I notice the ice volume is given in terms of anomaly. It’s understandable in many cases why anomaly is more convenient and meaningful, but in this case, I think it would be interesting to see the absolute value (because near zero is not inconcievable at some point in the forseable future).
Area and volume records should fall easily, but extent could be the sceptic’s salvation. Traditionally, the difference between extent and area was minutia. The vast chunk in the middle was solid and thick. Area pretty much equalled extent minus the comings and goings of ice on the fringe.
Now it’s all getting fringey.
Has any run of any major model approximated the actual behavior of the ice over the last 10 years, including a reasonable extrapolation of 2012? Or, what’s the worst run ever spit out by a current model? A graph would be wonderful :-)
IF the (truly visually spectacular) map on the Tamino link attributed to “apocalypse4real” can be trusted, then there appears to my untutored eye that a substantial portion of the remaining thick ice is teetering on the brink of being flushed down the Fram Strait.
(And I said “IF” because this map doesn’t seem to correlate particularly well with other Arctic Ice maps. A great pity if it’s inaccurate, because it could be a magnificent ‘at a glance’ resource.)
Please, not again a discussion about the surface of sea ice while thickness disappears faster below. UCL just announced CryoSat data reveals it’s melting faster than predicted (though no data published) and it is very hard to extrapolate PIOMAS data and see an ice free September any later than 2020.
In Summer 5 million square kilometers of extent might look healthy, but if thickness is just a few centimeters it’s all gone within days. There is thickness distribution – of course, but isn’t MYI not getting thinner every year?
Neither area nor extent tell the story of what the ocean is doing the ice from below. Ok, sometimes one can observe how fast ice or regions of floes melt out completely without any sign of drift. Is in situ flash melting a sign from the future, a precursor of what’s coming when sea ice is that thin that extent shows binary behaviour, meaning yesterday there was ice today surprisingly not?
When I go skating first acquisition is thickness information and not size. Everybody living close to a lake freezing in Winter knows thickness melts gradually extent not. Actually it reduces very fast from 100% coverage to zero during the last days.
Sure, it is a good thing there is a long record of extent data. The media will publish the message of a new extent record rigorously measured by all scientific means. But is the message being transported correct or is it just about the surface of the real problem?
Show any journalist the extent record of the satellite era and he will not consider an ice free Arctic during his lifetime. It is not enough to add we found the ice pack is shrinking faster than thought or predicted or calculated.
My impression is the science of the Arctic Ocean still reveals more surprises than facts. Hopefully, once the ice gone there is more to say as: Oups, that was not foreseeable. Given there are only a few years left to achieve the needed scientific self-awareness to give a proper forecast – what can be done?
How many buoys are needed for a complete assessment? Who (names + contact, please) decides Arctic science budgets? Why papers still get published behind paywalls? Where is a sea ice model describing single floes? How to get more reporters with cams on research vessels? Why do researcher stay away from blogging?
Personally I learned two things from the last storm: one million square kilometers of ice can disappear within one week and the public is aware and interested. The Arctic is the place where the consequences of anthropogenic climate change will be visible very soon, even from the moon.
Most probably half of the planet will have to adapt to new weather pattern an ice free Arctic spawns. What we can’t afford are surprises.
I think again it really useful to find and post 2012 minima past predictions by contarians on how wonderfully restored sea ice would have been if they were right.
No credible journalist will interview an utterly incapable with predictions contrarian . Yet they appear again and again as serious climate critics. But do not expect them to brag about their failures, we have to be vigilant in building their skill reputations. I liked Gareth’s piece, but there is plenty more satire from the contrarians if someone would be nice enough to do a collage .
It is a strange strange world we live in when a large community of people preclude correct science and predictions and follow made up on the go presenters as being equivalent to solid science work giving remarkable successes.
10 Arcticio said, “Given there are only a few years left to achieve the needed scientific self-awareness to give a proper forecast – what can be done?”
Well, we can pop some corn. We get to be front row center to the permanent transformation of summer conditions in the northern hemisphere. Brand new set of weather patterns! Reminds me a bit about the old short story The Lottery.
Thanks for all the links to great graphs, everybody. What does this Arctic Ocean heat mean for the weather/climate down here? I know you can’t answer this question, but I keep wondering if the farmers should plant wheat [a dryer land crop] instead of corn.
PS: The drought is over near Davenport, Iowa. Some corn has been de-tasseled, which leads me to believe that the farmers still have some hope for a corn crop this year. Some corn plants have turned yellow and died, but not many. The soybeans look taller.
The whole cycle of summer sort of got put on hold during the drought.
10 Arcticio is correct in saying: “What we can’t afford are surprises.” The year seems to have a new rhythm.
Is the Arctic cyclone something new, or just something that didn’t get reported before? Is there a new pattern of polar oscillations?
Arcticio: “Please, not again a discussion about the surface of sea ice while thickness disappears faster below.”
There’s a graph of PIOMAS volume, so I think you’re being a bit harsh. Further, I think the distinctions between extent, area and volume are going to be largely academic within the next ten years, or so. When there’s hardly any ice left at the minimum, it makes not a jot of difference how many dimensions you measure it in.
Arctic sea ice science will then rather swiftly move onto how well the seasonal opening and closing of the North-West Passage and Northern Sea Route can be forecast. A lot of that depends on the weather, of course, but I reckon that’s where a lot of the interest will be.
Patrick027 @3 & @4
If you poke about a bit at Neven’s website, you will note it has more than one page of graphs & enough interesting stuff crops up in the various discussion threads that would overload even ten pages of graphs.
Re the PIOMAS anomaly graph. It gives a spectacular slope but little else that is comprehensible. The more recent introduction of their ‘comparative-years’ graph doesn’t help so much either as the rate of decline or changing rate of decline is not particularly evident.
Happily, with PIOMAS data now public we don’t need to resort to scaling graphs with a ruler to produce graphs of our own.
One that I produced compares the anomaly for each year here (usually 2 clicks to ‘download your attachment’) which shows the melt is occuring earlier through the summer before a bit of ‘regrouping’ by the time of the annual minimum arrives.
1: Neven..Thanks for the multi graph link, it’s an excellent source of info even though some graphs still have an annoying 1 month delay.
10: Arcticio. I feel that when that fateful summer dawns soon when there is only open water in september that that will be the end of the anolomalies, from then I feel each consecutive summer from then on will be ice free as there is no hard perrenial left to act as a frozen raft for falling snow. I think you will see from then on a rapid lengthening of time when the area is ice free. Millions of square kms of open ocean in the middle of summer is going to heat rapidly and I think it would be quite possible that within our lifetimes assuming you are 40ish like me that we will witness at least half the year without ice cover.
Comment by Lawrence Coleman — 13 Aug 2012 @ 6:20 AM
Wow, hours later and there are numbers from UCL: 7,000 km³ last summer. Here the BBC interview with Seymour Laxon:
#19 Excellent work Tamino, it is a must see graph for especially journalists wondering if AGW is happening, post it everywhere like a coca-cola sign.
Arcticio, the large Arctic cyclones are part of a newish trend of Lows penetrating the North Polar regions with increasing frequency. It is linked with the disappearance of thick multi-year ice. Was seen this past winter a great deal. NASA models forecasted this quite sometimes ago. It is caused by the shrinkage of the entire polar-temperate to tropics atmospheric interfaces, in other words the hurricanes have moved Northwards with less juice to fuel them.
Seems ironic I was mountaineering in British Columbia doing my own personal research on glacial retreat and what should be up at Real Climate and Open Mind when I’m back but a discussion of ice loss. The route I’d climbed several years ago is “no longer practical on account of glacial retreat…”. How ‘bout that. Don’t argue Climate Change with a mountaineer.
Thanks for all the links to great graphs, everybody. What does this Arctic Ocean heat mean for the weather/climate down here? I know you can’t answer this question, but I keep wondering if the farmers should plant wheat [a dryer land crop] instead of corn.
I think the question is “How fast” we approach ice free conditions, which accelerate climate forcings. We approach chaotic new conditions for the entire northern hemisphere with all kinds of system interactions and new phenomena, of whom some can be observed already. For instance extreme weather phenomena from polar amplification and sea level anomalies. I would really like to read more of the impacts and the increase of forcings from albedo flip effects.
Re 29 Geoff Beacon – great link; I haven’t finished it yet but plan to (the youtube part) – PS I haven’t watched this yet either but it may be related (somebody else originally posted this link somewhere on RC I think; don’t remember where, though):
Maybe this is a good description what is happening and which is indicated by large scale arctic sea ice melt..
Elevations and ages of drowned Acropora palmata reefs from the Caribbean-Atlantic region document three catastrophic, metre-scale sea-level–rise events during the last deglaciation. These catastrophic rises were synchronous with (1) collapse of the Laurentide and Antarctic ice sheets, (2) dramatic reorganization of ocean-atmosphere circulation, and (3) releases of huge volumes of subglacial and proglacial meltwater. This correlation suggests that release of stored meltwater periodically destabilized ice sheets, causing them to collapse and send huge fleets of icebergs into the Atlantic. Massive inputs of ice not only produced catastrophic sea-level rise, drowning reefs and destabilizing other ice sheets, but also rapidly reduced the elevation of the Laurentide ice sheet, flipping atmospheric circulation patterns and forcing warm equatorial waters into the frigid North Atlantic. Such dramatic evidence of catastrophic climate and sea-level change during deglaciation has potentially disastrous implications for the future, especially as the stability of remaining ice sheets—such as in West Antarctica—is in question. http://geology.gsapubs.org/content/23/1/4.abstract
If you are looking for information on where the various sea ice concentration data in the graphs are coming from, check out the Climate Data Guide overview of long-term sea ice data sets. And yes, the major data sets all show the long-term (over 30+ years) decline in Arctic sea ice. There are some interesting systematic differences in estimates of absolute sea ice extent and area.
