tag:blogger.com,1999:blog-5337555368793819627.post3819331595642971282..comments2023-06-07T09:04:36.390-04:00Comments on More Grumbine Science: When will Arctic ice be gone?Robert Grumbinehttp://www.blogger.com/profile/10783453972811796911noreply@blogger.comBlogger13125tag:blogger.com,1999:blog-5337555368793819627.post-76733044592828491022011-04-02T07:17:14.550-04:002011-04-02T07:17:14.550-04:00Pretty good post. I just stumbled upon your blog a...Pretty good post. I just stumbled upon your blog and wanted to say that I have really enjoyed reading your blog posts. Any way I'll be subscribing to your feed and I hope you post again soon.Web Designhttp://www.boundlesstech.netnoreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-69952758663253525292010-07-01T20:29:26.445-04:002010-07-01T20:29:26.445-04:00> why a certain figure ... is ...
> meaning...> why a certain figure ... is ... <br />> meaningful to ... some ecosystem<br /><br />Perhaps after break you could invite some of the ecologists who study the area to comment here, unless someone knows of places outside the journals this is discussed. Plenty studied, e.g. limit this to recent papers:<br />http://scholar.google.com/scholar?hl=en&q=polar+arctic+"ice-dependent"+ecosystem+primary+productionHank Robertshttps://www.blogger.com/profile/07521410755553979665noreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-15964045880068902002010-06-19T15:52:13.679-04:002010-06-19T15:52:13.679-04:00I dissed this in a throwaway comment on my blog, s...I dissed this in a throwaway comment on my blog, so maybe I owe you a longer version:<br /><br />You only combine probabilities with "and" if they are independent (you know this). But the fact that you think they shift year-by-year rather suggests they aren't.<br /><br />Also, your choice of normal distribution for the variability is trivially wrong in detail (because it gives a finite probability of less than zero ice). You can obviously patch that up, but the question of the distribution still matters. The chance of zero ice this year is zero, not a very small number. Ditto next year. Quite when it becomes non-zero I don't know.William M. Connolleyhttps://www.blogger.com/profile/05836299130680534926noreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-12745105319325991322010-06-15T20:56:08.107-04:002010-06-15T20:56:08.107-04:00Brian:
Certainly (as certain as I can be) we'l...Brian:<br />Certainly (as certain as I can be) we'll see a day of 0 ice cover before we see a solid month of 0 ice cover (and even longer before it is exactly a calendar month). <br /><br />The thing is, the data I am working with is for monthly average extents. So the predictions have to be about that quantity. My approach could also be used for daily minimum extent, by anybody who looked up that data set. (Or constructed it -- the daily data do exist at http://nsidc.org/ but you might have to compute the extents yourself.) But I'm preparing for that summer break of my later post.<br /><br />Daniel et alia: <br />I agree entirely that there are other interesting points to look for when we'll pass below. I'm a little leery of a figure like '10%', because that's suspiciously round and I don't think polar bears really know about 10%. This is another point, though, I'd like to get back to, along with the seasonal daily minima. <br /><br />If you or anyone knows of research which sets and explains why a certain figure (whether it's 10% or 12.35% or 2.34 million square km, ...) is particularly meaningful to the climate system or some ecosystem, please do let me know what the figure is, and, of course, the source.Robert Grumbinehttps://www.blogger.com/profile/10783453972811796911noreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-3413453496495360582010-06-15T19:37:05.433-04:002010-06-15T19:37:05.433-04:00Seems like your definition of zero ice - that it l...Seems like your definition of zero ice - that it lasts a solid calendar month - is much tougher than a 'zero ice at any one time during the year' that I've seen more often. The other definition could presumably happen much earlier.Brian Schmidthttp://backseatdriving.blogspot.com/noreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-31554462993925783542010-06-08T08:32:34.575-04:002010-06-08T08:32:34.575-04:00I really admire this, I mean it really looks inter...I really admire this, I mean it really looks interesting! Very nice research.Term Papershttp://www.ghostpapers.comnoreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-43478314568983132792010-06-07T06:13:03.705-04:002010-06-07T06:13:03.705-04:00I would like to add my voice to doing the same wor...I would like to add my voice to doing the same work with endpoint at 10% being 10% of 1979-2000 average (for example), and maybe also a run for 15%. <br />Reason are <br />1)-with 10 or 15 % of average, on a ecosystemic basis you can say the arctic is gone (or call it Aarctic as its something totally different, see recent book called "Eaarth"...)