The accuracy of climate models is simply astonishing!
Assuming that greenhouse gas emissions rise rapidly in the future, the frequency of Arctic hurricanes could fall from an average of 36 per winter to about 17 by 2100, the model suggests. If emissions rise more slowly the number of hurricanes could fall to 23 per winter.
http://www.guardian.co.uk/environment/2010/sep/16/global-warming-arctic-hurricanes-oil
If I were betting, I’d put my money on 36 …..
Impressive stuff!
Does anyone have the first clue how many Arctic storms there will be this winter?
No.
And they don’t say they know how many there will be in 2100, either.
Does anyone here understand what the word “average” means?
Blatant misrepresentation.
As the quote clearly states–but your headline and snarky comment do not–this is not a forecast that there will be 17 storms in 2100.
It’s a projection of the average number of storms per winter by 2100 given certain conditions.
These are not the same thing, and you know they’re not. And the headline is just flat-out wrong.
The article itself is complete rubbish. What’s this “Global Warming could” crap ?
Besides meteorologists can’t even get this coming weekends weather correct, what chance 2100?
Indeed the fact it comes from the Grauniad is reason enough to ignore it.
But it’s not reason enough to misrepresent it.
The study does not say what this headline claims, and Steve knows it. There’s a word for this. I don’t often use it, but it has three letters. It begins with ‘L’ and ends with ‘E’.
Like I said, I am betting on 36 – same as the present.
Most people here understand the nature of Gaussian distributions. You completely missed the point of the article.
If you understood Gaussian distributions, you would also have specified the standard deviation, indicating the error bars on your prediction.
Yet again you are evading the point, which is that your headline is a misrepresentation of the forecast, which does not predict a particular number of arctic storms will happen in the year 2100. If you have any pretentions of being involved in a scientific, rather than merely a rhetorical, debate, you really need to stop continually misrepresenting your opponents position. If your science stood up to the test, you wouldn’t need to.
You too seem to have missed the point of the article. These people have no clue what the error of the model is when they make their predictions.
They should have said – “In 2100 we expect to see somewhere between 10 and 50 polar storms.”
That would be a proper use of an error bar. Most people here understand science, sarcasm and parody.
It doesn’t matter how you try to spin it, Steve. These people did not say what your page title and snarky comment say they said.
The page title in particular is just plain wrong. They did not forecast 17 storms for 2100. Your title says they did. What would you call that? What would you call it if Joe Romm did it?
An objective person would understand the utter folly of the prediction and laugh along.
That doesn’t justify misrepresenting their paper as having made a forecast of a particular number of storms occurring in a particular year. If you have identified a flaw in a paper, you devalue your criticism by suggesting they made a different even bigger flaw. IT just makes it look as if (a) you have an agenda and (b) making a mountain out of a molehill.
Now there is a reason why they didn’t say “In 2100 we expect to see somewhere between 10 and 50 polar storms.”, which is that the purpose of the paper was not to make a forecast of the number of storms we would see in 2100. The purpose of the paper was to demonstrate that a consequence of AGW was likely to be a change in the frequency of Arctic storms. To establish that, you compute the means and perform a test of statistical significance to see if it was explainable by chance. Which is EXACTLY what they did.
The fault is yours for not understanding the aim of the paper.
By the way, the interval was given in the paper, in figure 1, hich Steve would know if he had actually bothered to read it, looks like they expect between about 10 and 25 polar storms around 2100 assuming scenario A2 (and obviously different intervals for different scenarios, but they are given in the plot).
http://www.nature.com/nature/journal/v467/n7313/images/nature09388-f1.2.jpg
Ha. What an objective person would understand is that it isn’t a prediction.
They can claim to “expect” whatever they want, but the models themselves have shown no skill at these types of forecasts.
Exactly. The modelers claim is a blatant misrepresentation of the skill of their model, because they didn’t specify their 300% error bar.
Sorry, that is just your prejudice about climate models, nothing to do with science. Wriggle all you like, all you are doing is demonstrating that you are more interested in rhetoric and soundbites than you are in the science. TTFN
Did Zahn and von Storch predict that exactly 17 Arctic storms would occur in 2100? Please answer:
1. Yes
2. No
Thank you.
The article is rubbish. First there weren’t supposed to be any more winters (snow) in England. When that obviously didn’t happen, now the spin is that it may happen by 2100. Which is it? They were wrong before …the only thing different is that the prediction has now been moved back by 90 years. Who will be around in 2100 to say `crap’ or `I told you so’. If it doesn’t come to pass then my prediction is that by 2030 they will move it back to 2150. You can quote me on that. If I’m wrong in 2100 you can come and take my car…. honestly.