I’m wondering if there is a direct connection between ice minimum and methane emissions? Are there any graphs depicting that interaction? Are there any predictions regarding how much methane would be released if arctic ice disapears in Summer?
GIF versions at same location but with gif extension.
Comment by Andy Lee Robinson — 13 Aug 2012 @ 8:34 PM
Sea-ice switches and abrupt climate change
We propose that past abrupt climate changes were probably a result of rapid and
extensive variations in sea-ice cover. We explain why this seems a perhaps more likely explanation than a purely thermohaline circulation mechanism. W e emphasize that because of the signi¯cant in° uence of sea ice on the climate system, it seems that high priority should be given to developing ways for reconstructing high-resolution (in space and time) sea-ice extent for past climate-change events. If proxy data can con¯rm that sea ice was indeed the major player in past abrupt climate-change events, it seems less likely that such dramatic abrupt changes will occur due to global warming, when extensive sea-ice cover will not be present.
Keywords: climate; sea ice; thermoha line circulation ;
abrupt climate change; glacial cycles http://www.seas.harvard.edu/climate/eli/reprints/Gildor-Tziperman-2003.pdf
#38 Pprokaryotes, I would add : overall sea ice thickness dictates the height of boundary layers which in turn affect the lower atmospheric temperature profiles, in effect changing the weather of the entire lower atmosphere. Wrote a paper on that, looking for a publisher.
To complement Tamino’s awesome animated sea ice volume graph, here’s a very simple multi-year PIOMAS animation based on a polar grid, with radials set at single month intervals, and decadal averages thrown in just because:
I’ve wondered what was the purpose of the sea ice extent metric. It seemed goofy to me since Area x Volume is the calculation for the amount of ice. It was explained to me (recently and yet I’ve already forgotten where) that it’s of use to shipping. How far south can you still expect to find coherent patches of ice? That sort of thing.
1884 was the culmination of the first International Polar Year (more than one year; it was 1881-1884). First few minutes of googling didn’t come up with total area or extent in km3 (unsurprisingly), but I did find a map at http://www.arctic.noaa.gov/aro/ipy-1/Frontpage.htm
I’m not sure what the different colors indicate, but my best guess is that the darkest blue/green is open ocean, white is solid ice, and intermediate colors indicate some concentration of ice, but not solid. Even if only the solid white represents what was left at minimum, that’s a LOT more ice than we’ve got now.
I guess I don’t know for certain what AlaskaHound @8 was trying to say, but after poking around the data and images for IPY 1881-1884, and particularly that map, I’m pretty sure he/she was very wrong.
One indication of the minimum ice extent in the 1881-1884 is that the coast lines of many of the arctic islands were still unknown, including areas that are ice free today. The lack of mapped coasts implies multi-year ice that had prevented the coasts being mapped. The islands include; Spitzbergen, Franz Joseph Land, Severneya Zemlya(first reached by icebreaker in 1913 is not shown at all), Ellesmere land, North Devon Island, Cockburn island, Cumberland Island, Prince Albert Land, Wollaston Land, Victoria Land and the DeLong Islands. This from “North Polar Regions, Chart of the Arctic Ocean. Compiled from the latest information, 1885. This chart was included in Greely’s Three Years of Arctic Service and Account of the Lady Franklin Bay Expedition of 1881-1884, Hydrographic ofice US Navy.
Also interesting on the map is a stretch of NE Greeland coast, located approx. 60 to 100 miles north of C. Bismark (King William Land) that refers to it having been sighted in 1670, but apparently not evident at the current time(1885).
Folks, I just posted a similar question in the Unforced Variations thread, but this seems specific to ice, so if you’ll pardon this… over what time span does ice area reflect climate changes and why? I guess I don’t understand why this plot has a slope that stands out over much smaller time periods than the global average temperature. Does weather have less effect on the ice area as compared to its effect on global average temperatures? Thanks.
#47, Will, that is brilliant!!! Yes and yes, the biggest indicator of no (bi)cycles rendering the arctic ocean open at times is in the archeological record of aboriginal ruins throughout the Arctic. None are near the same places not mapped in 1885. Also lets not forget Bowhead whales extending the record backwards a whole lot. So I read a guy from Alaska knows better, we await the tangible records.
derek, I’m not sure which specific plot you are referring to but I think I can answer your question anyways. The time scale needed to separate the signal from the noise is a dependent upon the dataset, mathematically speaking it’s a property of the dataset and not a property of the climate. Certain metrics, such as global mean temperature, will have a different timescale on which the signal overcomes the noise than other metrics, like Greenland ice mass. It just so happens that for global mean temperature you need about 15 years on either side of a point to identify the trend.
For a much better explanation check out Tamino’s blog post: Fifteen
Comment by Unsettled Scientist — 14 Aug 2012 @ 5:50 PM
derek: I’m going to give a different answer than unsettled.
There are two sorts of questions that could be answered. One is if you have only a single
time series how long must it be accumulate significant enough statistics to establish a trend. That was what Unsettled answered. A second question is: starting at some arbitrary sea ice state, you impose a now constant climate, you would expect the ice state to asymptotically approach an equilibrium state. What is the time scale of that adjustment? I remember that having been discussed at RC at couple of months back, and the claimed timescale was roughly three years. That is of course a different experiment from that of trying to establish a trend in a presumably changing climate. Maybe one of the contributors to that discussion can chime in?
All — There are excellent Arctic ice graphics, animations, and interactive viz tools on the iPad/iPhone App “Climate Mobile,” free at the App store. (Best used on an iPad.) Click on the Interactive Sea Ice Tracker for lots of options, including derived “thickness” from PIOMAS volume and NSIDC area. Very interesting pattern there. One thing that stands out is that PIOMAS volume trend is not consistent with other data sets. Just how thin can the ice get before it disappears?
“I’m not sure what the different colors indicate, but my best guess is that the darkest blue/green is open ocean, white is solid ice, and intermediate colors indicate some concentration of ice, but not solid.”
So, if people like AlaskaHound try to say that the ice concentration in 1884 was nearly the same as today, this would be rather silly, as most of the Arctic wasn’t even explored back then.
The light blue color in the map indicates the southern limit of drift ice. So nobody back then knew the real extent of the Arctic Sea Ice at a given time. We can only try to derive estimates by putting together various observations. But those estimates would have huge error bars and when people do create such estimates, it indicates a much more bigger Arctic Sea Ice Extent than today.
I recall reading about the NW passage, let alone the rest of the Arctic, a few years back. Yes, prior to 1906 the passage had been made a few times. The ships which attempted it were destroyed or icelocked and the journeys were completed by dog sled…..
I say 1906 because the 1903 Amundsen expedition took 3 years of jumping through leads in the summer breakup before it could get through.
The timescale of response of Arctic sea ice area/extent is a mere matter of years. This is because the winter growth of new ice is a very fast process. Over the winter the limiting factor for thickening of new ice is not really time, it’s the thermal conductivity of the ice. The thicker the ice gets the better it insulates atmospher from ocean. And thickening of first year ice across much of the Arctic is a matter of sea water freezing onto the base due to heat being conducted away from the ice ocean intefacte driven by the intense cold above the ice.
The timescale of response of older multiyear ice is a slower process. Multiyear ice is mainly thickened by compression from the sides causing ridging. There is a mecahnical system involved in this. The transpolar drift draws ice from Siberia (creating open water which freezes in the winter making new ice) and crushes it against the north of the Canadian Arctic Archipelago and Greenland. This crushed ice is aged both by remaining off Greenland and the CAA, and by being cycled in the Beaufort Gyre ‘flywheel’ after which it gets drawn back into the transpolar drift. So because of the years involved, multiyear ice grows far more slowly than first year.
In terms of a warming:
The response of the ice is rapid, but the dynamics of the ice pack play a moderating role. Since satellite observations started in 1979 the ice pack has transitioned from a mainly multi-year mass, to mainly first year. This process is slow because it takes time to remove the multi-year ice and thin it down. Open water formation efficiency plays a role in this. Say in a given area the ice loses 1m thickness between March maximum and September minimum – with a starting ice pack of 3m thick (i.e. in the 1980s) there is no way that open water will be exposed, back in winter and the 1m lost can regrow. But with a continual +ve forcing the ice thins until in the area under consideration it’s under 1m thick. If that ice still loses 1m thickness during the season it will be able to expose dark ocean, which will then be warmed by the sun (ice albedo feedback). So using only an areal metric (area/extent) over most of the time the forcing applies it might seem like little is changing, but once the ice gets thin enough changes in area are rapid.
It might seem that this guarantees a rapid transition to a seasonally sea ice free state. But the fast growth of first year ice over the winter acts to limit the area/extent loss – energy must still be expended in melting that ice, and the open water exposed by increased open water efficiency vents a lot of heat into the atmosphere after the sun sets in the Autumn. This doesn’t however guarantee there won’t be a rapid transition.
The current summer is giving a solid glimpse of what the future will look like as far as weather and climate.
