My attention was directed to this post by Steven Goddard on WUWT. The theme is the rather silly claim that the arctic isn’t really warming — at least, not since 1958.
The gist of the argument is this:
Below is an overlay directly showing that 2009 temperatures (green) are similar to 1958 (red) and close to the mean.
The problem is, this graph (temperature north of latitude 80 degrees) shows no such thing. It shows the opposite. Of course there’s no analysis to go along with it; that would show how silly the claim is. And do I really have to mention how foolish it is to draw conclusions by comparing single years? But demolishing that post is not my purpose, that’s too easy. Instead I’d like to look at temperature north of latitude 80 degrees.
It’s easy to get monthly time series data for surface temperature north of 80 degrees here. Here’s the data:
It’s obvious that temperature in the far north has indeed increased throughout the time span covered by these data (since 1948). It’s even more obvious when we compute temperature anomaly (the difference between a given month’s temperature and the average for that month):
I’ve added two smoothed versions, a lowess smooth in red and a wavelet smooth in blue. We can get a better idea of the size of the warming by magnifying the y-axis, looking just at the smoothed curves:
It certainly has warmed since 1958, by 3 deg.C. For those who don’t trust fancy college-boy smoothing algorithms, here are simple 5-year averages:
Yeah … Warmer … A lot.
The graph of the temperature (not anomaly) time series indicates that most of the warming has occurred during wintertime. This is confirmed by using wavelet analysis to estimate the size of the annual cycle; here’s a plot of the smoothed semi-amplitude (half the difference between winter low and summer high):
Because the winter low has warmed so much more than the summer high, the amplitude of the annual cycle has decreased strongly. Again, if you don’t cotton to no fancy cityfied wavelet hogwash, we can just look at the data for individual months of the year. Here’s the average over time for January:
Here’s the same for July:
While the warming of January temperature has been a full 5 deg.C (according to linear regression), that for July has been less than half a degree.
How did Steven Goddard convince himself that his graph shows “2009 temperatures are similar to 1958 and close to the mean”? Perhaps more important, is there any limit to what Anthony Watts will post?
43 responses so far ↓
Deech56 // August 28, 2009 at 7:48 pm |
To answer your questions:
1. Make charts until he found one he liked?
2. Posts are limited to anything but mainstream climate science.
Timothy Chase // August 28, 2009 at 9:01 pm |
Tamino wrote:
I have some other questions…
How dishonest can posters like Goddard get before Watts begins to lose his audience? Or to put this another way, are there any limits to the degree of self-deception that members of that audience are willing to impose upon themselves so that they can continue to believe what they want to believe? And how much self-deception are they willing to endure so that they may continue place politics above reality, viewing the world in terms of and dividing it along the lines of Us vs. Them?
Not much to grab hold of when its all matters of degree.
David B. Benson // August 28, 2009 at 9:20 pm |
Timothy Chase // August 28, 2009 at 9:01 pm — It is the National Inquirer of climate blogs.
dhogaza // August 28, 2009 at 9:32 pm |
No, I don’t believe there are.
dhogaza // August 28, 2009 at 9:42 pm |
Oh, BTW, Stephen Goddard is no longer involved with WUWT because of the fiasco over the “does it snow CO2 snow in antarctica?” where even Watts became convinced (eventually) that standard physical chemistry textbooks were right, and Goddard wrong.
Apparently Goddard refused to admit to error, got pissed at Watts for saying that standard science is right, and walked off in a huff, never to be seen again.
KenM // August 28, 2009 at 9:44 pm |
Well yeah. But the sum of the anomalies adds up to 0. :)
Dan L. // August 28, 2009 at 9:55 pm |
>”…is there any limit to what Anthony Watts will post?”
Evidently not.
Anything to get the choir to sing, as they do loudly in the comments following this latest bit of twaddle.
MarkB // August 28, 2009 at 10:14 pm |
Nicely done…although it doesn’t take more than a very basic knowledge of statistics and an ounce of a desire to think critically to debunk such an argument, yet Watts and his merry band of zealots hoot, holler, and applaud any set of silly talking points that make them feel good. Can anyone explain to me why the Pielkes would want to associate themselves with this blogger who’s political agenda is painfully clear?
Gareth // August 28, 2009 at 10:27 pm |
The hilarity: in this post, Mr Watts explains why he’s accepting an ad from Energy and Environment…
Yeah, right.
None of the stuff at Watts (or Icecap or any of the others) has anything to do with reporting facts or genuine investigation. It’s about feeding material that supports a particular world view to an audience that needs constant confirmation in the face of reality. It’s a separate ecology, the ecology of denial, which draws its energy from sources well out of the mainstream. Like the communities of sea life round undersea volcanic vents, they’re life, Jim, but not as we know it.
(I may have a post developing the ecology theme at some point — it’s being mulled, like wine).
bouldersolar // August 28, 2009 at 10:36 pm |
One argument Goddard made was that it is generally believed that the 1940’s artic temps were warmer than today. Can you pull up a graph that refutes that claim?
[Response: Hard to tell whether he actually believes that or he was just lying.
http://tamino.files.wordpress.com/2007/01/gis5.JPG
http://tamino.files.wordpress.com/2007/10/arctic.jpg
http://tamino.files.wordpress.com/2006/11/iceland3.JPG?w=500
]
CapitalClimate // August 28, 2009 at 10:48 pm |
Wot the heck are those Goddard axes supposed to represent?
Also, what dhogaza (9:32) said.
[Response: The x-axis is day of the year, the y-axis is temperature in Kelvins.]
dhogaza // August 29, 2009 at 12:27 am |
The Pielkes appear to share the same political agenda …
caerbannog // August 29, 2009 at 3:59 am |
Timothy Chase // August 28, 2009 at 9:01 pm — It is the National Inquirer of climate blogs.