Steven,
I like the picture of the old main frames. Perhaps if the AGW crowd went back and plugged in some of them they would find all the old data. Not likely though. More likely it will be;`it is my data and you can’t see it, oops `I can’t find the data’, oops `it is to much work – I don’t have the time, trust me’, oops, `I didn’t know what a FOI was’, `it didn’t qualify under a FOI’, oops, `it is missing’, oops `I deleted it – I mean I will delete it if I have to give it to you’, oops, `I deleted it’, opps `it is missing’, oops, I just stepped in a big steaming pile of horse s$#*. Or the latest, oops, `It isn’t really official data and the guy who homogenized it is dead and we don’t know what methods he used and we don’t have his calculations’, …………. ooooooops. Trust me.
Hold it, I thought when it warmed up we get bigger and more hurricanes? Oh, not right again. We’ll, what they actually meant was `by 2100, trust me’.
ChrisD says:
October 13, 2010 at 12:29 pm
“And they don’t say they know how many there will be in 2100, either.”
—————————————————————————————
So, if you don’t know how many there will be in 2010, 2100 and presumably, every year in between, how can you offer a meaningful prediction about the average number of storms in 2100? You can’t.
Proponents of AGW poo-pooh these objections by saying that weather is chaotic, but it really isn’t – it operates within the laws of physics after all. The more exact truth is that weather is extremely complex – so much so that our presently poor level of understanding of these complexities makes it appear to be chaotic.
If scientists concentrated more on improving that understanding, and less on throwing out wild speculation about what may or may not happen in the future – which the MSM gleefully turn into horror stories designed to scare the **** out of everybody (it’s what they do) – then they might actually start getting somewhere.
Not holding my breath, though.
THE PAPER DOES NOT MAKE A PREDICTION OF THE NUMBER OF STORMS THIS YEAR, NEXT YEAR, 2100, 21001 OR ANY YEAR.
sorry about the shouting, but there are some who don’t seem to realise that the title of this article is a misrepresentation of the scientific claimn actually made in the paper. What they claim is that *according to the models* a consequence of global warming will be a reduction in the average number of such storms, and they give figures for the averages.
Now it is perfectly rational and scientific to reject the finding as is is based on climate models if you have identified a flaw in the models. What is not rational or scientific is to pretend that they had made a far more specific and claim than they actually did and show that to be unrealistic. They would probably agree with you, which is why they made NO SUCH CLAIM.
By producing numbers like 17 and 23, they are misrepresenting their accuracy. The can not claim accuracy to a single digit, because the error of their models is more than an order of magnitude higher than that.
Once again you demonstrate a complete failure to grasp the topic of discussion.
Steve, this is getting beyond a joke, the paper does not say anywhere (or even imply) that there will be exactly 17 storms in 2100. They say that the average number of storms is 17; if they meant there will be exactly 17 storms in 2100, why did they say an AVERAGE of 17 storms for an event that can only every happen once?
I understand the point you are trying to make. I agree, nobody can reasonably expect to predict the exact number of storms in 2100. The problem is that you have not yet grasped that the paper in question does not make any such prediction.
Yes, and furthermore, that number’s not even for 2100, it’s for the 30-year period 2070-2099. “17” isn’t for any particular year, it’s just the center of the error range for that 30-year period.
They don’t. The only misrepresenting that’s going on here is yours.
<So, if you don’t know how many there will be in 2010, 2100 and presumably, every year in between, how can you offer a meaningful prediction about the average number of storms in 2100? You can’t.
Please listen carefully: The authors DID NOT offer a prediction for the “average number of storms in 2100”. The blog author is Making Stuff Up.
What the authors actually said was that over the period 2070-2099, given a certain emissions scenario (IPCC scenario A2, to be specific), the model shows an average of 17 polar lows per season, with an error range of 11-25.
The post title here is flat wrong, and I believe intentionally so.
You and Dikran are doing a fantastic job of highlighting the problems with global warming “science.” People on your side of the fence fail to grasp the folly of these model predictions.
These people have absolutely no clue how many storms there will be in 2100, and to claim otherwise is ludicrous.
How many times do you need to be told, THE PAPER DOES NOT MAKE ANY PREDICTION OF THE NUMBER OF STORMS IN 2100.
There is no such claim. You are making it up.