Without sea ice the polar weather zone will become temperate, temperate weather zones will become tropical, without sea breezes for those inside the continents along with a jet stream at about 50 degrees latitude Northwards, the summers will be great for heat lovers in Canada, kind of horrendous in the US with droughts a plenty. For North Africa and Euro-Asia much of the same. It is simple, yet I have watched many TV meteorologists fumble the correct version, they tend to say the “jet stream is up North” without explaining why. The jet stream is roughly at the boundary of atmospheric zones, wipe out sea ice and the summers will be minus the coldest zone with the tropics extending North. The implications need be modeled further but we got a lesson from nature, as Ray wrote, oblivious to our needs.
That capitalised sub-headline you copy from the site you link to – that sub-headline is a bit misleading, me thinks.
It refers to the Cryosat-2 data described in a Guardian article (on Sunday, so actually printed in The Observer) which procaryotes also links to @51 and which describes measuring a 900 cu km annual ice loss between 2004 (measured by ICESat) and 2012. This level of annual loss is also what PIOMAS is suggesting See graph (Usually 2 clicks to ‘download your attachment’).
There was a bit of confusion caused by a BBC Radio interview with Prof Laxon on Monday linked by Arctico @18 which referred to the same data but suggested it came from 2011 not 2012. Neven managed to contact Dr Laxon and get to the bottom if it. The data (‘summer figures’ 7,000 cu km) was for October/November 2011.
What is important about this 7,000 figure is that it is the same ice measure that PIOMAS models & so as well as the average 900 drop per year, the 7,000 also allows a confirmation of the absolute PIOMAS values.
The confusion by the Guardian probably arose because there is a winter figure as well (February/March) and that is for early 2012. As the Feb/Mar ice extends beyond the Arctic Ocean, this figure will not equate to PIOMAS.
(And if simple arithmatic had been applied, 13,000 (2004) minus 7,000 (2011) yields the ~900 cu km annual drop when 2012 would not.)
Does that also mean that Britain will be looking at cooler summers from now on? Will South Africa start to see snow regularly now? Will droughts be a thing of the past in Australia? Be careful when substituting current weather conditions for climate. There have always been heat waves (at least since the current interglacial started), and global warming may make them worse, but there is no indication that they will become more plentiful.
[Response: Not so. The Hansen paper indicates that heat waves (defined by a threshold with respect to a fixed climatology) do become more frequent, and they become worse too. Your point about focusing too much on current weather is however sound. - gavin]
Thanks all for explaining to me the time scale issues with the ice pack’s response to climate change. Is it fair to say that this response is consistent with global warming, but not sufficient to prove it, because there have been other natural variations in the ice extent in the past? I guess I am thinking that when the Vikings were in Greenland the ice pack was probably smaller.
Hi prokaryotes @63
I know no more than the PIOMAS figures & the data gleaned from ICESat & Cryosat-2. It is dropping at some 900 cu Km or so pa. From POIOMAS, the annual minimums (daily data) from 2005 on are 9200, 9000, 6800, 7000. 6400. 4400, 4000 suggesting an average annual drop of 870 cu km over the period. Unlike Area & Extent, volume is dropping similar amounts winter and summer.
PIOMAS being a model, it must give a lot of soul-searching for those involved so it is encouraging that their hard work is vindicated by the Cryosat-2 data. Well done, PIOMAS!!
Now when you say “conservative” estimates, PIOMAS say their output (PIOMAS v2 IC-SST) is “a conservative estimate of the actual trend” but this is a comparison with PIOMAS v1 which gave the trend since 1979 as 25% higher.
Derek, the rapidity with which the ice is currently disappearing is caused by global warming which has already been sufficiently proven previously. Why would you assume that there was less ice for the Vikings, or that it would have relevancy to today’s unprecedented melting. Is there some specific evidence about the time of the Vikings on Greenland? What about the direct ancestors of the Innu who came later? What about the North Americans that preceded the Vikings on Greenland? Is that remark simply based on an assumption that if people were there it must have been warmer with less ice?
Comment by Unsettled Scientist — 15 Aug 2012 @ 5:28 PM
Thank you MARodger for helping to improve my article, i’ve updated my post now accordingly.
The ice pack was smaller than what? Than it is now? Very unlikely.
The Vikings sailed as far as they could and got as far as Newfoundland, but no further. Not through the North West Passage, which is open now. It’s fair to speculate about provisions and why they didn’t go further, but an ice sheet is the most likely obstacle.
It’s important to remember that Greenland was a very marginal proposition for existence when the Vikings found it. A small change in climate wiped them out. The temperatures at present are higher and far more accomodating for agricultural purposes in Greenland than at any time in human history.
I’d very much like to hear opinions on the believability of the PIOMAS volume trend, something that was discussed in a previous thread. Projections of that trend are far more dire than of any area/extent trends and I have to question their accuracy. A naive projection of PIOMAS has Sep ice gone in 6 yrs. It was argued here earlier that that is not plausible on statistical grounds — natural variations will bump it back up — yet the trend continues. It doesn’t appear to be a succession of chance events. The 33-year pattern is very systematic. Others suggested that maybe “PIOMAS is crap.” The vast weight of expert opinion by modelers holds that we will not see an ice-free Arctic for 20-50 years. So what’s the deal here? My guess is that there will still be Sep ice in 20 years. At some point PIOMAS will have to bottom out. No?
Pretty much every night for a decade, the arctic conditions gets a quick check. The Jaxa site referenced in the article is the baseline, but since AMSR-E went offline last October it has a problem of possible recalibration issues. The comparative site is:
It uses a 30% concentration threshold, so there’s an inference on both extent and pattern; draw it with a comparison to the Jarxa site. This became pretty valuable when the 2007 shrivel was diss’d off: a freak weather anomaly that caused a massive arctic exit through a clear southern channel.
In the Daily Kos article, there’s a picture of a decline graph that uses the same 15% concentration threshold as Jarxa – and the interactive ability is worth the price of admission:
“At some point PIOMAS will have to bottom out. No?”
Unfortunately, that point is zero.
IMO, the statement that PIOMAS is inconsistent with other measures is not true. Because the relationship of ice thickness to extent or area is not fixed a priori in any way that we know, it’s quite conceivable that the thickness (and thus, volume) can decrease much faster than the other parameters mentioned. (Perhaps there are even physical reasons to expect thickness to decrease most rapidly.)
AFAIK, PIOMAS is consistent with in situ measurements of ice thickness–indeed, they use such measurements to constrain the model. So, no reason to think PIOMAS is terribly far off the reality.
To illustrate my point above, here are some 33-year [http://climatevo.net/IcePlots.png Ice Plots] for Area, Extent (both interpolated NSIDC monthly means), Volume (PIOMAS), and “Thickness” derived from PIOMAS volume divided by NSIDC area. If all of these are correct we would have to conclude that thickness has not yet fallen to the critical level that would result in a wholesale collapse of area/extent in a short time; and that if the Sep volume trend were to continue at the present rate, within a few years we would see such a collapse in order for all of them to reach zero at the same time.
Possibly. Alternatively, if the ice were to persist for another 20+ years the volume trend will have to change. Either it’s wrong to begin with or whatever is driving the current decline will have to lose its force. I tend to suspect that if the volume data were truly accurate the area would be declining more rapidly already — though maybe this year heralds a new trend. I guess we’ll find out soon enough.
This paper includes intercomparison with available measured thickness. The recent Cryosat results indicate that the recent strong decline in volume shown by PIOMAS is real, and that the good agreement with observation in the past is maintained, i.e. the recent crash in volume is not an artefact of the model.
It is reasonable to expect that the current acceleration of trend may not continue. I think that being able to add this season to 2007 and 2011 will help answer that question. 2007 and 2011 were joint record lows, and both showed a reduction in loss during and after the last week in August. If this year follows suit then it will be reasonable to expect that volume loss may pause while the ice loss advances in the season. I think the reduction of loss in 2007 and 2011 in late season is probably because the melt edge had extended well into the 80degN area and reduced insolation by late August meant the area began net heat loss rather than gain. So the melt stalled.
Furthermore the massive loss of volume between 2009 and 2010 was due to the virtually complete annihilation of grid boxes reporting thick ice. This process cannot repeat, and the behaviour of the ice pack since then may be epiphenomenal of thin ice, rather than part of a process of rapid transition. Then again it may lead to a rapid transition. At present I’m not convinced anyone knows either way.
If it was just PIOMAS that was predicting an early ice-free summer, then it would be hard to justify it. But you hit on something broader than just PIOMAS when you say “The vast weight of expert opinion by modelers holds that we will not see an ice-free Arctic for 20-50 years.”
The climate models that suggest 20-50 years (or more correctly 30-60 years) also seriously underestimate present Sea Ice Extent & Area which is something we can measure with confidence. The question is then, why are the models not modelling the ice? For some time, the answer given was “natural variation” but the variance has continued to grow for many years and is now far too big to be explained away as “natural variation.” Even the latest AR5 models are not capturing the decline in the ice! (I have seen a graph of this but cannot place it to link to. Here is an AR4 version.)
So why are the models off target? I think it’s because general climate models of increasing sophistication take longer and longer to build and to run. Ice data relies on the short satellite record. And it is only 10 years ago that predictions of 2030 first started appearing & 5 years ago for sub-2020 predictions.
