Arguments
Recent Comments
Prev 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 Next
Comments 48451 to 48500:
-
ubrew12 at 11:34 AM on 22 February 2013Low emissions are no justification for Kansas scaling back renewables
There's a groundbreaking book in American politics, Thomas Franks 'Whats the Matter with Kansas?', that helps explain what is going on here, "where small farmers cast their votes for a Wall Street Order that will eventually push them off their land". Where former lobbiests for tobacco grower Phillip Morris joined hands with the Koch Brothers to create a 'Heartland' American Astroturf movement of 'Salt of the Earth' dupes.
The bizarre truth is that Heartland farmers in Kansas are only now in business thanks to 'Big Government'-backed farm insurance programs that tax the ordinary American to help them through their global-warming-induced drought. And, thanks to 'Heartland-values' propaganda, they then USE their solvent status to proselytize against those very 'Big Government' programs, and the Climate Change 'nonsense' they would otherwise seek to redress.
"We're not in Kansas anymore, Toto"
-
scaddenp at 11:25 AM on 22 February 2013Mars is warming
And also worth noting (as this and articles on other planets show) that real evidence for climate change on the other planets is somewhat lacking. See the intermediate versions.
-
Tom Curtis at 11:02 AM on 22 February 2013Mars is warming
Novemdecellist @33, there is a very simple reason to think that our motion through the galaxy has very little do do with current changes in climate. Specifically, it will take us 1360 years to travel just one light year at our current rate of travell. In the direction we are travelling, there are no nebulae to obscuring local stars, so we know the nearest nebula to be many light years away. Apparently the nearest nebula of any sort, L134, is around 300 light years away, and not in our direction of travel.
It follows from this that the radiation, and molecular density of the space through which we are travelling has not changed appreciably in hundreds of thousands of year; and will not change appreciably for hundreds for thousands more years except in the cases of nearby supernovas. And if it is not changing, it is not the cause of change.
-
Novemdecillist at 11:00 AM on 22 February 2013Mars is warming
We have NOT been measuring the other planets temperatures systematically since the 70's.... and I did not state the the majority of EARTH warming does originate with our mismanagement of the planet. I was only stating that IF other planets in our system might all be experiencing warming, then we should look at other possibilities be they "weird" or undiscovered!
-
scaddenp at 10:14 AM on 22 February 2013Mars is warming
Whether travels on the galactic path had anything to do with past climate change is harder to decide, but we can see clearly that it has very little to do with post-1970 climate change. Why? because we have good measurements of the climate determinants. The only "space" factor affecting climate is TSI, whether from change of angle, change of solar luminense or "space dust". This is accurately measured by satellites since 70's and is stable if not decreasing (see "its the sun" argument), and yet the earth warms. Why look to weird, unknowable, out-there sources of warming when there is a perfectly reasonable, physically plausible solution coming right out of the smoke stack? Or to put it another way, we have measurably increased the energy flux onto the surface of the earth with increased GHGs. What mechanism do you propose by which this would not cause warming?
-
Novemdecillist at 09:35 AM on 22 February 2013Mars is warming
Thinking 'outside the box' is a school not often visited by some. In all the discussions of planetary global warming (i.e. Earth, Mars Jupiter, Saturn, Pluto) very little attention is given to our solar system's travel above, below, and around the galactic plane on its cyclical route round the Galaxy. (approx 240k+ years). Conditions along the path are hard to predict since historic records are only now being established. Therefore any theories or calculations of planetary temperatures from volcanic, atmospheres, rings of dust, or human intereactions, etc. must include calculations of the density and position of materials encountered along our solar systems path and its interactions with the Sun's activities. These conditions make it extremely difficult to predict any astronomical algorithm's absolute forecasting the reason for Earth's warming trends. In other words, the best thought humans put forth probably has not reached a definite answer in totallity to the question, "Why is Earth warming?"
-
Roger D at 09:04 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
My "prediction" is 4.2million Km2, based on cheating, i.e. putting a straight-edge up to the computer screen, and eyeballing a vertical line through Gavin Crawleys prediction line in Fig. . The intersection for 2013 minimum is ~4.2 million Km2 (with a low of ~3.2 and a high of ~5.1 million Km2)
Given that Gavin's predictions have proven better than average, and also that there is an ongoing trend of less multi-year ice (oops, sorry Steve Goddard, I mean an ongoing trend of "increasing extent of Jan-Sept. ice growth") it doesn't make any sense to choose something like 5million Km2 as a likely ice extent, and then put an error bar on that.