You are just not ever going to admit that the paper does not make the prediction you say it does, are you?
Global warming could cut number of Arctic hurricanes, study finds
——-That is the headline.
Money Quote
“Zahn and his colleague Hans von Storch, of the Meteorological Institute at Hamburg University, used a global climate model to project the impact of three scenarios on temperature, humidity and other variables in 2100. They then fed this data into a regional model to assess how polar lows may respond.
Assuming that greenhouse gas emissions rise rapidly in the future, the frequency of Arctic hurricanes could fall from an average of 36 per winter to about 17 by 2100, the model suggests. If emissions rise more slowly the number of hurricanes could fall to 23 per winter.
Polar lows are less likely to form in the future because climate change will warm up the air in the north Atlantic faster than it warms up the ocean, reducing the thermal difference and reducing the risk of convection currents forming.”
The prediction of the future Polar weather is the point of the exercise.
So when you say
“How many times do you need to be told, THE PAPER DOES NOT MAKE ANY PREDICTION OF THE NUMBER OF STORMS IN 2100.”
then I have to ask what you thought the point of the study was ?
And how can Chris argue that they have not made numerical predictions fro the year 2100 when they most obviously have ?
Arguing that they can predict a 30 year average is no less ridiculous than than thinking that they can specify the number of storms in the particular year 2100, can’t you see that ?
Once again you guys split hairs like a razor and it is a sure sign that you don’t even believe this poppycock.
Saying that the AVERAGE number of storms will decrease to 17 does not mean that they are predicting there will be exactly 17 storms in 2100.
If I ask you to roll a pair of dice in 2100, then if I predict that the average result will be 7 does that mean I am predicting that you will will roll a total of exactly 7? No of course not, but that is exactly the way Steve is misrepresenting the paper.
Dikran I will quote (paraphrase) myself:
“Arguing that they can predict a 30 year average 70-90 years in the future, from their flawed assumptions and incomplete understanding of processes, is no less ridiculous than saying that they can specify the number of storms in the particular year 2100, can’t you see that ?”
It is neither my fault nor Steve’s that the Irony and sarcasm go over your head, and rather than argue the plain validity of the statement they did make, you’d prefer to argue and make demands concerning Steve’s Ironic/Sarcastic portrayal of it, as presented in the mushy and kowtowing press.
I especially liked the bit at the end speculating abut less snow on British motorways.
Sadly Steve also wrote:
“By producing numbers like 17 and 23, they are misrepresenting their accuracy. The can not claim accuracy to a single digit, because the error of their models is more than an order of magnitude higher than that.”
http://stevengoddard.wordpress.com/2010/10/13/forecast-for-the-year-2100-17-polar-storms/#comment-4612
which makes it clear that he (incorrectly) thought they were making a exact prediction.
No, it says that Hanz and Fritz are pretty cavalier about tossing out numbers .
Jimash, I already said it: Read the paper, not the newspaper. Nowhere in the paper do they come even close to making a prediction like this. They did not predict a specific number of storms in 2100 or any other year. It did not happen.
“Jimash, I already said it: Read the paper, not the newspaper. ”
It is behind a pay wall.
Do you think that the newspaper people made up the numbers ?
I doubt that.
And you quoted it yourself dude !
“What the authors actually said was that over the period 2070-2099, given a certain emissions scenario (IPCC scenario A2, to be specific), the model shows an average of 17 polar lows per season, with an error range of 11-25.
”
This kind of equivocation and hairsplitting on your side
is very unbecoming, and to think that it only takes
a simple if simplistic portrayal of the news that you support to twist you guys up like pretzels, it is really something to see.
Kind of makes you look like humorless drones.
I will repeat, the point here is that it is JUST as absurd to claim to have divined a numerical average for a 30 year period 60-90 years in the future, based solely on models, as it is to portray those findings as specific for one year.
At least Steven does it with a smile.
It’s not hairsplitting, dude. They didn’t say what Steve says. They didn’t predict a specific number of storms for any year. They projected that the number of storms per year over a 30-year period would average somewhere between 11 and 25. The post is wrong. Period. If pointing out that a post is wrong is “hairsplitting,” I’ll keep on doing it. Sorry.
I will point out that 17 is their number.
But there is no number they could offer with certainty,
and even this halfbaked average is offered with the caveat
“assuming that GHG emissions rise rapidly ” in the future.
They are assuming facts , not in evidence, feeding those into models, and offering up numbers.
You are not objecting to the study or its halfbaked conclusions.