Conversely, the thickness data PIOMAS produces has been checked against the thickness measurements that are available and it has not been found wanting. The latest Cryosat-2 data will allow yet more checking. Prelimenary data suggests again PIOMAS is an accurate reflection of the ice. So PIOMAS is being questioned but the answers are not good for Arctic summer ice.
The size of the ice pack during the reign of the vikings is highly speculative, and neither your claims nor Derek’s can be substantiated. Given the pospect of more fertile land to the South and West, why would the Vikings even venture through the Northwest passage? Your last statement about Greenland temperatures is contradicted by ice core data.
Here are some observations by Jim Pettit at Neven’s blog on the event:
* That’s only the third time in the record area has been below 3 million km2 (2007 and 2011 were the others, of course).
* It only took 138 days this year for area to fall below 3 million from this year’s maximum. 2011 accomplished that feat in 169 days–one month longer–while 2007 needed 178 days, nearly six weeks more. Perhaps more astonishing, that 138 days is less time than it took most years that fell below 5 million km2 to reach that point from their individual maximums.
* 2012 SIA fell below 3 million km2 nine days ahead of 2007, and ten days ahead of 2011. 2012 area is currently 405k km2 ahead of 2011 on the same day, and 356k km2 ahead of 2007.
* As others have noted, an area record is just 81,585 km2 away. Even if area were to see an additional decrease this year equivalent to only the lowest post-6191 drop on record–1997′s 194k–2012 would still end up with just 2.793 million km2. At the other end of the scale, 1984′s post-6191 drop of 1.113 million km2 would render a very far-fetched 1.874 million km2.
* 2012 SIA has now been in first place for the past 47 consecutive days–since June 30–and 63 out of the last 68.
* 2012′s SIA anomaly has been below -2 million km2 for 14 consecutive days. That’s 20% of the 70 days on record with an anomaly greater than -2 million. (The greatest anomaly so far this year ranked #20 on that list of 70; the 19 largest negative anomalies were all in October of 2007.)
I make that about 2m sqkm of pack ice and lots of floating chunks to keep your drinks cool. Even Trolls can’t argue with 0. Although they can start talking about November to March and how it’s “recovering”….
That’s what Internet trolls do. They tell lies for effect.
‘Greenland Melting Breaks Record Four Weeks Before Season’s End’
Similar articles on Greenland’s 2012 melt have been circulating these past few days. In particular, what struck me as dramatic is this part:
This year, Greenland experienced extreme melting in nearly every region — the west, northwest and northeast of the continent — but especially at high elevations. In most years, the ice and snow at high elevations in southern Greenland melt for a few days at most. This year it has already gone on for two months.
If that last sentence is the beginning of a new level of melt that continues into subsequent years, then the Greenland melt has entered a new, much greater era of Summer melt, but along with the dramatic melt of arctic ice in 2007 and subsequent much lower yearly ice extents and volume, suggests the situation is accelerating. The question is; Is the acceleration exponential? And if not, what is the best way to describe the acceleration curve?
Reply to #69: I’m not sure how there can be a strong argument against piomass, because regardless of the accuracy of the volume measurements, there is at minimum a comparison of different years and the trend is clear in the direction of approaching zero in some future melt season.
Dan H, what does the Younger Dryas have to do with the Vikings? I think you are confused about history. The Vikings were in Greenland in the 10th Century AD and the Younger Dryas was about 10,000 years before that. Looking at the PDF you linked which does include a temperature graph for the period of the Vikings it shows a strong decline in temperature around the time of the Vikings settling Greenland down to -32.5 C (Figure 1) which is lower than today’s temps. Yes the paper shows there have been warmer temps in the past (at the Summit region of Greenland), but not at the time of the Vikings. This paper actually supports the idea that it was very cold indeed for the Vikings in Greenland.
As that Kobashi et al paper you linked notes it’s conclusions are different than others for a couple reason, one main reason being they aren’t using summer temp but an a mean-annual temp. When placing the current temps in context with the past 4000 years that paper actually uses decadal-mean temp, so an average of 2001-2010, over which we’ve seen accelerating increases of ice mass loss. They even point to Kaufmann et. al as an indication that for summer temps, this is the warmest it has been since humans have been to Greenland.
So depending on how you measure it, mean-annual temp or summer temp when the melting occurs, this is the warmest it has been since humans have set foot on Greenland. Regardless of how you measure it, it was colder for the Vikings.
Lastly, one quote I’d like to pull from the conlusions of the Kobashi paper you linked. When discussing why their conclusions differ somewhat from other recent reconstructions they write “our site is not necessarily representative of the the Arctic as a whole, and may respond in opposite ways to annular mode fluctuations.”
Comment by Unsettled Scientist — 16 Aug 2012 @ 12:41 PM
” why are the models not modelling the ice? For some time, the answer given was “natural variation” but the variance has continued to grow for many years and is now far too big to be explained away as “natural variation.” Even the latest AR5 models are not capturing the decline in the ice! ”
I wonder at the seeming mule-headedness of ice scientists. I asked at the beginning of this thread if a SINGLE run of a SINGLE model has produced results approximating reality. Nary a peep in response. Truly amazing. “There is SUPPOSED to be more ice!!!!” while stamping their feet….
re: Vikings, Greenland, and sea ice.
My recollection (which may be faulty) is that even during this “warm period”, south Greenland was only seasonally accessible by boat, and in later years visits were not possible every year. That is to say, unlike today, no part of Greenland was ice-free year round. By contrast, today Sisimiut is Greenland’s northernmost year-round port.
There’s an interesting problem with the idea of heatwaves becoming more frequent: there’s a fixed number of days in a year. If every day were above average for the year, there would be 1 single heat wave that year.
The problem is, of course, that “heat wave” is a vague term, and as the climate warms, the significance of individual weather events might blur. Like this year: was July several heat waves or one monster one?
Thanks again for the comments everyone. I wasn’t trying to suggest that climate change isn’t causing the drop in ice coverage; mainly I was just trying to get at whether there would be longer term (say 100-year) variations in ice coverage. I am curious whether, if you ignore everything else, one could use the Arctic Sea Ice Volume Anomaly plot to show that climate change is occurring, or if that is consistent with, but insufficient to prove, a general increase in the heat budget of the Earth.
I picked the Viking example out of the air, and phrased it poorly. I meant to suggest that maybe there was less ice in the 1200s relative to surrounding centuries.
Dan H. #77 “Does that also mean that Britain will be looking at cooler summers from now on?”
You confuse cool summers with a wet climate because the temperate to tropic boundary is just at or South of Britain. That statement is typical of name games trying to be clever, however intelligence is elsewhere, I foresaw it in my long range projections. The wetter weather in the UK was due to a great deal of heat and moisture combined with the presence of cooler upper air.
“Will South Africa start to see snow regularly now?”
It is winter there , so what is your point? The long Antarctic night will be cold for a long time to come.
” Will droughts be a thing of the past in Australia?”
Don’t bet on it you might be poorer from doing so. ENSO is a strange cycle, I understand it better, but much more is needed to figure it out.
“Be careful when substituting current weather conditions for climate.”
The new world climate order will be an extended tropical zone well into the Northwards for the NH. The weather it gives depends on location. Be careful when making not so bright statements about weather, makes you sound like a TV meteorologist.
Once again, your links don’t really support what you claim they do. The first link is only concerned with decadal temperature averages, which is not going to tell you much about year to year ice variations, particularly in comparison to the most recent years. Your second link is to a paper discussing temperatures in “central greenland.”
It appaears to me that you are scrambling to generate evidence that does not exist.
There is evidence that Viking technology spread into the plains north of the treeline and west of Hudson Bay in central Canada in pre-Columbian times, but that it came overland from the east, and not down from the north. The better question is why WOULDN’T the Vikings have taken avantage of a clear northwest passage, if it had existed.
The Younger Dryas has nothing to do with the Vikings. It was a period shortly after the planet started emerging from the last glacial maximum when temperatures plunged suddenly. It was 11,000 years before the Vikings ever built their first ship. Also, I think you may be confusing the Little Ice Age with the Viking colonization of Greenland, when temperatures plunged to -32.5C. That period was several centuries later, from about the 16th to 19th centuries. The Viking era was associated with the Medieval Warm Period, which was centered around the year 1000.
While summers may be warmer (this paper does mention the 2000-yr cooling trend, but shows higher temperatures prior to its onset), the paper clearly states that the current surface temperature (-29.9C) was similar (-29.7C) in the Medieval Warm Period. Also, 72 decades were warmer than the present, some by 1-1.5C, during which time Greenland was inhabited by several other groups of humans, before the Vikings arrived. The present decade is not outside the envelope of variability. So unless you which to redefine “human,” your statement still contradicts the data.
Getting back to Derek’s original question, it cannot be answered, as historical extent of the floating sea ice cannot be determined by proxies. Some studies present the sea ice extent off Greenland and Iceland during the past millenia, or so.
We can get an indication of what the sea ice extent may have been like, by using temperature proxies for nearby sites. The local sea ice, which is indicative of the maximum sea ice extent in the North Atlantic, correlates with temperatures, which would indicate that the greater Arctic sea ice would likely follow. Since the mean decadal Greenland temperature has varied by 4C over the past four millennia, it is reasonable to conclude that large variations in sea ice also occurred. During the past two centuries, temperatures have risen ~2.5C, and there has been a significant decrease in Arctic sea ice.