-
Bob Wallace at 08:54 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Keith #43 - my observations and expectations match yours.
I have a small pond outside my living room window that freezes to 100% extent and melts to 0% multiple times per year. After being frozen over for days, it goes from almost frozen over to ice free in short hours.
The Arctic sea ice has gotten (comparatively) very thin. A few decades back the Russans set up their ice observatory stations on 20'/6m thick ice. Now they report that they have to hunt for 6'/2 meter ice. And they have to be prepared to move quickly if that ice begins to break apart.
I think that in the next few years we'll see a rapid melting of first year ice and that will clear the Fram and CA channels for the last remaining multi-year ice to be flushed out of the Arctic. I won't be surprised to go from significant extent to "Where's the ice?" in less than a month.
-
dana1981 at 08:40 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Note that the summer storm had a small impact on the 2012 Arctic sea ice minimum. There almost certainly would have been a record even without the storm.
-
Djon at 08:39 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Kevin,
Following your suggestion, when I look at figure 2 I see:- 1986 more or less matches your description, apparently fully recouping the loss that occurred in 1984 and 1985.
- 1992 followed a year of recovery not loss, and 1992 alone didn't have a recovery as large as the preceding two loss years, 1989 and 1990, put together
- 1997 was a year of loss rather than recovery. You presumably meant 1996, not that it recouped two immediately preceding years of loss either. If I were in a generous mood, I guess I'd allow that 1996 recouped the two years of loss that occurred in 1989 and 1990, though I don't see how that helps your argument that the losses of last year should be expected to be recouped this year.
- 2001, like 1992, followed a year of recovery. Unlike 1992, it looks as though the 2001 recovery was about the same size as the preceding two years of loss put together.
- 2006 didn't recoup the loss in 2005, let alone the loss of 2004 and 2005 combined
You're failing to convince me that any careful analysis underlies your beliefs about sea ice.
-
cRR Kampen at 08:30 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
#43, Keith - that is my point exactly. This is why catastrophe theory applies, and not gaussian thinking. Even my hypothesis on the evolution of arctic sea ice these years is based on the same sort of observations re canal, pond and lake ice.
That hypothesis dates from 2005 and predicted first arctic ice free summer at +10 to +15 years, which means I can seriously fail this year or next.
The prediction was based on the miracle Britain-sized hole appearing in the north of the East Siberian Sea in August '05, in an area where for centuries to many millenia a massive 2-4 metres ice, usually affixed to the Siberian coast including during the minimum, was the norm.
I am actually quite aghast at the amount of underestimation of this process even in (semi-)professional circles.
-
cRR Kampen at 08:20 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
#40 Tom, I am aware of that. The question is: what do we mean by 'the mean' here? Obviously we do not mean the 1979-2000 ice extent/area/volume mean for this number has (literally...) no meaning anymore. Therefore expectations based on this regression concept for 2013 running a higher minimum than 2012 are rather gratuit.
As for the discussion on error bars, as of this year one side of such a bar is very easy to place: it is on the zero of the vertical axis.
-
Kevin8233 at 07:16 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Dikran,
Some of the other reasons,
1. 2012 was warm
2. 2011 was warm
3. etc
4. Put several years together in a row that were warm and the logical deterioration of the thickness of the ice occurs. Look at the trend for ice volume. All of these things point to a decreasing ice extent, my point was that the storm was a major cause (not saying the only) for it to be lower than expected. Looking at the "professionals'" predictions, they were all above the actual, so obviously they thought the actual was exceptional as well.
Looking at fig2, you can see several recovery years that do recoup a couple of years losses - these do not portend a general recovery of the overall trend - nor am I implying that - look at years 1986,1992,1997,2001,and2006.
-
2013 Arctic Sea Ice Extent Prediction
Ok, I'll throw a few cents into the ring, based on a simple polynomial extrapolation (R² = 0.8194).
- 4.04 Mk^2 +/-0.8 (2σ) extent
-
scaddenp at 06:19 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
I'll go with 4.1 +/- 0.9, error bars based on variability, number on crude extrapolation.