You are merely nitpicking Steven’s approach to the title of the article as posted here.
Ok, I’ll play. Steven says “Forecast for the year 2100 17 polar storms”
He does NOT specify whether that number is meant here, to be an average or a specific number, for that year.
He does offer the quote that supports the use of the number and reports it as an average for that year.
If one were going to take the project seriously ( which I don’t)
then in fact one could and would pencil in the 17 storms for 2100 .
So the post is perfectly reasonable in light of the unreasonable certainty of the polar researchers.
That it is also a funny rip on the whole thing obviously eludes you. Too bad, that.
I will point out that 17 is their number. But there is no number they could offer with certainty,
There is no number they can offer with certainty, so they don’t. Steve just pretends they do.
offered with the caveat “assuming that GHG emissions rise rapidly ” in the future. They are assuming facts , not in evidence
Wrong. They can’t know what will happen to emissions, so they repeat the analysis with 5 different emissions scenarios and provide the results for all of them. What you are objecting to is caution and honesty.
Ok, I’ll play. Steven says “Forecast for the year 2100 17 polar storms” He does NOT specify whether that number is meant here, to be an average or a specific number, for that year
Oh, please. He says that the forecast is for 17 storms in 2100 and then goes on to say that “the accuracy of climate models is simply astonishing.” Now tell me honestly that, before reading the comments, you didn’t take that to mean that these guys predicted there would be 17 storms in 2100.
If one were going to take the project seriously … then in fact one could and would pencil in the 17 storms for 2100 .
If you can say this, you simply do not understand what the paper actually says–which is not surprising, considering the page title and comment. If I tell you that the average daily high temperature for next summer in East Mudflap, IA, will be between 85F and 95F, centered on 90F, am I predicting that it is going to be 90F on September 1? Really? Would this headline be justified:
“Scientist predicts high of 90F on September 1, 2011”?
Seriously? The fact is, you guys will defend anything Steve says, no matter how obviously wrong it is.
Have you even noticed the references to humor, sarcasm, irony etc . ?
Have you even noticed the references to humor, sarcasm, irony etc . ?>/i>
Yes, I’ve noticed some ex post facto attempts to back away from what he said without admitting that it was wrong by (a) claiming that the whole thing was supposed to be funny or (b) claiming that what it’s really all about is the generic “folly of model predictions”. Doesn’t excuse a blatantly wrong title. Saying that the paper predicted something it didn’t predict isn’t “sarcastic,” it’s just false (and there’s a stronger word I could use).
ChrisD sort of hyperventilated
“Yes, I’ve noticed some ex post facto attempts to back away from what he said without admitting that it was wrong by (a) claiming that the whole thing was supposed to be funny or (b) claiming that what it’s really all about is the generic “folly of model predictions”. Doesn’t excuse a blatantly wrong title. Saying that the paper predicted something it didn’t predict isn’t “sarcastic,” it’s just false (and there’s a stronger word I could use).
”
I will ask you again; Do you think someone just made up those numbers ? Steven didn’t. I assume and it’s only fair, that
the researchers gave those numbers to the reporters .
But the study itself, I do not see you commenting on.
It is of no value. The climate models are incomplete and wrongly biased.
They are making a prediction and I will say to you again that predicting an average for a thirty year period, 60-90 years in the future, as if it were a game of roulette is insufficient for any prediction at all, much less a 30 year average 90 years out, and not one iota less ridiculous than a claim of one year precision .
Finally it is obvious that you are reading this blog.
Are you really saying that you are completely blind to the fact that Steven writes with some sarcasm wit and humor regularly ?
Because I did not need to read the comments to be in on the joke.
I will ask you again; Do you think someone just made up those numbers ? Steven didn’t.
They aren’t made up, but they’re also not specific predictions. Not even the Guardian says they are. Only Steve did. If you don’t understand that these aren’t predictions, you’ve bought into what Steve claimed, whether you realize it or not.
You’re right Chris.
I have bought into what Steve says
And what he says is that the projection of averages
in polar storm activity whether for 1 year or 30 years, some 60-90 years in the future, through the use of computer models better suited to predicting the hands in a blackjack game, is a flawed process and NO number of specific prediction or average can possibly be accurate or represent some future reality except by random chance.
But you have maintained that climate is not random,
therefore the whole thing is an exercise in futility and hubris,
and so he mocks it as would I , had I the wit .
I quote:
“The study is claiming precision far beyond their accuracy.