Did you read Gavin’s response about weather vs. climate? You claim in your post superior intelligence, but your response suggests a lack of comprehension. My previous post, which was bathed in sarcasm, cautioned against using recent weather patterns to claim climate shifts. Repeated weather patterns would be a better indication. By the way, snow is a rarity in South Africa. It has not snowed in Pretoria in decades, and this was the first recorded snowfall in all nine South African districts simultaneously.
Not sure what Antarctica night has to do with South Africa. Being just south of the Tropic of Capricorn, the minimum winter sunlight is still ten hours.
Looking at the arctic seaice graphs for area, extent and volume using decadel averages since 1980-89 etc till today. https://sites.google.com/site/pettitclimategraphs/home All three demonstrate that the ice loss is accelerating rapidly every single year. However the sea ice volume graphs sends shivers up my spine. Here you can extrapolte that the summer sea-ice will be long gone within 10 years. The decadel volume graph since 1980 is practically exponential. Send these graphs to your policy makers urgently and there is absolutely no mistaking their significance!.
Comment by Lawrence Coleman — 17 Aug 2012 @ 8:38 AM
Although most of the record minimums of the Arctic Cryosphere are looking shaky as the melt season forges ahead, as of today none have fallen yet – except that is for Land Snow Cover. (There is an issue with pre-satellite Greenland measures of snow cover but that can be dodged by excluding Greenland.) I reckon it stands as the first scalp of the 2012 melt season, the 24 year record tumbling during the second half of July.
Minimum Land Snow Cover
(Data – Rutgers Uni Snow Lab weekly)
Pos.. Year….. Min (Sq Km)
1….. 2012….. 159,325 to date.
2….. 1988….. 202,563
3….. 2011….. 204,226
4….. 1990….. 204,658
5….. 2009….. 281,217
6….. 1992….. 282,868
7….. 2008….. 286,271
8….. 1968….. 286,507
9….. 2010….. 286,788
10….. 2004….. 288,068
You do realize that we have limited temperature data for Greenland, and what we do have does not offer much in the way of specific annual temperatures. The paper does compare the data to recent thermometer measurements with reasonable agreement. Yes, we were discussing temperatures in Greenland during the Viking settlements. There have also been Viking artifacts found in New England and the upper Great Lakes region, but there is doubt as to whether the Vikings actually ventured that far, or traded with others who carried them inward. We can only speculate as to the route travelled, or their extent in North America.
The evidence is sparce, but if you have anything to add, please do so to further the discussion. Otherwise, we will have to go on what evidence we have, which does exist, as per the papers.
Dan H. It is always easier to destroy than to build, and you have been busy here and no doubt elsewhere, trying to destroy the painstaking work of scientists trying to build a better understanding, and bystanders like me who at least appreciate what they are attempting to do in the face of a lot of dirty fighting.
Your reckless disregard for the truth is exceptional.
I am willing to bet you long odds that in 20 years (I’d say 10 or even 5, but IMNHSO the kryptonite shell of ignorant resistance you are wearing will require your little island of sand to be completely surrounded before you acknowledge you are wrong) you will be completely bereft of reality-denying arguments, say $1000 at odds 5:1.
Comment by Susan Anderson — 17 Aug 2012 @ 11:57 AM
Dan H, you’re all over the place and didn’t even read your own source. I should have known better than to address you directly. I’ll do it once more just to point out how you’re spreading confusion.
> I think you may be confusing the Little Ice Age with the Viking colonization of Greenland, when temperatures plunged to -32.5C. That period was several centuries later, from about the 16th to 19th centuries.
No, I got the temperature from the paper you linked. Please open the PDF, scroll down a few page to Figure 1 like I said in my first reply, and look at the year 1000 in the middle graph. Once again, you fail to understand your own source. It helps the conversation if you read what I wrote about your source, then go read the paper to see that what I’m talking about. It is before the MWP, as were the Vikings. But I should know better than to expect you to read your own sources and represent them faithfully. I’ll definitely not be going down this particular rabbit hole any further.
PS – as Craig Nazor notes using decadal averages to discuss interannual variability is pointless. If I go from 100 to 0 in 10 years, you can point at a decade where it was 50 all ten years and they have the same decadal-mean. Not useful.
Comment by Unsettled Scientist — 17 Aug 2012 @ 12:03 PM
Given typical regional temperatures in SA during the winter but adding atypical atmospheric moisture for that season, why should snow be terribly surprising, or notable except as confirmation of what reasonable to expect?
Come to think of it , the fact that the North Pole is ongoing a dramatic seasonal ice-scape change while Antarctica remains essentially intact has very large consequences outside of the hemispheres isolated by the equatorial divide. In Particular ENSO, but not uniquely so. I Imagine that a warmer Northern hemisphere as opposed to a cooler Southern Hemisphere has equally consequences needing modelling. For instance Hadley cell winds should be diminished (a la Lindzen) exacerbating the formation of super El-Ninos.
Has any information surfaced whether or not the salinity of Arctic surface water increased following the storm of Aug6. If so, this would indicate that the storm mixed deeper, warmer, saltier Atlantic water into the surface and would help to explain the increase in the slope of the ice extent graph. This water is only one to 2.5 degrees above the melting point of ice but one meter depth of 1 degree water has enough heat to melt 12.5mm of ice. There is more than enough heat available to melt all the ice many times over. How effective was this storm in bringing up this deep water.
Wayne Davidson @108
You say “For instance Hadley cell winds should be diminished (a la Lindzen)…”
Is this the Lindzen assertion that because of polar amplification, the temperature difference between equator & pole will diminish, and thus extreme weather will diminish?
I have always found this somewhat nonsensical. It is akin to asserting that racing car drivers are safer driving faster because they complete the race more quickly and so have less opportunity to crash.
In a warmer world, the reason the poles are warmer compared to the equator is surely entirely because of all that extra (more extreme) weather pouring polewards.
Given the amount of cods wallop that Lindzen spouts, I consider that it is not good to use him to support a line of argument.
101 “we will have to go on what evidence we have, which does exist, as per the papers.”
Comment by John E. Pearson — 17 Aug 2012 @ 9:03 PM
> Given the amount of cods wallop that Lindzen spouts, I consider that it is not good to use him to support a line of argument.
I’ve got a couple of issues with Lindzen, not the least of which that he will draw one picture in front of a scientific audience, then turn around and draw a different picture in front of an audience he thinks is incapable of calling him on it. It’s a character flaw that makes it so it doesn’t matter how good his science is, I cannot trust what he is saying. If his science is sound, it will be recognized by the community and I can ignore his presentations and non-peer reviewed writing.
Comment by Unsettled Scientist — 17 Aug 2012 @ 9:48 PM
110 MARodger said, “I have always found this somewhat nonsensical. It is akin to asserting that racing car drivers are safer driving faster because they complete the race more quickly and so have less opportunity to crash.”
I think the analogy describes the effect well. Less temperature difference makes for a slower and so more wobbly jet stream. Another analogy would be riding a bike in a straight line. Go too slow and you’ll fall.
Not sure how to respond to this topic. Does this mean Santa Claus won’t be bringing Xmas gifts much longer, or they’ll be all wet and soggy?
RE “It’s worse than we thought.” I just posted to a blogger who raised the typical red herring re “and how many of the climate scientists’ predictions have panned out?” To which I replied, not many actually, because what they’ve been finding instead over and over again in the 22 years I’ve been following the issue of CC is “it’s worse than we thought,” “It’s much much worse than we thought.”
Comment by Lynn Vincentnathan — 18 Aug 2012 @ 12:25 PM
“In a warmer world, the reason the poles are warmer compared to the equator is surely entirely because of all that extra (more extreme) weather pouring polewards.”
Not quite, Arctic permafrost ( a major player in Arctic climate) is weakening and retreating to the North, http://www.reuters.com/article/2010/02/17/us-climate-permafrost-idUSTRE61G5BG20100217, the Arctic ocean sea ice is lesser expansive causing great Arctic Ocean warming. Even during recent Maximum extent, the lesser sea ice thickness has a significant impact (allowing more cyclones to penetrate the North Polar atmosphere). In effect the Arctic is becoming warmer, the Arctic Ocean is becoming a more important heat source over the long night. This summer showed what happens when there is far lesser ice with a warmer ocean.
Although I am a great fan of criticizing Lindzen especially when he acts like a standup spokesperson for contrarian interests, he didn’t get that chair at MIT by being controversial, that came after.
Imagine a world with the oceans having lesser sea surface temperature differences, the doldrums come to mind.
Looks like the arctic ice extent melt is slowing down. The link above (from today – updated at exactly 5:00pm PST every day) shows it to be about 4750 million sq. kilometers, but yesterday was about 4800. During the recent storm the melt was around 200 thousand kilometers a day, then went back to about 100k a day (which is what it was previously averaging), but now seems to have slowed considerably. Even at 50 k drop a day the record will probably still be broken.
Just as a follow up to my post yesterday (#118 with link to NOAA) the slower melt has continued into a 2nd day with approx. a 50k melt, down from yesterday’s 4750 million sq. kilometers to approx. 4700. So two days in a row of 50k melt (the new lower melt rate).
From where we are now: 4700, and with an average melt per day of 50k, a new record would be set at 4300 million sq. K in 8 days, on Monday the 27th of August. Since the melt towards the end of the season slows, let’s say from Aug. 27th to Sept. 15th the melt averages 35k a day, would be 19 days x 35 = 665k, bringing the new record down to 3635.