Got to ask Kevin - if it goes below 4, what conclusions are going to draw? What sort of data is required to change your mind? At the moment, I would say 5 is "Math You Do as a "skeptic' to Make Yourself Feel Better", and as likely to be as accurate as Rove.
-
Djon at 06:12 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
If you wanted to say my error band was nonsensical, just say so. I think your unwillingness to have your understanding of what is likely to happen to sea ice extent in the next year judged based on your error band is a much more convincing demonstration that your error band is nonsensical than my simply saying so. Just because you don't like being mocked for writing nonsense isn't a good reason to retire mockery as a debate technique. -
keithpickering at 05:12 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Personally I'm skeptical of all statistical methods to predict SIA. I say this based on my personal observation over decades of the way ice melts on the lake behind my house. The area of the ice is extremely non-linear during ice melt. During the winter, ice covers 100% of the lake. During spring, as the weather warms, the ice still covers 100%, because it's melting mostly from the bottom up. At some point the edge gives way to open water, and the covered area is 99% to 98%. And then, in about a day, it goes from 97% to zero, as the ice is no longer thick enough to withstand the wind.
We're seeing basically the same fall-off-the-cliff behavior in the Arctic, starting in the Kara Sea and working east. Last year it made it all the way arount to the Beaufort, and it will again this year too (although at even higher latitudes).
If you want to know when summer ice will hit zero, ignore extent and look at the volume. Ice melts from the bottom up.
-
Dikran Marsupial at 05:09 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Kevin, well quite, which is one of the reasons that your very confident prediction of a record September recovery seems somewhat peculiar.
By the way, in this post, you ducked the question posed by Phillipe here. He asked you to list the "number of factors" [note plural] involved in last year's record other than the storm that had already been mentioned. Pointing out that "that [the storm] was "one" of the other reasons" suggests you either didn't read his post with sufficient assiduity, or that you were being deliberately evasive. Please could you answer the question, so that we can understand your reasons for your prediction. -
Kevin8233 at 04:59 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Dikran (35):
That is what I was trying to say, it is easy to set a re-freeze record when so much melted previously. The re-freeze record is not indicative of a new trend, as you point out, the new max is still less than the previous max. I was not trying to state otherwise.
-
Tom Dayton at 04:57 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
cRR Kampen, there is no "physics" behind regression to the mean. Regression to the mean is a human perceptual/judgment of the mathematical fact that (by definition of the mean) observations are more likely to be closer to the mean than farther from it. When observations are independent of each other, an observation happening to be far from the mean does not change the probability of the next observation most likely being closer to the mean than farther from it. The human perception/judgment of those two observations is that the second observation moved toward the mean (regressed toward the mean).
-
Kevin8233 at 04:56 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Djon,
I did not expand my error bars. the +/- 4.5 is 5 times Tamino's error band. Since you want to place a consequence to my being wrong, I suggested using Tamino's error band. Maybe you would prefer to use his band without multiplication? (5.0 +/-.9) If you wanted to say my error band was nonsensical, just say so.
-
Djon at 03:51 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Kevin,
Do I have to have made a prediction for your prediction to be wrong? In school, did you complain to teachers who gave you a bad grade on a test because I hadn't taken the same test? Maybe, unlike you, I'm not overconfident enough to think I can predict the sea ice minimum to within +/- 100,000 square kilometers seven months in advance. Although "overconfident" doesn't seem like the right word since you don't seem willing to stand by your prediction, judging by your apparent desire to expand your error bars from +/-0.1M to +/-4.5M. -
Dikran Marsupial at 03:33 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Kevin, I'm sure Tamino would concede that his prediction was falsified if the observations were outside his stated error bars. There is a good reason why Taminos error bars and mine are much broader than yours, which is because that is the basic scale of inter annual variability around the trend.
I suspect Djon was being lenient as the error bars on your prediction are clearly unrealistic, unless there is something fundamental that you know about Arctic sea ice that the rest of us don't know (including the scientists who work on this for a living).
-
Kevin8233 at 03:26 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Djon,
How about 5 times Tamino's error band. That would make it 0.5 to 9.5 million Km^2.
Where is your prediction?
-
Dikran Marsupial at 03:22 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Kevin, see figure 4, there was indeed a record gain in Arctic sea ice this year, but the January sea ice extent was still lower than last year nevertheless, so the trend in winter extent is still downwards.
