songster

No. East Siberian methane releases are estimated at ~0.5 Mt per year (= 0.5 x 10^9 kg). Methane's density as a gas is 0.66 kg/m^3, so that comes out to be 0.75 x 10^9 m^3 of methane released per year. The area of the East Siberian shelf is >2 million km^2 (=2 x 10^12 m^2). Divide one by the other, and you have 375 cubic centimetres of methane released per square metre per year. That's one bubble with a diameter of ~1.5 cm per day per square metre. That one bubble per day is supposed to mix the entire water column of 50-100m depth? It's not even within four or five orders of magnitude!

It looks to me as though the losses in the Laptev at least are illusory. If you look at it on MODIS, you'll see it's still a solid ice sheet, but over the last few days it's turned blueish (which I believe indicate melt ponding). Melt ponds are often erroneously seen as open water by the microwave satellites. http://rapidfire.sci.gsfc.nasa.gov/imagery/subsets/?subset=Arctic_r05c05.2012155.terra

To be fair, I think the relevant picture is this one, where the coverage is almost complete, but there's a hole in Hudson Bay: http://www.ijis.iarc.uaf.edu/cgi-bin/seaice-monitor.cgi?lang=e The one you linked is today's data, which will (presumably) get more swaths added in the next few hours, and then a preliminary extent figure published tomorrow.

Since the satellite change, the latest (pre-adjustment) daily IJIS value has been useless. It's always showing a loss of ~200k or more, and then a huge upward readjustment the next day. Just don't look at it, it's obviously only a partial reading. Possibly something to do with the satellite's orbital parameters meaning they don't have full coverage in a single 24 hr cycle?

Cameras are up, but the feeds aren't yet grabbing images from them. Their main homepage is here: http://psc.apl.washington.edu/northpole/ ... which has links to the pages for each camera http://psc.apl.washington.edu/northpole/NPEO2012/webcam1.html http://psc.apl.washington.edu/northpole/NPEO2012/webcam2.html ... from which you can get to the archives http://psc.apl.washington.edu/northpole/NPEO2012/WEBCAM1/ARCHIVE/ http://psc.apl.washington.edu/northpole/NPEO2012/WEBCAM2/ARCHIVE/ ... which has the actual latest pictures for each http://psc.apl.washington.edu/northpole/NPEO2012/WEBCAM1/ARCHIVE/npeo_cam1_20120508100338.jpg (as of 8th May) http://psc.apl.washington.edu/northpole/NPEO2012/WEBCAM2/ARCHIVE/npeo_cam2_20120508123257.jpg (as of 8th May) Camera 2 seems to be slightly damaged electronically, with some imaging artefacts and loss.

Don't be silly. If you spend 5% more on groceries this year than last, it doesn't mean the price of a slice of bread has gone up by 0.25 pence.

Yes, but the point is the only "recovery" cohort is the ice that froze during 2008 winter (and is now ~3.5 years old). There is less 1-2 year ice this year than last, therefore next year there will very likely be a reduction in 2-3 year ice There is less 2-3 year ice this year than last, therefore next year there will very likely be a reduction in 3-4 year ice Overall, MYI is going down, and you can't keep looking to that one good year for salvation since we already know that the "recovery" hasn't been sustained. For MYI to increase year on year, then the summer minimum has to also increase year on year, not just blip up and then back down again.

http://nsidc.org/arcticseaicenews/files/2012/04/Figure5.png Multi-year ice (defined as 2+ year ice) is down on last year, as is perennial ice (all >1 year). There is however an increase in one specific cohort of multi-year ice, namely 3-4 year ice. This cohort is expected to be larger because of the temporary recovery in minimum extent in the summer of 2009. Ice that formed in winter 2008, and which didn't melt out in summer 2009 due to the temporary recovery, is now ~3.5 years old. Similarly, you can see that last year there was a "bulge" in the 2-3 year cohort, and so on. However, that's one cohort, and is only a small proportion of the total ice.

The only thing you can conclude from this is that the aerial survey gives (slightly) larger numbers than previous catch-and-release data. That's not surprising as they're surveying a wider area. Any comparison between methodologies is meaningless until validated, and thus far the aerial survey data has 1 (one) data point. You cannot build a trend on one point. A more extensive Q&A with a polar bear scientist from the area is here: http://mediamatters.org/blog/201204090004

As for drift off from the pack edges (spread from 100% concentration to less) this has tended to 'bully up' some posters as they mistake this as growth. This can't be a very significant effect as area and extent are staying pretty much in lock step. At this time of year there are no melt ponds to confuse the issue, so it's safe to compare area and extent directly. The growth is real, if thin.

Nope, gonna wait for some early season indications. It's down to how well the Beaufort and Chukchi seas hold up this year. They've been colder than usual, and had a lot of compression that may have ridged/thickened the ice. Kara and East Siberian seas are however looking vulnerable. If we have a repeat of the last few years' early melt in the Beaufort/Chukchi, then I'll agree a new minimum or near-minimum is likely.

The PIOMAS team don't make any projections whatsoever from their graphs. All those graphs were drawn up by people in the blogosphere, not the scientists actually working on the PIOMAS model. Note also that they were explicitly drawn up as thought exercises, not as formal predictions. The scientists did however produce a prediction a while back based on running the model forward (not simply extrapolating from graphs). It forecasts a collapse of Arctic ice extent some time in mid 20's - 30's, i.e. a bit ahead of other models, but not drastically so. http://psc.apl.washington.edu/zhang/IDAO/multi.html

To be fair, the Greenland and West Antarctic ice sheets did melt (or at lease shrink significantly) at that time, it's only the East Antarctic sheet that didn't. Losing any one of them would be bad news for us.

