Looking at the PIOMAS volume anomaly curve, one sees three distinct downward peaks in the last 3 springs/summers. The last 3 years, the volume development over the year was definitely and consistently different from the mean. This could be another indication of a new state of the weather system, although those are at most semi-stable. The coming summer will corroborate - or weaken - the assumption of a new weather state. My personal guess is 0.6 probability for a corroboration.

Concerning the map: my favorite is the Goode approach, an equal area projection minimizing the distortions at the price of cutting the map up. http://en.wikipedia.org/wiki/Goode_homolosine_projection

This Cryosat data release story is absolutely sensational, a true scoop!

Found the problem - a ")" at the end included in the URL. http://www.physicalgeography.net/fundamentals/6i.html is correct.

@bouke van der spoel: your link to physical geography seems not to work - could you repost it?

@SATire: Most of "Spiegel"- commenters (and everywhere in the world) just can't deal with uncertainty. They didn't learn it. They have no feeling to weight the importance of this or that kind of news. They are lost and resort to dumb boldness, just to reinstall their feeling of self security. To blame are partly the media, which prefer clear and dramatic news over subtle and sound analysis and, after all, are made by journalists who themselves all too often didn't learn do deal with uncertainty. But fortunately the dumb-bold type - though often dominating the comment columns of newspaper websites - seem to be a minority.

Not only sea ice but also snow cover development isn't reflected in numerical models: http://www.nature.com/news/arctic-snow-cover-shows-sharp-decline-1.11709: "Derksen says that scientists need to understand why the observed changes do not match up with the projections of widely used models. He found that the snow-cover projections generated by the climate models being used in the Fifth Assessment Report of the Intergovernmental Panel on Climate Change underestimate the extent of spring snowmelt in the Northern Hemisphere. 'Even if we’ve become a bit more willing to be aggressive with the scenarios we use to drive these models, it still doesn’t seem to be enough to describe what we’re observing,' says Sharp."

Thinking about it, I had to give my original blog post an overhaul. As often, things are more complex than they appear at first sight: The snow data are “extent” – data which means, that they contain 10 – 15 % snow free area (not 100% sure about this point). The albedo of land is bigger than that of water (ca. 0.2 vs. ca. 0.1,http://www.climatedata.info/Forcing/Forcing/albedo.html). Contrary to intuition, the mean solar irradiation in mid summer in those areas, where the main snow melting difference occurs (around the 60° N) is lower than on the polar sea around the north pole!(http://tamino.wordpress.com/2012/10/01/sea-ice-insolation) The data for other months than June are not given in my source, probably in the paywalled original paper, so they might look less crassy. On the other hand, the capacity of heat absorbtion of land is smaller than that of sea surface, which will direct the absorbed energy earlier into atmosphere instead of the water body. Also, the speed of recent summer snow loss seems to be higher than that of sea ice.(http://www.livescience.com/26091-arctic-snow-cover-decline.html)

This is all quite heavily OT, but anyway: With cap&trade vs. carbon tax, there are pros and cons for each. Predictability: there are statistical ups and downs of energy price anyway, which make planning for industry difficult. On first look, a thoroughly planned rising carbon tax gives more predictability. But c&t can act stabilizing on economy, as with low demand, the certificate price will drop also. Ease of implementation: carbon tax wins, but: noone knows which tax path yields which emissions reduction path. With c&t, reduction is well-defined. At the end of the day, these are details of lesser importance. Hansens proposition is perfectly tailored for the notoriously state hating US. If it is necessary to make enough GOP diehards to embark on climate saving, it's the best.

http://ocean.dmi.dk/arctic/meant80n.uk.php shows consistently warmer winters the last years which cost us 5 - 10% freeze rate.

@Aaron: You spoke of freshwater lenses under the ice (supposedly MYI). Where they melt water from the last summer, staying at place, or river influx (then only at the fringes)? Just for my understanding. @crandles: Yeah, wasn't aware melt is mainly radiation driven and then those absorbtion effects kick in. "Um the freeze rate determines the maximum, which seems a strange thing to call a baseline." The idea behind is that a stationary minimum summer thickness for a given sort of year supposed to repeat itself a number of times is where winter freeze-up and summer melt-down equal. Both depend on summer minimum thickness. So I called this "baseline of the thickness oscillation". If there is no summer thickness of course a "baseline" makes no sense. The absorbtion increase of thin ice neutralizes somehow the stabilizing effect of fast freezing and make the temperature dependency of summer ice thickness much more hefty.

The thickness growth rate of thin ice - with all other conditions kept constant - is bigger than that of thick ice. The effect of 10 cm ice more or less is most pronounced if the ice is thin; the thickness-freeze-rate-relation of thick ice is much weaker. So the effect of a late freeze start on maximum thickness is diminished by this, but not cancelled out. The thickness-freeze-rate-dependency is stabilizing the ice sheet. (Also, it is one function, which determines the baseline of the thickness oscillation. The other is melt rate, which has a weak thickness dependency, afaik.)

There is not only a spring- and summer- snow cover feedback from less snow. More snow _on the ice_ in winter will decrease ice thickness growth rate, as it will prevent the heat from below to escape to the atomsphere.