That was just an example, because the melt can occur on many different dates in Sept. and the average daily melt may be > or < than 35k.
However, it's starting to look like we might get a final melt count down in the 3's, i.e. less than 4 million sq. kilometers. That in of itself would herald in a new threshold the MSM might sit up and take notice of, even if it was for one news cycle.
“Getting back to Derek’s original question, it cannot be answered, as historical extent of the floating sea ice cannot be determined by proxies.”
Science disagrees. “These results are consistent with the northern Ellesmere Island record indicating an abrupt termination in driftwood stranding along most of the north coast (>250 km) after 5.5 ka due to the development of exceptionally thick, multi-year landfast ice (ice shelves) with possibly only short intervals of lesser ice since then (England et al., 2008). The growth of these ice shelves indicates very high ice concentrations in the adjacent Arctic Ocean since ca 5.5 ka, which apparently expanded and ice-locked Northeast Greenland coast by ca 3 ka.” And which will cease within a few decades.
“On suborbital time scales, ice distributions varied in the Holocene, but no evidence exists for large, pan-Arctic ﬂuctuations. Historical records indicate that Arctic sea-ice extent has been declining since the late 19th century. Although this decline was accompanied by multidecadal oscillations, the accelerated ice loss during the last several decades lead to conditions not documented in at least the last few thousand years. Taking together the magnitude, wide geographic distribution, and abruptness of this ice loss, it appears to be anomalous in comparison with climatic and hydrographic variability observed on submillennial time scales and longer-term insolation changes.”
History of sea ice in the Arctic, Polyak et al. Quaternary Science Reviews 29 (2010) 1757–1778 http://www.cgd.ucar.edu/staff/mholland/papers/Polyak_2010_historyofseaiceArctic.pdf
Is there a nice brief comparison of the methodologies used by NSIDC to determine sea ice extent vs. the sources used by Cryosphere Today? With over a million square kilometer difference in their estimates and Cryosphere Today, today showing a new minimum it would be nice to understand how each is derived.
Comment by Will MacKinnon — 20 Aug 2012 @ 11:47 AM
“I’ll bet on Sep 1st for the 2007 record low to be broken.”
@Will MacKinnon @125
The two measures of Arctic Sea Ice you ask about are very different, one called Extent the other Area. (NSIDC measure both but usually only mention Extent.) Area is as you would expect, the area of ocean covered with ice. Extent is the area of ocean that is covered with more than 15% ice (DMI use 30%.) and was adopted to make the measurements of melting ice flows easier.
A graph of NSIDC Area & Extent is linked here (usually 2 clicks to ‘download your attachment’).
A suggestion; would it be possible to ‘slow it down,’ such that a person could see it unfold more slowly, and to be better able to tell the beginning date? I had to watch it a few times to figure out the start date.
Thanks Harry, I really intended it to be looped with VLC or mediaplayer, but youtube and other online video methods (excluding animated gifs) don’t really take looping into consideration.
With my initial version, anyone could download and adapt the timings to fit any dialog as appropriate, though I can redo easily using existing frames. I’ll add a pause at the beginning to allow viewer to digest the scene, then animate more slowly by duplicating frames.
It should be more publishable with Aug and Sept results when they are in, and easy to do – just another linux command with the new PIOMAS source data and an 8 hour wait for the results at 1080p.
Comment by Andy Lee Robinson — 21 Aug 2012 @ 12:27 AM
Look at sst’s all over the Arctic, Stretches the imagination:
#91, you asked why I am curious what we can deduce from the loss of Arctic ice? I am not trying to debate global warming as a reality, but am just trying to understand what can be inferred form sea ice, and get a feeling for the magnitude of the current drop compared to variations over a longer time span.
Thanks for all those people that provide links on proxy reconstruction of Arctic Ice. I found these papers very interesting. Cheers.
If you put a ruler across from the bottom of the dotted 2007 line, it appears the ice melt in 07 reached 4200 million sq. kilometers. However, 2007′s melt record was actually not that low, it was 4330. Now look at the left hand side legend where it has the years 2012 over 2007. 2012′s legend position is arbitrary, however 2007′s dotted line is correctly located at 4330 (evidently purposefully done to mark that record melt point while also using it as a legend). Now put your ruler across from the legend and you will see the dotted line on the graph drops below that point to about 4200. That’s a false deviation of 130 thousand sq. kilometers (or about 2.3 days at current melt rate of 50k a day), making it look like the 2012 melt has further to go to reach 07′s melt than is actuzlly the case.
2012 is at about 4550 and 2007 was 4330. In approx. 4 days the 07 record will be matched.
The DMI 30% extent index also reached a new low point today (dated Aug 20). The Uni Bremen 15% graph too is in record territory, although the UB team views their present data as provisional. Over at Neven’s folks are noting the records falling like dominoes one after another; it’s quite possible they all will in the next week or two.
This is off the cuff without checking the numbers myself but are you sure you aren’t being confused by the distinction between the lowest 5 day average extent NSIDC recorded in 2007 and the monthly average extent for September 2007, which is officially recorded as 4.30M at ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/Sep/N_09_area.txt. Naturally, the lowest daily value in the month is lower than the average for the month so I can’t see how the lowest daily value could have been 4.33M and the monthly average 4.30M.
Perk Earl, I think you are confusing the September 2007 monthly minimum extent with the daily record minimum, which was obviously lower; looking at the archives, NSIDC at the time put the 5 day record minimum at 4.13 million square kilometers and I doubt it was revised after the fact by that much:
Tenney Naumer @138
I’d reckon a link to the actual story might be more appropriate than that link to the “accepted wisdom” graphic (that graphic being the only bit you can’t blame me for). :-)). The Untold Drama of Northern Snow Cover now showing at Neven’s.
For anyone interested in the opening of Northwest passage I’d recommend the detailed maps from Canadian Ice Service (www.ec.gc.ca), en français aussi bien sûr (mais pourquoi pas également en inuktitut, c’est pas juste :-( ). They present composite images from Modis and Radarsat satellites. According to the images I browsed through, the opening seems imminent if not already done, though ground confirmations are necessary.
From the ground, I’d recommend northwestpassage2012.blogspot.com where several ships are attempting the passage. Quite interesting!
Looking at the JAXA graph, it appears that it usually takes about 40 days for the derivative of the sea ice extent to go from roughly constant and negative in the June-July decline to positive sometime in September. This year, the period of (roughly) constant negative derivative seems to have extended to the present. It will be intersting to see if that means that the sea ice extent minimum will come in October this year rather than September.
The emergent seasonal structure in the sea ice volume anomaly plot may get more pronounced.
Hi, out of interest, what’s the criteria for ice free arctic ocean, what if it mostly disappears (say within 5 – 10 years), for at least one day in summer, but small frozen areas remain around the edges, perhaps in sheltered spots ?
Richard D @ #146–over at Neven’s Arctic Ice blog, they seem to have settled on 1 million k^2 area as their definition of ‘virtually ice free’ Arctic Ocean, and we are getting perilously close to that this year.
140 Hank “Is there a shift toward a later date and a longer period of open water, along with the increase in maximum area melted, over the past decade or two?”
Between the Arctic Islands yess, not quite for the Arctic Ocean the refreeze is relatively fast but this has slowed down due to warmer sst’s, its the regular yearly latest contrarian trap, they fool themselves silly that the ice has recovered, but its an illusion they love to fantasize about. The biggest question is what are the impacts of thinner sea ice during the long night? The largest effect so far is the lessening of winter atmosphere build up, and more frequent penetration of cyclones from the South, a feedback loop causing the ice to be even thinner.
Richard D @146
Ice-free is usually defined as Sea Ice Extent dropping below 1 million square kilometres.
At the start of the satellite era, the summer Extent was dropping to about 7 million. The record from 2007 is 4.16 million & yesterday’s figure was 4.19 (down from 4.29 the day before). So, although this will be the first year that makes it over half way to an ice-free level, the other half will go with a rush as year-on-year the ice is a whole lot thinner. For instance, you won’t now get very long odds on an ice-free 2013.
#151 MARodger, Ice-free is sort of problematic in the near term because of daily tides which continuously ridge first year ice next to Greenland, the last standing obstacle slowing AGW appreciably. In the long term ice free Arctic ocean every year is possible as sst’s become much warmer. In the mean time, the next real attention grabber is an ice free North Pole accessible from the Thames in a few days or so of sailing, which can happen at any upcoming septembers, especially if La-Nina rages strongly in April to July, making the Arctic atmosphere more cloudless. When it comes to weather, the world is one, every region plays a role with one region or another. This year near or present El-Nino saved the NP yet to be seen blue.
We have now set two dubious records, a new low for sea ice ‘area’ and also a new record low for ‘extent’. As we are probably going to sea a ice free summer arctic within the next 5-10 year, does anyone know how soon the longer and longer periods of ice free conditions (or another 5mil sq kms of open dark ocean) will really begin to principally affect the northern hemisphere climate? A assume that to a degree the climate is already changing as a percentage of ice albedo forcing. Say that the arctic is totally ice free for about 3-4 months of the year,(very possible in the not too distant future) what changes in ther NH climate are we likely to see and how soon?. I think these are vital and pertinant questions that need answering.
Comment by Lawrence Coleman — 25 Aug 2012 @ 1:53 AM
Wayne Davidson partially answered my question, “more frequent penetration of cyclones from the south” smashing the arctic even more and accelerating the +ve feedback loop.