The fact that the record gain is very likely to be simply due to the record loss last year is exactly why Goddard and Watts are wrong, it is exactly what would be expected, and doesn't signal any kind of recovery of sea ice, because the extent itself is still on the decline even with a record gain.
-
Djon at 03:14 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
@Kevin
So you predict the NSIDC September average northern hemisphere sea ice extent in 2013 will be 5.0+/- 0.1 million square kilometers, right? For future reference, how much lower than that do you think the actual value needs to be to prove beyond reasonable doubt that you are utterly wrongheaded about climate science and that everything you say about it ought to be ignored? Would 4.5 be low enough? Letting you miss by anything less than 5 times your estimated error without considering it proof positive that you're suffering from the Dunning-Kruger effect seems pretty generous to me but maybe you're more forgiving of your own errors than that.
-
Kevin8233 at 03:12 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Dikran:
I would however point out that to get back to five million km^2 would require around the largest annual increase in sea ice seen in the satelite record, which seems rather unlikely as the trend in winter sea ice extent/volume are still downwards.
I believe we just had exactly that. And no, I don't believe that a record freeze means the downward trend is over. Is was a record freeze becauseit followed a record melt.
-
Kevin8233 at 03:07 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
John Hartz:
Re your forecast that 2013 will be a recovery year for arctic sea ice and for future reference, do you like your crow baked or broiled?
I don't seem to be able to find your prediction. Is it here?
Phillippe:
It was clear before the storm that the minimum was going to be extremely low, the storm simply made it impossible to not beat the 2007 record
Agreed, that was "one" of the other reasons.
-
stonepig at 03:06 AM on 22 February 2013In Wall Street Journal op-ed, Bjorn Lomborg urges delay with misleading stats
Somewhere I read reports written by insurance companies regarding the future of global climate changes and that industry. It's about the cost of replacement, not what it is pitted against.
-
TerryM at 02:35 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Last year the Caa opened up and swallowed the MYI that previously had found a safe haven north of there. This year we could have a repeat of this new avenue for advection and melt or this winter's MYI having slid down from above Ellesmere Island may plug the labrinth of channels.
My WAG is that the Caa opens again & that SIA, SIE & volume will all fall below this years minimum. If the Caa opens, Arctic ice will find that the only place of refuge is above Northern Greenland, ballancing between advection through Nares Strait or advection through Fram Strait.
While it's possible that the remaining ice can evade both traps in any one year, at some point an unfavorable wind or current will ship the bulk of it south. I 'd consider lingering year round ice in 2033 much less likely than an ice free period in 2013.
Considering the probability that we'll go into the coming melt season with less volume than last, and that we'll probably have an open Caa like last year, I'd be surprised if any of the metrics were as high as 2012.
Terry
-
John Hartz at 02:29 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
@Kevin #26
Re your forecast that 2013 will be a recovery year for arctic sea ice and for future reference, do you like your crow baked or broiled?
-
Philippe Chantreau at 02:22 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Kevin, would you care to list the "number of factors" involved in last year's record?
You mentioned only that one storm. It was clear before the storm that the minimum was going to be extremely low, the storm simply made it impossible to not beat the 2007 record. Considering that this record was not beaten but shattered, I doubt that the storm was that much of a factor. That was also the opinion expressed by Zhang and colleagues from UW.
Overall the conditions over much of the season were "a mixed bag" (NSIDC's words). In contrast, 2007 had a perfect combination of melting conditions.
http://nsidc.org/arcticseaicenews/2012/08/a-summer-storm-in-the-arctic/
-
cRR Kampen at 02:07 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
#27 - "The reason why there are "recoveries" in the observations is essentially "regression to the mean", after an extreme year, you will generally get a less extreme one."
What physics is behind this?
There is a severe circularity in this reasoning involving the definition of 'extreme year'. Apart, of course, from the meaning of 'mean' in cases of accelerating trends on dynamical systems.
-
Dikran Marsupial at 02:01 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Kevin making testable predictions is what science is all about. However, your two arguments both seem to me to be saying more or less the same thing twice. The reason why there are "recoveries" in the observations is essentially "regression to the mean", after an extreme year, you will generally get a less extreme one.