I can't help but I dislike the attribuition of any event, be it a storm or a winter, may it have soandso many people killed, be they as improbable as it may be, to global warming. It all sounds sensational and interesting, but under the bottom line I only trust statistics. There have been harsh and early winters before in Russia, e.g. the one that helped keep the German Army 1941 from taking Moscow.

I drew a little bit of Flak because of the somewhat megalomanic proposition with the mass-produced thickness sensors. I was of course well aware that an outsider project of such size, which may or may not work out, has no real chance. AFAICS the 70s-Sandia-penetrator project was a ballistic penetrator with deceleration sensors radioing its data up during or after impact - not a soft landing automatic station meant to remain stationary for months. It may have a lot of problems hard to solve but surely not deceleration on impact, as Peter Ellis so sarcastically suggested. Could be an interesting interdisciplinary development project of higher level science, engineering and manufacturing technology students... So, and now I give You a break with this.

Concerning volume: If it's not taken by IPCC (I did not read the draft and also, there will surely be modifications) then for the simple reason, that it cannot be measured directly and so any data have a pretty wide uncertainty bar and are open to attack. A very unsatisfactory situation for years now. I propose herewith to invent cheap mass produced air droppable floating ice probes, which drill automatically a sensor stick through the ice and keep track and transmit ice & snow thickness, salinity, temperature, radiation balance. Then spray the ice sheet with thousands of them. In the end cheaper than cryosat, and very lsmall error bars at least pointwise. A few million Euros - one single cleanroom lab for semiconductor processing costs several times this amount.

@Boa05att: Yes, this is what has not been done in Fukushima, where the risk of a big earthquake had been calculated as once in 1000 years or so, which seem small at first glance , but yielding 1/30 within nominal reactor lifetime, which is very big, taking into account the potential (and then real) damage.

The IPCC statement might be correct, say: still not completely false - but a note on the recent deviation is the least to expect. Then, how about ice volume? We wait for the first model to get the accelerated melting right in the middle of its probability field. The interesting question is, where the main differences lie between models and reality. Models failed, but where exactly? Is it ocean currents? Is it atmospheric heat transport? Is it pressure distribution, cloud cover and insolation? Is it thicker snow cover with subsequent slowing down of winter ice thickness growth rate? So many questions and no comprehensive emission certificate system in sight...

IMHO the fear of exponential growth is exaggerated, because although there are positive feedbacks, none of them will grow infinitely. E.G. albedo will never decrease below a certain value, and of course solar irradiation won't. Atmospheric heat transfer is limited by both circulation and heat transfer coefficients at the surface, and so on. As the mechanisms are complex and contradictive no simple function is distinct in its predictive capability and earns more trust than a linear or at most a quadratic. Only functions based on a - however simple - physical modelling are of any value. That said, I put my penny in with the value of sea level rise above now in the year 2050, supposed the quadratic holds for that time (which is of course questionable): delta_t = year - 2012 melt rate i Gt/y ~ 450 delta_t + 30 delta_t², integrated over 40 years = Delta_V = 450*40+30*40²=66000 Gt = 66 e3 km³, earths water area = A = 361 e6 km², sea level rise = Delta_V/A = 66/361 m = 0,18 m.

One humble remark: I think one of the most important factors concerning melting is solar irradiation. So any reasoning on happening or not happening melting in a certain region should be based on irradiation history of the weeks passed. Unfortunately I dont have one. Dominik

Could you post a link to where one can find this beautifully colored map above?

With this heat transfer thing, I had in mind something like changing medium time scale weather patterns, like the north atlantic oscillation, which may change its sign and consequently break the trend. Actually this is a denialist way of arguing. You are of course right insofar as any trend already started some years ago is incorporated in the statistical model. (An other interesting question is, whether the north atlantic oscillation isn't itself broken by sea ice levels so low - that is, whether there is a "weather pattern tipping point". This would then be the kind of nonlinearity you mentioned.) Concerning the packing of ice against the Greenland and Canadian Islands north coast as cause for the higher ice thickness there i would be thankful for one or two links.

Actually I don't understand many of the details of the above analysis, but as it uses a purely statistical model with no physical elements in it, its worth might be limited. As I read, the Gompertz - function is used in population or usage statistics, where it describes some kind of saturation behaviour. In our case, the driving force, i.e. the heat transfer into the arctic, may well have significant variations, which may accelerate or slow down the process considerably and unexpectedly. Also, with steady driving force, the behaviour of the now-thickest areas of ice is of importance - it might well be, that it melts all away! There we have to enter some discussion about where those areas are (in the nearest vicinity of Greenland and the Canadian Islands north costs), why there of all places (influence of ice shields???) and wether some protecting function of the adjacent land masses will be sufficient to keep a rest of summer ice over a longer time "alive".

Ah, yess, I imported the data in excel and said to excel: "use blanks as separator". Somehow I guess it took the blank in the first line and split title *and* the numerical data into 2 columns. Something like that. Thanks for the remarks.

Hi, I appreciate Your work very much. I'm interested in the data table linked to above. If you could give a hint, why for each region not only colums "extent", "area" and "volume" are given but also 3 colums with the area title on top (e.g. "baffin", "baffin" and "baffin"). Also a word about the units would be helpful.