<br />2) Given the properties of the (inverted) logistic curve, with its long asymptote to 0, 10 or 15% might happen A LOT sooner than 0%, right? <br />Thanks for the exercice, very informative.Danielnoreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-3700162954175897832010-06-03T20:45:43.005-04:002010-06-03T20:45:43.005-04:00So, in your last post, you listed three estimation...So, in your last post, you listed three estimations: one based on statistical fit, one based on ensemble statistical fit, and one "Wu and Grumbine". Is this last one also a statistical fit, or is it something more complicated?<br /><br />If this is only statistical fit work, then despite a long history of people publishing papers based on such things (in some cases famous - the Hubbert curve, for example - and in some cases infamous - various skeptic attempts to show climate sensitivity must be low by making an equation of T = CS*LN(CO2) + ENSO correction or whatever), I would advise against trying to publish in a serious journal. As your ensemble fit shows, there are a wide range of possible ways to fit a given curve shape to the data... and if you allow other curve shapes, why, as the saying goes, you could fit an elephant... <br /><br />(but if Wu and Grumbine refers to something that is more complex than a straight logistic fit, then, go for it!)<br /><br />-MarcusAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-39624910809878758312010-06-03T20:13:58.376-04:002010-06-03T20:13:58.376-04:00Anon:
Any suggestions on where to submit it for pu...Anon:<br />Any suggestions on where to submit it for publication? <br /><br />All:<br />Good points. <br /><br />Partly, you illustrate why it is I specified sea ice. The fast ice is not, properly speaking, sea ice. It also isn't included in the remote sensing that, for instance, the NSIDC does (too close to land).<br /><br />Fast ice is its own interesting creature. I should take it up in its own post at some point.<br /><br />Greg:<br />I'm not sure about the shipping question. For the Northeast passage, you might be right. The Northwest passage, though, should be less easy. The standard circulation pushes ice towards the Canadian Archipelago. Since we have recently opened, if only briefly, the Northwest passage, I'd expect the openings to become longer and more frequent. But the chance of ice getting blown into the passage would make it a concern for a long time.Robert Grumbinehttps://www.blogger.com/profile/10783453972811796911noreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-80280127470766554782010-06-03T18:38:27.182-04:002010-06-03T18:38:27.182-04:00Like Peter, I think you have to allow a small, but...Like Peter, I think you have to allow a small, but non-zero, amount of ice to persist on a longer time-frame than the majority of the ice.<br /><br />Overall I think your time-frame is about right though. By 2030 the September ice, although still present, will be irrelevant to shipping.Greghttps://www.blogger.com/profile/07116646136992710754noreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-59356409904902596072010-06-03T14:28:49.286-04:002010-06-03T14:28:49.286-04:00Bookmarked. Peter, if by fast ice you mean multiye...Bookmarked. Peter, if by fast ice you mean multiyear ice attached to the shores, there has been a very large reduction of the attachement points in recent years, and most of it flows like any other sea ice. The shores of the islands get warm during summers and free the multiyear ice to go with the flow. Of course ice shelves from glaciers on land will be there for a long time though some are disappearing in Antarctica.jyyhnoreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-61983423137984092912010-06-03T07:22:11.878-04:002010-06-03T07:22:11.878-04:00Is zero the most appropriate endpoint? The fast i...Is zero the most appropriate endpoint? The fast ice around Northern Greenland and the Canadian Archipelago will presumably be more resistant to melting out completely, just like the Antarctic shelves. <br />Thus, loss is likely to slow dramatically once all the floating ice has gone - do you have a prediction for that?Peterhttps://www.blogger.com/profile/12559721137290332762noreply@blogger.comtag:blogger.com,1999:blog-5337555368793819627.post-10118686008654396792010-06-03T04:48:46.548-04:002010-06-03T04:48:46.548-04:00This is good stuff. You should quickly write a pap...This is good stuff. You should quickly write a paper on it.Anonymousnoreply@blogger.com