Comment by Lawrence Coleman — 25 Aug 2012 @ 2:04 AM
NSIDC now shows 2012 as lowest extent in their records . . .
#156 Ozajh yes time to reminisce enjoy found memories of great predictions, because we have endured them time and time again, got a solid memory of all time greatest moments in contrarian science achieving epic failures worthy of Greek mythology, something our descendants wont believe, they are our modern day legends when some people believed in people failing again and again in being correct:
wattsupwiththat.com/…/arctic-sea-ice-about-to-hit-normal-what-will-t…31 Mar 2010 – 2010 is looking promising for sea ice recovery again. After all, who wouldn’t want the Arctic Sea ice to recover? WUWT is predicting a recovery …
The Economist Provides Readers With Erroneous Information About …
wattsupwiththat.com/…/the-economist-provides-readers-with-erroneo…16 Jun 2012 – By WUWT regular “Just The Facts” A June 16th article in the … In fact, the largest influences on Arctic Sea Ice appear to be wind and Atmospheric Oscillations, i.e.: … for declines in how much sea ice covers the Arctic Ocean, with near ….. The Arctic ice has recovered a little since 2007 which correlates with …
Sea Ice News: Arctic sea ice “may” have turned the corner | Watts Up …
wattsupwiththat.com/…/sea-ice-news-arctic-sea-ice-may-have-turned-…13 Sep 2011 – Alerting message from the Arctic: The extent the the Arctic sea ice has … Stay updated with all of the latest plots and maps at the WUWT Sea Ice ….. “too low” in Sept, just wait a couple months — it’ll make a dramatic recovery.
Ignoramus web site looks better than ever now. The oscillations of the Arctic and cloud cover were not so favorable for a great 2012 melt. There was mostly lows over the Arctic ocean.
Not to forget Joe Bastardi “Trace gas have no effect” guy
Actually there are myriads of contrarian websites really missing this great melt:
DenialDepot: Arctic Ice Continues it’s Recovery
denialdepot.blogspot.com/2010/…/arctic-ice-continues-its-recovery.h…10 Jul 2010 – Arctic sea ice point blank refuses to melt. Continues to Defy alarmists. Maunder minimum ice recovery confirmed. Ice age imminent? I won’t …
The Climate Scum: Arctic Sea Ice Recovery
theclimatescum.blogspot.com/2011/08/arctic-sea-ice-recovery.html7 Aug 2011 – The Arctic sea ice extent is now above what it was in the corresponding period in 2007 according to NSIDC. This means 4 years of a steady …
“For example, he presents this slide to argue that the Arctic sea ice “death spiral” is nothing to worry about”
We can go on and on, I think this site should show more examples of completely erroneous long range sea ice forecasts because contrarians belief of AGW as a Myth. We must expose those who believe in the wrong things so that their reputations gets enhanced correctly.
Part of what Nevin wrote in his blog before leaving on vacation: “I really don’t know what to expect next, but what I do know is that all the records on daily graphs have been broken in the past week. Regardless of how this melting season ends, we know the ice is getting thinner and thinner, and so it’s time for the whole world to look ahead at what an ice-free Arctic could imply. Not for the polar bears or the ice itself, but for modern civilization, for us, humans.”
It probably means a greed-grab for whatever riches can be excised from the Arctic; precious jewels & metals, oil, gas, and who knows what else. Most likely a lot of militaristic posturing to secure territoriality. A lot of environmental analysis that gets ignored in favor of mo money from extraction, all the while failing to observe the most basic fact, that if we had not spewed so much GHG’s we wouldn’t be chasing our tails in a negative feedback loop that will end up with us with our tail between our legs begging for runaway GW to stop even if it means going with the most absurd geo-engineering idea! “Ah, no need to worry folks. The fix is on the way, and soon we will be back to scouring the Arctic for goodies!”
“Getting back to Derek’s original question, it cannot be answered, as historical extent of the floating sea ice cannot be determined by proxies.” You wish.
Science disagrees. “These results are consistent with the northern Ellesmere Island record indicating an abrupt termination in driftwood stranding along most of the north coast (>250 km) after 5.5 ka due to the development of exceptionally thick, multi-year landfast ice (ice shelves) with possibly only short intervals of lesser ice since then (England et al., 2008). The growth of these ice shelves indicates very high ice concentrations in the adjacent Arctic Ocean since ca 5.5 ka, which apparently expanded and ice-locked Northeast Greenland coast by ca 3 ka.” And which will cease within a few decades.
“On suborbital time scales, ice distributions varied in the Holocene, but no evidence exists for large, pan-Arctic ﬂuctuations. Historical records indicate that Arctic sea-ice extent has been declining since the late 19th century. Although this decline was accompanied by multidecadal oscillations, the accelerated ice loss during the last several decades lead to conditions not documented in at least the last few thousand years. Taking together the magnitude, wide geographic distribution, and abruptness of this ice loss, it appears to be anomalous in comparison with climatic and hydrographic variability observed on submillennial time scales and longer-term insolation changes.” History of sea ice in the Arctic, Polyak et al. Quaternary Science Reviews 29 (2010) 1757–1778 http://www.cgd.ucar.edu/staff/mholland/papers/Polyak_2010_historyofseaiceArctic.pdf (This also has some discussion of IP25 biomarker to which Kevin Mckinney referred.)
DanH and other skeptics often have problems distinguishing their beliefs, which are different from mine, and the facts, which are the same for everybody – even though no one knows all the facts. The resistance to learning unpleasant facts, a long recognized part of human nature, is why we have peer review, and multiple expert medical analyses.
Skeptics only accept peer review if it supports their beliefs; peer review that challenges their beliefs is see as conspiracy. Their most prominent “experts” include the likes of Monckton, Watts, Morano, and “iron sun” Plimer.
Why has the PIOMAS volume been reaching and bouncing back from its annual low point a month or more before the other measures? It’s monthly, so won’t update from its end-of-July chart (seen above at top of post) for a few more days. I’m going by the past few years’ behavior.
Estimated sea ice volume from UW PIOMAS (updated every month)
P.S. — responses 3, 4, and 5 to this topic are about the PIOMAS anomaly, with some pointers — I think there’s an explanation in there about the reason it shows an uptick while the other measures are still declining, but I haven’t figured out what it is well enough to describe it, nor found it explained clearly anywhere I can point to it.
In July particularly for 2010, 2011, and 2012, more of the ice melt has been happening faster and earlier — the anomaly is greater, and the graph dives to the low point on the PIOMAS chart for 2010, 2011, and 2012. In July it’s ‘more different’ from the past history — eyeballing, around minus 9, minus 9-1/2, minus 10-1/2 ‘thousand cubic kilometers’ as of the end of July).
By August and September, the months that always have the most ice gone and the most water exposed, the difference between the amount melted historically and the amount melted the last three years is not as much.
And that’s important because early melt increases how much open water gets sunlight for how many days — before night falls — so how much added heat gets into the Arctic ocean.
By September and October, water and ice both lose heat to the night sky very efficiently). Unless it’s cloudy, of course.
Trying for 7th grade language comprehension here. Corrections anticipated.
That is the ice volume. It is still falling, not bouncing back. The chart that is shown in the article is the ice volume anomaly, a daily departure from the (daily) average since 1979. One of the interesting things about the volume chart is that 2011, not 2007 holds the prior record for minimum ice volume. You might read that off the anomaly chart if the annual minima were marked but this would clutter the story told about the anomaly which is pretty steady decline at all times during the year. The winter ice volume does not recover even if the extent usually does. The summer ice volume falls steadily even if extent can make a bit of a come back after 2007.
Which means tomorrow it will go sub 4 into the 3′s! Look at the line for 2012 – that’s sharp downward line for this late in the season! Wonder what the final tale of the tape will hit. I’m guessing the new record will be +-3600. 500k sq. k’s less than 07 really ought to turn some heads.
#153–”As we are probably going to sea a ice free summer arctic within the next 5-10 year, does anyone know how soon the longer and longer periods of ice free conditions (or another 5mil sq kms of open dark ocean) will really begin to principally affect the northern hemisphere climate?”
Lawrence, here’s what Neven and I came up with in regard to the large question:
Some preliminary questions and thoughts, in line with what research we could find. Comments and further suggestions are very much welcomed; as I commented on the thread for the post, it seems to me (as to you, apparently) that we should be looking toward consequences a bit more than it seems we have heretofore.
Am I overlooking something, or is my bafflement with 2007 as a particularly standout year justified?
I admit that 2007 was what got my attention, but then I wasn’t exactly focused on the issue. Leaving that aside, what was unique about 2007?
In retrospect we been banging off records since the very first satellite derived extent measurements. Really, the satellite record doesn’t even appear able to tell us what’s “normal.” It seems fairly obvious we started watching the series quite a few episodes into the plot.
See this record of anomalies for July, for instance. Looking at that, what’s so special about 2007, or this year for that matter?
Let alone Watts, many of the rest of us were pretty slow on the uptake.
Wow are we ever going to be judged harshly, particularly the better the hindsight we have.
2007 broke the record for lowest NSIDC Extent by a HUGE margin, largely due to more-or-less perfect melt and ice transport conditions for the entire summer. This captured a lot of media attention at the time, which continued as several subsequent years got close but failed to reach the 2007 low.