I would however point out that to get back to five million km^2 would require around the largest annual increase in sea ice seen in the satelite record, which seems rather unlikely as the trend in winter sea ice extent/volume are still downwards.
-
Kevin8233 at 01:43 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
yes, I did mean in millions. As to why, which is important.
2012 was very low for a number of reasons, one of which was the storm which dispersed ice. Doubt that will happen again like that this year. As I indicated, I believe it will be a recovery year, which happens with some regularity. Couple the two tegether and there is some pressure to be dramatically higher than last year.
-
Tristan at 01:42 AM on 22 February 2013Reconciling Two New Cloud Feedback Papers
The thing I find difficult to match up in my head is the idea that (judging by the likely ranges) the TCR can be as much as 2/3rds of the ECS.
A TCR of 2 and an ECS of 3 implies that 66% of the total warming associated with doubling GHG concentrations would have been already realised by the time the GHG concentrations had doubled.
It doesn't seem plausible.
-
meher engineer at 01:40 AM on 22 February 2013In Wall Street Journal op-ed, Bjorn Lomborg urges delay with misleading stats
what is Accumulated Cyclone Energy?
-
cRR Kampen at 01:08 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
#24, I might acknowledge your reproach of 'evasion' after you have checked my arguments for disposing with the concept of 'mean' in this case, but not before.
Interestingy 'The Black Swan' is much a report dealing with this kind of 'argument' leveled against Taleb all the time by people who do not seem to understand the concept involved.
Your (and Tamino's et al's) statistical approaches have fared pretty well, but we are increasingly looking at a Turkey Syndrome here -> http://www.attaincapital.com/img/newsletters/20111121_Fig1.gif
-
Reconciling Two New Cloud Feedback Papers
Tristan - Quite correct on the three sensitivities; clearly I needed more coffee when I wrote that.
As to the TCS range, the IPCC AR4 discussion on sensitivities states that:
Equlibrium climate sensitivity is likely to be in the range 2°C to 4.5°C with a most likely value of about 3°C, based upon multiple observational and modelling constraints. It is very unlikely to be less than 1.5°C.
The transient climate response is better constrained than the equilibrium climate sensitivity. It is very likely larger than 1°C and very unlikely greater than 3°C.
ESS as Hansen defined includes ice sheet changes, feedbacks including release of GHG's (such as ocean CO2, permafrost methane, etc). I find it (IMO) a bit more clear to group the different sensitivities by what feedbacks occur over that time period, rather than just the timescale itself.
-
Dikran Marsupial at 00:47 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
cRR Kampen You made an clear error in asserting that regression to the mean applies only to Gaussian situations, but when I explained why that is incorrect, your response was to argue instead that in some situations the mean is irrelevant. While this is technically true, it is also evasion, and I have learned over the years there is little to be gained discussing science where such errors cannot simply be acknowledged.
Perhaps the easiest way to decide would be for you to provide a numerical prediction for NSIDC September extent, with a credible/confidence interval and we can see how our predictions fare.
My statistical approach has fared pretty well for the last couple of years, I see no reason to expect a spectacular recovery suggested by Kevin's prediction, but at the same time I rather doubt that Arctic sea ice will vanish as quickly as you suggest either.
-
cRR Kampen at 00:34 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
[DM] "... new ice forms accross much of the Arctic during winter, and this is steadily pushed throughout the year towards the Canadian archipelago and northern Greenland throughout the year. Even when the Arctic is substantially ice free during the summer, this "stubborn ice" is likely to continue to be sufficiently replenished in the winter to last all summer, even when there is regular shipping across the Arctic."
This is simply not true: http://i1340.photobucket.com/albums/o728/OlTom67/cryosat_zpsf2aa8e10.gif (borrowed from a comment under http://neven1.typepad.com/blog/2013/02/cryosat-2-reveals-major-arctic-sea-ice-loss.html ).
Why would this trend reverse - especially as the albedo feedback kicks in ever more?
[DM] ]"This is simply incorrect, regression to the mean ocurrs in non-Gaussian situations as well..."
What 'mean'?
There are situations in which 'mean' is meaningless. Cf catastrophe theory, http://en.wikipedia.org/wiki/Catastrophe_theory .
Many examples are given in 'The Black Swan' by Nassim Taleb, http://en.wikipedia.org/wiki/Black_swan_theory .