They have no clue what the storm trend will be in 90 years. Every climate model storm prediction so far has been completely bass-ackwards.
If people don’t get sarcasm and parody, they don’t get it. ”
Don’t you get it ?
And what he says is that the projection of averages
in polar storm activity whether for 1 year or 30 years, some 60-90 years in the future, through the use of computer models better suited to predicting the hands in a blackjack game, is a flawed process
Unfortunately, he didn’t say any of that, did he? What he said was that they predicted 17 storms in 2100, which they did not do.
“The study is claiming precision far beyond their accuracy.
No. Wrong. It did not. It used standard statistical techniques and assigned a probability to a range of values from 11 to 25. Steve stuffed the claim of a precise prediction that they never made in the authors’ mouths. That’s the whole point. It’s what you don’t seem to be getting here.
Please listen carefully: The authors DID NOT offer a prediction for the “average number of storms in 2100?
“the frequency of Arctic hurricanes could fall from an average of 36 per winter to about 17 by 2100,”
That’s NOT a prediction?!!??
OK, fine. Whatever……
Read the paper, not the news article. And certainly not this blog post, because it is wrong. That number is the center of the error range for the average number of storms over a 30-year period.
In other words, what the paper actually says is, “Over the 30-year period from 2070 to 2099, the model puts the average number of polar lows per season at somewhere between 11 and 25, centered on 17.”
Compare that to “The authors predict 17 storms in 2100.” Do you seriously not see the difference?
The paper does not predict the number of storms in any year.
They key to the question is why they say it is an average rather than just a specific number. If they meant there will be 17 storms in 2100, they would have just said that, but they didn’t. They said the average number of storms. That is because climatologists have a tendency to think of probable outcomes in terms of ensembe forecasting, where you perform a large number of simulations of the physics of the climate, count the number of storms you see in each run, and then take an average. The real Earth is then assumed to be statistically exchangable with the model, so the number of storms you would expect to see in reality will be within the distribution of values you get from the ensemble. In the paper this is very clearly shown in FIgure 1. Assuming the A2 emissions scenario, the number you get is somewhere between 11 and 25, with a mean of 17. So if the models give a prediction at all, it is not that there will be exactly 17 storms, but that it is most likely to be 17, but it could plausibly be between 11 and 25.
The thing is climatologists aren’t remotely interested in the actual number of storms in a particular year, that is “weather”, they are interested in trends and long term averages, so they report the mean of 17 as that is what other climatologists will find interesting.
Looking at it another way, it is basically saying if you had a large number of parallel Earths with the same climate physics and forcings, but different initial conditions so they have different “weather” variation, then the mean number of storms will be 17, but that doesn’t mean that all of these parallel Earths will have 17, some will have more, some less. We can’t predict which of these parallel Earths will have 17, just as we can’t predict the number of storms there will be on any particular Earth (including this one).
That may sound a little bizarre, but that is the basis of Monte Carlo simulations, as used in climatology and a whole range of other applications in physics.
Monte Carlo simulations ,
can work when there are a limited number of known outcomes as in a game of chance, or a nuclear reaction.
Climate, not so much.
In other words, what the paper actually says is, “Over the 30-year period from 2070 to 2099, the model puts the average number of polar lows per season at somewhere between 11 and 25, centered on 17.”
STILL a prediction.
Keep digging (:-
It doesn’t actually make that prediction explicitly as it isn’t what the authors seem to be interested in, but that would be a valid interpretation of Figure 1.
They key point is that a prediction (based on assumptions stated in the paper) of a distribution of values for the number of storms with a range reflecting the variability is a reasonable thing to do. Predicting there will be precisely 17 storms in 2100 would be clearly unreasonable. However they made no such claim, Steve’s article is a misrepresentation of what they actually did claim, I hope a misrepresentation due to ignorance rather than willful disingenuousness.
The key point is that they are making unsupportable predictions and your personal bias makes you completely blind to the problem.
They have no idea if there will be more, less or the same number of storms in that time period, and publishing meaningless numbers like 17 and 23 for 90 years out is pure blather.
No, the key point is that your post title is dead wrong, and you won’t man up and admit it.
That is not what they published. Have you even read it?
Nice job quoting a mass media article about the paper. Why not quote the paper instead?
Is it a prediction that there will be exactly 17 storms in 2100, as the blog author claims?
Yes or no?
Do you honestly not get it, or are you being intentionally obtuse?
Gosh, you know that I could ask you exactly the same question, right? Do you honestly not get it, or are you being intentionally obtuse?
