Apocalypse, Those thickness charts are very concerning to me when combined with my Barents Sea current-Lomonosov Ridge theory. It sure looks like the ice is thin along the ridge, then along Greenland where the current would turn. Maybe this is always the case, as that current must dip under the ice no matter what (are past years' images available)? But it makes me think even more that we could very suddenly see the ice melt "from beneath" along that line... and it looks like such melting is continuing this year. That would be very frightening, to think of the ice continuing to melt from underneath in autumn even as it rather inconsequentially freezes (thin) and extent increases around the fringes.

Terry, Perhaps my thinking on it is muddled. I was unsure if something along these lines happened in the past -- a warm current dividing the pack but without necessarily melting it in the Beaufort Sea), if an "open water path" down the middle of the ice pack could give the impression of less ice in the studies (by cutting the transport of the wood from Tiksi to Greenland via ice) while not actually reducing the extent to as great a degree as has been inferred.

Thanks. And is it also possible that more of the current is also staying near the surface, so it melts the edges that way? This is interesting. It would suggest that as melt progresses this year and in future years, we will see the warm current follow the Lomonosov Ridge (but near the surface) and so split the ice pack in two before melting it completely away. It could even start to hit the coast of Greenland and melt the ice along the coast before it melts completely elsewhere. I wonder too how this affects those studies that made assumptions about driftwood found (and carbon dated) on the coast of Greenland. The possible dynamic is intriguing.

Addendum to my last comment. Note that the Lomonosov Ridge is just about where that "hole" is in the Laptev Sea. A layman's guess would be that some of the warm water from the Barents Sea which has been forced under the surface hits that ridge and is deflected upwards, causing the hole in the ice.

I recently posted the following comment on Skeptical Science: http://www.skepticalscience.com/news.php?n=1615&p=2#85009 Basically, I've noticed two things: 1) Extent has stopped dropping for various compensating reasons, but the ice continues to melt at a rapid pace in the area between the North Pole and Severnaya Zemlya. 2) Combined with the Nasa recognition that the freshwater flow of the Canadian currents have changed (http://www.nasa.gov/topics/earth/features/earth20120104.html) Can this mean that the warm saltwater current from the Barents Sea is pushing further towards the pole (rather than dipping under the freshwater and icepack, neither of which is there anymore), which in turn causes the continued fast-melt... and would be a mechanism that would tunnel into the main icepack off of Greenland over the years? Does anyone understand this stuff well enough to comment knowledgeably?

I'm concerned that the calculation is inadequate. First, it doesn't seem (at first glance) to consider the angle of incidence, the length of day, and most importantly the distribution (sparsity) of sunlight over a larger land area near the poles. Second, and more importantly, it doesn't appear take into account the time factor. Is ice gone by August, July or June? I've said in various places that we will soon switch from watching minimum extent (which will always be near zero) to minimum extent date (which will become earlier and earlier in the year). The length of time that the ice is gone will be important. Or does this calculation presume that all ice is gone all year, in which case we are still (I hope, please I hope) some time away from such a scenario?

I'd like to ask again. Is anyone looking at and able to read the weather maps at http://www.wetterzentrale.de/topkarten/fsecmeur.html ? The next five days show a series of three tightly wound lows taking up position in the Arctic. They differ from the Great Cyclone in that they are bounded mostly by other lows rather than highs, but they do have closely wound contours and some spread in vorticity (not that I really know what I'm talking about -- I haven't yet gone through enough material to really, properly read those maps). Does anyone who understands weather know what this could portend? Could the low just off the Fram Strait create winds that really empty it out, while the one just over the pole compacts and chops up the ice at the edges in the Beaufort and East Siberian seas, and the third out in the Bering Sea waits its turn to move in and do whatever it wants to do?

Can anyone read the weather maps at: http://www.wetterzentrale.de/topkarten/fsecmeur.html Is that jumbled mush of lows looking like anything scary developing? [Okay, okay, I've added Great Arctic Storms to my list of "Things to Be Afraid Of"]

Any comment on the ice drift as reported/projected by PIOMAS? What does it mean? It looks to me like it's mostly going to compact the ice further, and maybe push more of it (some of it the loose ice in the Laptev Sea) down the Fram Throat. I mean Ice-Grinder. I mean Strait.

L. Hamilton, [Sorry, I got distracted by my paying job.] Do the people who have Chrome see the old one after they hit reload? I'm using Firefox, and I got 8/22 first, then 8/23 after a reload. Which browser are you using? My advice would be to find another upload site... maybe Mediafire? Look for one that either sets no-cache, or else lets you set the cache control yourself (which probably doesn't exist... I can't imagine many sites bothering with that capability). There are some other tricks you can try. For instance, add a meaningless but changed parameter to each new link, like http://img.photobucket.com/albums/v224/Chiloe/Climate/sea_ice_DMI_delta_this_month.png?d=08232012 That extra, meaningless (and therefore ignored) parameter makes the browser think the result may be different from what it got the last time, so it ignores the cache. Of course, if someone follows the same link twice, they may still see the old cached entry (depending on the browser), so that parameter value has to change. So this is like changing the tinyurl, but at least when you embed links into comments you can get the right behavior for people.

L. Hamilton, RE: Graph/website question Data caching is controlled by the host, meaning photobucket. They are telling your browser "don't bother checking for a new copy of this image for X days." Photobucket decides/controls X. So the short answer is no, there's nothing anyone can do about it, short of: a) You can turn off caching in your browser (but not by web site or URL... it will affect everything everywhere). b) You can hit "reload" after you see the old/out-of-date image, and that basically tells the browser "ignore any cached thingies and go get everything fresh", so you get the latest image. Alternately, if people put photobucket images up with entirely new names (like by appending the date to the name), obviously you won't have a cached version of a file with an old name and there's no problem, but then you need to know the name of each new image. One way to get around this (but it takes work on the part of the uploader) is to use a tinyurl. a) Upload the image, named by date b) Create a tinyurl to the image, and give that to people for access c) When you upload the new image, use a new name and change the tinyurl to point to the new image.

Glacier watching on MODIS this afternoon I spied yet another large chunk that just calved off of the Steensby Glacier (if I'm reading my maps right). Not as big as the Petermann Ice Island 2012 -- maybe a fifth of that -- but still... One more big block of ice floating away, one that was there in every pic I look at from last year and two years ago (clear view on 2010217).

Oh my gosh, it's so hot, even the islands are melting! [Sorry. This is all so depressing, I can't take it. I need some levity in here somewhere.]

Just as an observation... the paradigm is changing. I don't think it's necessarily correct to assume that we'll continue to lose more ice from this point because we have in the past. We almost certainly will, but the sun is getting very low in the sky, and what ice remains is getting closer and closer to the pole. We may well find that we've reached something of a wall, in that what doesn't melt by now stays, while the new "usual point of minimum" is early instead of mid-September (or even late August?). We will reach a point at some time in the future when the melt happens fast enough (weakened by year after year of more and more multi-year ice loss) that it all goes very quickly, but I think at least for this year there is some chance (barring things like cyclones churning up what's left) that extent loss could slow to a crawl or even halt, because the ice that remains is out of serious reach of the sun by now. I'm not really saying that will or won't happen. I'm just saying we can't assume that we'll have a loss from this point equivalent to previous years -- because the model has now changed substantially. That's not to say that melt won't continue to occur, which gets us further and further behind for next year... but the melt may not have much impact on extent. It may focus on the thicker ice near the coast of Greenland, or on thick ice that's been pushed into the Fram Strait where it will now thin somewhat, but not get below that 15% threshold.