What does 'mean' mean if you take the mean daily financial volume on Wall Street while knowing that over past half century only ten loose days account for half the total volume of the entire half century and realizing these are the days with true meaning?What is the mean melting rate of an icicle on the roof - when it includes its sudden moment of dropping off?
[DM] "Also, just looking at the data suggests regression to the mean is just what we should expect, extremes in the observations are generally followed by less extreme observations the next year. Every record low in the data set is followed by a higher extent in the following year. Why should 2012/3 be any different?"
Again, what 'mean'? To answer your question directly, the chance that 2013 will be different is considerable and actually increases every year - momentarily that is - because the drop of the 'mean' which is really a point value accelerates. Guassian reasoning fails here (and actually forces me to write gibberish like 'a mean being a point value').
Another reason to suspect relatively considerable chance of a new record in 2013 is simply the fact the total volume of ice in March 2013 will be smaller (again) than it was in March 2013. There is simply less ice to melt. And more heat te melt it, too.
[DM ]"Note that the minimum in 2012 came very close to being inconsistent with last years projection, being only just within the 95% credible interval, which suggests that last years minimum was a very extreme event judging by the long term trend."
No it was not, it was spot on considering the long term trend, which exhibits accelerating ice loss.
[DM] "Note "regression to the mean" doesn't mean that the minimum extent this year can't be lower than in 2012, just that "all things being equal" we would expect it not to be lower."
But not all things are being equal. This year the ocean-atmosphere system contains more heat than it did last year. This given, plus the fact total ice is less now than it was a year ago can only lead to the expectation next minimum will be lower. Only thing to save this is sheer luck e.g. a cool summer with benevolent stratus and winds.
Kind regards/cRR -
Dikran Marsupial at 00:25 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Kevin, fair point, everybody should give a credible/confidence interval on predictions, rather than a point estimate. Note I actually did give a credible interval, so it isn't absolutely correct to say "nobody else has".
5 km^2 +/- .1 is a very brave prediction indeed (I presume you meant millions of km^2, I don't think anyone is quite that pessimistic! ;o), given the inter-annual variability is about ten times larger than this.
-
Kevin8233 at 00:08 AM on 22 February 20132013 Arctic Sea Ice Extent Prediction
Kevin would you like to add error bars to your prediction, so that it can be unequivocally falsifiable?
No one else has, but sure, +/- .1
-
Dikran Marsupial at 23:56 PM on 21 February 20132013 Arctic Sea Ice Extent Prediction
Kevin would you like to add error bars to your prediction, so that it can be unequivocally falsifiable?
-
Dikran Marsupial at 23:55 PM on 21 February 20132013 Arctic Sea Ice Extent Prediction
cRR Kampen writes From what is that ice being replenished? There is nothing to replenish it!".
I am not an expert in the physics of Arctic ice, but my understanding is that new ice forms accross much of the Arctic during winter, and this is steadily pushed throughout the year towards the Canadian archipelago and northern Greenland throughout the year. Even when the Arctic is substantially ice free during the summer, this "stubborn ice" is likely to continue to be sufficiently replenished in the winter to last all summer, even when there is regular shipping across the Arctic.
"This is no principle in statistics, this is only a principle belonging to gaussian distribution statistics and these are NOT applicable here."
This is simply incorrect, regression to the mean ocurrs in non-Gaussian situations as well, although the motivation for regression to the mean is most easily justified by considering what causes a random variable to exhibit a Gaussian distribution (summation of a number of component random variables). If you have a small number of random variables, the sum may not be sufficently Gaussian, but it will still mean that extreme values will be the result of extreme values of the components, and hence you will get regression to the mean.
However, that is a moot point, as Gaussian statistics are appropriate. Both the model that I use, and I strongly suspect that used by Tamino, assumes the errors of the model are normally distributed, and indeed that looks reasonable from analysis of the residuals.
Also, just looking at the data suggests regression to the mean is just what we should expect, extremes in the observations are generally followed by less extreme observations the next year. Every record low in the data set is followed by a higher extent in the following year. Why should 2012/3 be any different?
Note that the minimum in 2012 came very close to being inconsistent with last years projection, being only just within the 95% credible interval, which suggests that last years minimum was a very extreme event judging by the long term trend.
