that there will still a long tail of a small amount of sea ice in this region until enough heat builds up around it for heat transport into the region to overwhelm annual regrowth of ~3m ice. While this tail is melting out, the largest albedo forcing is earlier summer and fall meltout of annually melted ice, and not the melt of the last remaining perennial ice.

The comments to this entry are closed.