This low has now been broken, so you may see 2007 treated more as a strong exemplar of an underlying trend, and less as a freak one-off.
ozajh: …2007 broke the record for lowest NSIDC Extent by a HUGE margin…
Larger than the mostly unremarked previous record-holder 1995, sure. On the other hand obsessing on the grandiosity of one year’s anomaly is just a slightly less myopic way of failing to see the trend. Two lesser anomalies added together can be more important, more are worse but those go pretty much unremarked in the daily noise.
“Nothing to see until 2007′s record is broken.”
Are we going to see a replay of this? I suspect so; there’s nothing in the record suggesting we’re likely to see year-on-year annual anomalies, though monthly records have examples.
So, we’ll keep on getting used to this, or being used to it in actuality. I’ll be surprised if this year’s anomaly generates the same press notice as 2007, more surprised if in 2-4 years when we see this year’s record broken there’s much attention paid at all.
We’re just not very good at this sort of thing. Same deal as ignoring 10,000 deaths per year in automobiles that could be reduced by half with some engineering. A steady 10,000/year excess morbidity is boring. Twelve dozen killed by a defective skybridge at a hotel is the sort of thing is guaranteed to get our collective notice.
We have some very bad problems with our perceptions and cognition not scaling very well.
dbostrom, ozajh -
If you look at http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/sea.ice.anomaly.timeseries.jpg, they eyecrometer shows that from 2007 on, there is a strong annual cycle. The 2007 melt was he result of enough prior year thinning that the sea ice was no longer an integrator representative of climate (warming causing declining ice), but a responder to weather. Unfavorable weather was able to compact the thinning ice and flush larger quantities out the Fram Straight, where in previous years that bad weather wouldn’t have had such a dramatic effect on the thicker ice. This year, which hasn’t had such bad weather for ice loss, but still has impressive declines, is a second transition, probably a tipping point. The ice thickness last winter finally became so low that even normal weather is causing a crash – and the thickness isn’t going to recover anytime soon to where it will take a bad years weather to cause a 2007 style precipitous drop, let alone to where it can maintain a pre 1950 10 million km^2 extent.
A potential feedback for thinner winter ice(some ice will always form when the sun sets in the arctic) is that so much open water will be a large new source of atmospheric water vapor over the pole, which will result in rapid early accumulation of thick insulating snow cover. Once the ice reaches a thickness that is mechanically stable to wind and wave action, preventing wetting of the snow(0.5 meter?), the insulation will prevent much further growth in thickness even with -40 degree surface temperatures – and it’s possible that thicker snow will allow much lower winter temperatures than currently seen in the Arctic. When there are outbreaks of Arctic air in coming winters to New York or London, they will be deadly, IMHO.
Here’s a chart showing maximum potential solar insolation for 0, 30, 60, and 90 degrees north. Notice that for the arctic, the second half of May, and June and July are the times when the most sunlight is available to penetrate open waters, with about a month of shoulder on each end. September is about residual heat in the water. Even this chart is probably misleading because water reflects better at lower angles.
Here’s Arctic Roos’ area graph. Note that for the high solar period in 2012 ice area started out high and ended low as compared to recent years. It looks pretty typical given a visual weighting using the potential insolation chart. Thus, sea ice albedo isn’t significantly involved in the situation yet (beyond what it already was), but in a year where area doesn’t start so high, a lower April 15 through May 15 albedo could prime the system for yet another record smashing year.
#173–Mmm, that “shoulder” continues to deliver better than 200 W/m2 to the end of August. Clearly, it’s much smaller than the insolation during the height of the season, but I don’t think that even this amount is negligible–after all, the relevant comparison isn’t to the height of the season, it’s to previous comparable periods.
2005: 5.315 million square km
2006: 5.781 million square km
2007: 4.255 million square km
2008: 4.715 million square km
2009: 5.250 million square km
2010: 4.814 million square km
2011: 4.527 million square km
2012: 4.189 million square km (and running)
(h/t Neven’s Arctic Sea Ice blog)
We’re 70,000 km2 and a couple of weeks ahead of 2007, and more than a million km2 ahead (and again, a couple of weeks) of some other recent years. It should be pretty easy to do ‘back of the envelope’ calculations for just how much energy that represents in each case; I haven’t the time or the heart. But though I’m willing to believe that it’s not a huge proportion of the seasonal fluxes, I’m not willing to believe it’s negligible.
Moreover, if you look at the extent curve for the season, here:
…you see that 2012 has been tracking below 2007 since about the second week of June–IOW, right through the heart of the ‘insolation season.’ More energy than ever before (disregarding the effects of weather, of course–it’s very possible that 2007′s sunny weather more than compensates) has been reaching the sea for that whole time.
So, Jim, all in all I think you are overstating the case WRT insolation, at best.
Re the PIOMAS anomaly bounce-back after July: This year the late-August melt has been unprecedented, continuing to head steeply downward (area/extent) when all other years had begun to bottom out. This would suggest that the PIOMAS anomaly may be headed for an atypical Aug downturn this year. We’ll see in a few days.
176 Kevin said, “e that 2012 has been tracking below 2007 since about the second week of June–IOW, right through the heart of the ‘insolation season.’”
I wasn’t paying any attention to 2007. It’s all 2012 VS average. Before the solstice the ice was high (but only in context of this decade), and afterwards it was low by a similar magnitude. It crossed median a bit before the solstice, according to the area chart I found.
I agree with your definition of “shoulder” and I don’t know what the energy loss is, so I don’t know when solar input is matched by outbound IR, so I couldn’t tell you if 200W is significant or not.
Examined up close, the PIOMAS ‘bounce back’ seen since 2010 is well into it’s stride by the end of July as graphed here (usually 2 clicks to ‘download your attachment’)
It is probably fair to anticipate 2012 will become a ‘double dip’ year with the impact of the cyclone. All will be revealed in the next PIOMAS update.
#182 dingibily (nice nickname). No. A cyclone consolidates ice towards its center, a strong Anticyclone would have been more devastating. The ice was melting quite fast despite a generally cloudy arctic summer.
Comment by Andy Lee Robinson — 28 Aug 2012 @ 6:51 PM
Since December, 2011 I have been giving presentations linking Arctic sea ice destruction with extreme weather, etc. Since August 10th, 2012 I have been predicting that the sea ice will be completely gone from the Arctic by the end of this years melt season. Here are presentations that lead me to conclude this…I post all this stuff in real-time on facebook (Paul Beckwith) and twitter (PaulHBeckwith); please friend and follow me. Note that the last link is a presentation I gave on Jan 17th this year discussing all the sea ice links to extreme weather, etc. and the imminent sea ice failure…rather than publishing papers on this for my thesis I have been focused on getting this information out to the public, and contributing to AMEG (Arctic Methane Emergency Group) work.
We went to Alaska this summer. And even as far south as Anchorage it was easy to see that the Arctic is going to be cold for a long time. A complete melt-out will never be seen by me or thee. By the time it warms enough to have produced a melt-out, we’ll be dead.
If it happens quickly, there won’t be hominid witnesses at all.
Paul B, 185: Arctic ocean will contain nary an ice cube some time in 2012.
dhogaza, 186: Eschew wild-eyed predictions.
Jeffrey D, 187: No complete melt-out (c’mon, Jeff, definition, please) in the next several decades.
What do I mean by “complete melt-out”? I didn’t think that was hard. No ice in the Arctic. As in “complete”.
I’ve been wrong before, but there’s still a lot of ice up there. Think about all the ice in Greenland. It will really need to warm up A LOT to melt that. Do you think you’re going to live long enough to see that? I don’t think I’ll see it, but I’m in my 60s.
For me to see it would require real fast warming. The political/agricultural instability that kind of rise would entail would preempt witnesses, don’t you think? I think it would require ~1C more of warming to accomplish a melt-out. If you’re in your 20s now, you’ll be in your 70s by the time we get another 1C of warming. D you think you’ll see it?
I’m a genuinely bleak catastrophist by the way. The political inertia/reluctance to tackle this issue just confirms my misanthropy.
The PIOMAS data along with estimates from the Greenland an Antarctic Ice Sheets indicate we are permanently melting around 800-1000 Gigatonnes of Ice per year. A back of envelope calculation suggests this requires more than 1% of the total solar input to earth each year.
Has anyone calculated the impact this is having on the rate of warming?
[Response: You have made a mistake somewhere: 1000 GT ice melting a year is roughly 3x10^20 J/year. Solar input is 240 W/m2 so over a year is 240*5.1*10^14*365*24*3600 ~ 4x10^24 J/year. So more like 0.01% of solar input. Alternatively, the energy in W/m2 required to melt the ice is about 0.02 W/m2 compared to a current radiative imbalance of ~0.5-1 W/m2. - gavin]
To my eye, nice graphic on sea surface temps, with two steep drops. Will there be another one? Time will tell.
On ice-free, I seem to remember official definition is for a short period (day?) and with a small proportion of ice remaining (thin, lose, broken?). Sloppy layperson that I am …
Meanwhile, with Leslie and Son of Isaac (see Wunderground, S of I off topic a bit for my point, and your detailed weather reports) there appear to be some issues about weather perturbation in process. In particular, those northward heading low pressure whirligigs are getting a bit closer to the Arctic etc. given the current state of agitation and meltdown. Time will tell.
It has occurred to a number of observers that models are now lagging real-time. I know modeling has been the norm, but at this point I’m wondering if a better incorporation of real-time observations needs to occur. I know, it’s easy to shoot from outside, but since there is so much criticism of models it bears repeating that so far they are understating the case in multiple ways.