Note "regression to the mean" doesn't mean that the minimum extent this year can't be lower than in 2012, just that "all things being equal" we would expect it not to be lower.
I agree with Walt Meier on this one, the definition of an ice free Arctic is a bit tricky. It will be functionally ice free (i.e. ships can go more or less where they like, significant decrease in summer albedo) in a matter of decades, but it won't be completely ice free for quite a long time afterwards. I am sure skeptics will be claiming that the IPCCs predictions were wrong on these grounds long after shipping regularly passes over the north pole!
-
Kevin8233 at 23:46 PM on 21 February 20132013 Arctic Sea Ice Extent Prediction
5.0 km^2
Recovery year.
-
CBDunkerson at 23:21 PM on 21 February 20132013 Arctic Sea Ice Extent Prediction
The arguments for a 'slow down' in ice loss which I have most frequently seen are;
- Once all of the thin ice melts there will be a core of thicker multi-year ice which just hangs on year after year.
- Most of the volume loss is due to export of thick ice. Once all of that thick ice is gone the volume loss will slow down.
The fact that these arguments contradict each other indicates that at least one of them must fail. Indeed, it seems clear that both are incorrect. The past few years we have seen huge expanses of thick ice just wither away in days once the thin ice barrier between them and open water is gone... that demonstrates both that this thick ice will not hold out for years on end AND that much of the loss has nothing to do with export.
That said, land ice which extends out over the ocean or breaks off into it can be MUCH thicker than ice formed from frozen sea water. That kind of 'former land ice' will remain around for decades after all of the 'sea water ice' is gone. So there WILL be a 'leveling off' at above zero arctic sea ice... it will just be a very small amount above zero.
-
cRR Kampen at 23:11 PM on 21 February 20132013 Arctic Sea Ice Extent Prediction
“The predominant ice circulation pushes ice toward those coasts resulting in thick ice that tends to get replenished.” (Walt Meier)
From what is that ice being replenished? There is nothing to replenish it!
Reality is the ice on those coasts is also getting thinner, while the last shreds of multiyear ice might – just might – survive for another half year.
“There is a principle in statistics known as "regression toward the mean," which is the phenomenon that if an extreme value of a variable is observed, the next measurement will generally be less extreme, i.e. we should not expect to observe record lows in consecutive years.”
This is no principle in statistics, this is only a principle belonging to gaussian distribution statistics and these are NOT applicable here.
“ This is because when extremes are reached and records are broken, a number of different variables generally have to align in the same direction to make this happen. ”
This is not true. Sometimes extremes are reached and records are broken because one single strengthening driver pushes the system that way.
That is the case here (the driver is known as ‘global warming’ and never forget this pertains primarily to the oceans!).
Last year’s extreme Arctic ice low was realized during a summer that was in no way remarkable. There was no ‘alignment of different variables’ as was the case in 2007. Moreover, the year of greatest volume loss, 2010, was no remarkable summer either. Those of us focusing on the Arctic Summer Storm of 2012 should realize first that the ice was already far below record levels before that storm moved in, and secondly that 2010 featured no such storm.
We can realistically start betting on this or one of the coming couple of years for the first ice free Arctic summer. Forget 2034. Really.
/cRR
-
curiousd at 20:43 PM on 21 February 2013Introducing the History of Climate Change Science
I purchased the text by Farmer and Cook from Amazon. It will be a great "sit down and read" for me. I gather from the threads by rockytom that there are a wide variety of types of courses by which colleges teach climate science in some form. In the "101" level course we teach in one semester for non science majors, Farmer and Cook is too dense to be the main student text book, but it would be ideal as a reference for the professor or as an assigned reference for student in - depth study. I am holding a workshop for outstanding high school physics teachers in March, a workshop that will introduce climate science as enrichment. As part of the workshop I will recommend Farmer and Cook as an outstanding reference text.
-
Daniel Bailey at 15:27 PM on 21 February 20132013 Arctic Sea Ice Extent Prediction
Remember that the bottom-melt contribution to Arctic sea ice mass losses has been increasing over the past several decades as the heat sequestered in the oceans continue to manifest itself in the Arctic. At 50%+ now, the melt contribution from the oceans has become the full equal of the sun itself. And all the cloudy days in the world cannot stop that.
Prev 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 Next