In the WUWT thread, Julienne Stroeve posted the link to the NSIDC's unified dataset going back to 1972. They don't generally show it because of the technical issues already covered, but it's on their site and freely accessible. She also offered to provide datasets going back even earlier to anyone that contacts her, on the understanding that these are works in progress with (likely) large and unknown errors associated with them. ftp://sidads.colorado.edu/pub/DATASETS/nsidc0192_seaice_trends_climo/esmr-smmr-ssmi-merged/

Click on the year number to toggle any given line on/off.

Interesting, but frankly tangential to the real issue, which is the speed of change. If (say) the Greenland ice cap is doomed, and destined to melt fully away over the next 50,000 years... should we really care? Even with 8 degrees C warming, their model says it still takes 2000 years to melt the Greenland ice cap - i.e. a sea level rise of ~0.8m / century, and noticeably less than that for the first few centuries. Perhaps perversely, I find this reassuring. However, I understand that the rate-of-melt question - despite being the most important - is the one that's most poorly characterised. At this point I'm not sure we need more models of equilibrium climate change: we need much better understanding of the non-equilibrium dynamics of ice cap melt.

Fair enough. The planet is (almost) a sphere. Therefore, pretty much any coordinate system you choose will have pi in it somewhere. That's not only irrational but transcendental. How accurately will you represent that in your model? The fundamental point is that the real world has infinite precision, while computation does not. The minute additional error from choosing one digital representation over another isn't exactly the largest source of error in the process!

Rewrite these things to use a decimal representation At which point you can't represent 1/3 accurately, although you can now represent 1/10 accurately. Pop quiz: Do you think the real world actually uses exact rational numbers?

Because Hansen was writing a paper about his climate model, which happened to show a climate sensitivity of 4.2 degrees. As you quote, this was one of two models that collectively defined the agreed consensus, and at the time it was not known which model was right. In the 20+ years since then, it's been shown that the truth most likely lies somewhere in the middle, but higher climate sensitivities are not yet ruled out. Basically, it boils down to how much sulphate aerosols China has been emitting, and how much those have held global temperature down in the last decade or so. Note also that in the very first paragraph describing the model (in the 1988 paper), you find this section: "The equilibrium sensitivity of this model for doubled CO2 (315 ppmv -> 630 ppmv) is 4.2°C for global mean surface air temperature (Hansen et al. [1984], hereafter referred to as paper 2). This is within, but near the upper end of the range 3° ± 1.5°C estimated for climate sensitivity by National Academy of Sciences committees [Charney, 1979; Smagorinsky, 1982], where their range is a subjective estimate of the uncertainty based on climate-modeling studies and empirical evidence for climate sensitivity." It's hardly some kind of grand conspiracy given that the very first act when discussing the model is to discuss the then-existing disagreements over climate sensitivity! Moreover, the sensitivity is an output of the model, not an input - it would be utterly wrong for him to arbitrarily pretend that his results didn't give a figure of 4.2 degrees.

Essentially they've started as a wide scatter and converged somewhat. Wikipedia has a reasonable overview: http://en.wikipedia.org/wiki/Climate_sensitivity

http://www.realclimate.org/index.php/archives/2009/12/updates-to-model-data-comparisons/ Covers this quite nicely. The 1988 Hansen paper used a value of 4.2 degrees C for climate sensitivity - it's now believed to be about 3 degrees C. Once you adjust for this, the forecast agrees with measurements to within the limits of accuracy. Note that in the real world, emissions have been running slightly below Hansen's scenario B in the 1988 paper.

Hum, I wouldn't be so certain. Yes, the southerlies have kept the pack low on the Kara/Barents side. However, they've also compressed (and thickened?) the central pack and shifted the distribution of thicker ice over into the Beaufort and Chukchi seas. To an untrained eye, it also looks as though there's also been less export through the Fram Strait than usual, aside from a couple of weeks near the start of the freeze when it was absolutely booking it out of there. If you're not careful, there's a danger of falling into the trap of seeing every possible wind direction as catastrophic. Winds blowing towards the Pole? Then the heat transport will be keeping the ice thin. Winds blowing away from the Pole? That's the old ice being flushed out to melt in warmer waters. Winds blowing round in circles? That's churning up the ice and proving it's in poor condition. All told, there's been a lot of very cold air bottled up north of Canada for much of the winter, more than the last few seasons. I suspect therefore that melt in the Beaufort and Chukchi seas will be slower in 2012 than in 07-11. However, I agree that if losses in these seas are <i>not</i> substantially reduced this year, then yes, we're probably looking at a record or very near. That's simply a logical consequence of how low it it on the Atlantic side of things.

My understanding is that there's some evidence for a small effect (1-2% increase in earthquake frequency) associated with lunar tides: especially for (a) very shallow earthquakes and (B) slow energy release through microquake clusters without a major rupture. As far as I know any potential solar gravitational effect on earthquakes and/or volcanoes has been debunked many times over.

I can buy an effect on cloud formation, weather etc. - but volcanoes? What on Earth could the mechanism be? And what evidence is there of any such effect?

Um... so the Sun causes volcanoes now? Please tell me I misread you.