Does the paper predict exactly 17 storms in 2100, as your post title states, or not?
Yes or no?
I’ve already asked you this question. You didn’t respond the first time. Will you now?
Why don’t you start your own blog and post your straw man questions over there?
You obviously are not interested in having a serious discussion.
Steve, if it is true that ChrisD and I do not understand what you have been trying to say, then giving a direct answer to a question that would greatly clarify your position would be a very good step in helping us to see your point.
However, I have noticed that in the blogsphere people are generally reluctant to answer direct questions of that nature because it gives a hostage to fortune. If they give an unambiguous answer and it proves to be wrong, it leaves no room to wriggle out of it. Not answering looks (and indeed is) evasive, but at least you haven’t committed yourself.
I can see why you don’t want to answer the question.
I don’t know how many different ways I can say it. The study is claiming precision far beyond their accuracy.
They have no clue what the storm trend will be in 90 years. Every climate model storm prediction so far has been completely bass-ackwards.
If people don’t get sarcasm and parody, they don’t get it.
Steve, there is one rather obvious way of answering, just give a yes or no answer to ChrisD’s question. Far less effort that what you have just typed and likely to be more informatve.
Straw man? Oh, please. I’m asking if they actually said what you claim they did.
The fact that you won’t answer “Yes” pretty much answers the question, though.
There are 111 types of people in the world
Those who understand binary
Those who don’t
Those who can understand nuance
Those who can’t
Those who can’t see the forest for the trees
Those who can
Those who produce long term predictions from models which have no such skill
Simple yes or no?
No, you don’t get it. Yes, you don’t seem to be making any progress in that direction.
LOL, a few posts ago you wrote:
“You obviously are not interested in having a serious discussion.”
Very obviously neither are you! Way to go Steve, well done.
LOL.
You aren’t interested in the subject, you just want to make noise.
Enjoy yourself. I don’t censor specifically so that people of your ilk can air out.
Ain’t nothing “nuanced” about saying that they predicted 17 storms in 2100 when they did no such thing. That isn’t “nuance,” it’s just false.
You are cracking me up.
Again, this would be far more convincing if you could quote the actuial paper rather than a newspaper article. But you can’t.
But even what you quoted doesn’t predict 17 storms in 2100, does it? That pesky word “average” managed to creep in there, eh?
Do you really not understand how ridiculous this study is? It was obviously done by someone with very little familiarity of the state of GCMs
One last attempt at serious discussion. Is the basis of your argument that the spread of predictions is from about 10 to about 25 (for the A2 scenario), a range of 15, so the authors were wrong to quote the mean to the nearest whole number as the range is about 10 times larger (“an order of magnitude”) than the resolution of the natural numbers (i.e. 1)?
Seems a couple of you are arguing that the article Steve referenced didn’t say anything. If it doesn’t say anything, then it doesn’t mean anything. Why did Matthias Zahn bother to publish his study, and why did the Guardian quote him?
No. It said plenty. But it didn’t say what Steve claims it said.
And he knows it, too.
which is why he is refusing to give a direct unequivocal answer ChrisD’s perfectly reasonable question.
LOL – The data shows no change, but their projections do. It is ridiculous to describe the output of a computer model using the word “found.” It is a projection.
Can you imagine a weather forecaster saying “we found 10 inches of snow next Tuesday?”
Goddard sez: “Assuming that greenhouse gas emissions rise rapidly in the future, the frequency of Arctic hurricanes could fall from an average of 36 per winter to about 17 by 2100, the model suggests. If emissions rise more slowly the number of hurricanes could fall to 23 per winter.”
Zahn & Von Storch say: “All of the future scenarios experience a reduction of the number of polar lows per PLS during 2070–2099, resulting in averagenumbers
of 22.7, 16.9 and 17.6 in scenarios B1, A1B and A2, respectively (lefthand side of Fig. 2). According to a t-test based on their seasonal distributions (Fig. 1), all of these average numbers are statistically significantly different (at the 99.5% level) from the mean in C20.”
C20 being a control tested against 1960-1989 to test the strength of the model prediction.
Spot the difference.
The paper can be accessed here: http://www.indiaenvironmentportal.org.in/files/Decreased%20frequency%20of%20North%20Atlantic%20polar%20lows.pdf
to see the context.
It’s particularly interesting because the paper that our august host is attacking is one that is saying that, on this issue at least, climate change would be a good thing reducing threats to offshore activities such as shipping or oil and gas exploitation in the polar region.