Actually, the Inquirer has, from time to time, been out in front on the news (using questionable tactics, admittedly). The OJ trial is a case in point.
A better comparison would be with the (now defunct) Weekly World News. And let’s hope WUWT suffers WWN’s fate.
Berbalang // August 29, 2009 at 12:03 pm |
“A better comparison would be with the (now defunct) Weekly World News. And let’s hope WUWT suffers WWN’s fate.”
I think they have already suffered that fate, web only.
http://weeklyworldnews.com/
yenna // August 29, 2009 at 9:58 am |
I have a question on the interesting plots Tamino came up with (I´ve some natural sciences education but no special knowledge in physics or climate). Tamino´s plots show warming north of 80 deg, mostly in wintertime. Is that not night-time there then? How does that work, such a warming in the ground air over ice in the polar night? Can anyone explain?
one would expect that this should be a rather astronomically controlled radiation situation, not? length of the night hardly changes, so if that spot loses less T nowadays to space in the same night, would this warming not have to be a rather direct image of the change in atmospheric composition? (but thats just guessing)
Mikkel Østergaard // August 29, 2009 at 10:25 am |
Nice job. Short and to the point.
Slioch // August 29, 2009 at 11:27 am |
“are there any limits to the degree of self-deception that members of that audience are willing to impose upon themselves so that they can continue to believe what they want to believe? ”
It is an example of “confirmation bias”, see:
http://en.wikipedia.org/wiki/Confirmation_bias
“Confirmation bias is an irrational tendency to search for, interpret or remember information in a way that confirms one’s preconceptions or working hypotheses.”
Any suggestions as to why Arctic July temperatures have not increased as much as January? Is it that summer temperatures are just above 0C and are therefore held back by the latent heat absorbed to melt the sea-ice? Is it possible to compute the heat content of the Arctic sea/atmosphere system over the years, and if so does this show a more similar increase in heat in July as in January?
Nick Barnes // August 29, 2009 at 11:53 am |
Mark 1 eyeball on Goddard’s graph says that 2009 is *much* warmer than 1958, and not just by one or two degrees either. Just look at what the green line is doing at the start of the year. Subtracting the mean curve would show this even more clearly, which is why he didn’t do it.
Hank Roberts // August 29, 2009 at 2:16 pm |
Nighttime warming, 6-month night.
http://pubs.acs.org/doi/abs/10.1021/es00045a001
Hank Roberts // August 29, 2009 at 2:19 pm |
hmmmm. Tamino, this ring any bells?
Science 7 April 1995:
Vol. 268. no. 5207, pp. 59 – 68
DOI: 10.1126/science.268.5207.59
Prev | Table of Contents | Next
Articles
The Seasons, Global Temperature, and Precession
David J. Thomson 1
1 AT&T Bell Labs, Murray Hill, NJ 07974, USA.
Analysis of instrumental temperature records beginning in 1659 shows that in much of the world the dominant frequency of the seasons is one cycle per anomalistic year (the time from perihelion to perihelion, 365.25964 days), not one cycle per tropical year (the time from equinox to equinox, 365.24220 days), and that the timing of the annual temperature cycle is controlled by perihelion. The assumption that the seasons were timed by the equinoxes has caused many statistical analyses of climate data to be badly biased. Coherence between changes in the amplitude of the annual cycle and those in the average temperature show that between 1854 and 1922 there were small temperature variations, probably of solar origin. Since 1922, the phase of the Northern Hemisphere coherence between these quantities switched from 0° to 180° and implies that solar variability cannot be the sole cause of the increasing temperature over the last century. About 1940, the phase patterns of the previous 300 years began to change and now appear to be changing at an unprecedented rate. The average change in phase is now coherent with the logarithm of atmospheric CO2 concentration.
Submitted on January 4, 1995
Accepted on March 3, 1995
Derecho64 // August 29, 2009 at 6:25 pm |
Thanks, Hank – I’ve been bashing neurons trying to remember that paper!
Hank Roberts // August 29, 2009 at 2:39 pm |
That 1995 article is being cited more recently:
http://www.sciencemag.org/cgi/crossref-forward-links/268/5207/59
Petro // August 29, 2009 at 3:42 pm |
It is rather obvious that in July the Arctic is still melting, which is enhanced by the extra energy in the atmosphere. In recent years more open water is present after June, so surface temperatures in July have started to rise more.
Additional CO2 absorb IR radiation from surface and reradiates it back, therefore more pronounced warming in winter/nighttime.
Hank Roberts // August 29, 2009 at 4:41 pm |
More along the same line:
http://www.agu.org/pubs/crossref/2009/2009GL040133.shtml
MarkB // August 29, 2009 at 4:50 pm |
“3) I think some WUWT contributors might find E&E a place to publish some of the works they have put forward here, in the harshest peer review environment of all; the online scrutiny of thousands.”
So one can conclude that publishing in E&E is no better than assertions made in the blogosphere.
tamino // August 29, 2009 at 4:56 pm |
The decision by E&E to advertise on Watts blog, as well as their offer of a discount subscription rate, is truly bizarre. Can you imagine Geophysical Research Letters buying an ad on my blog?
I can imagine two reasons: 1. They’ve concluded that their true core readership does not include the scientific community; 2. They’re in financial trouble.
dhogaza // August 29, 2009 at 6:08 pm |
This is for an online subscription, and personal online subscriptions appear to be new.
That was my first reaction … or hope, maybe. Certainly they’re seeking a new source of revenue.
Can you imagine *any* serious science journal advertising itself in the style of “The Journal that smashed the Hockey Stick”? Cult science, even if it were true.
I just can’t imagine an ad from a legitimate journal screaming something like “The Journal that Flogged the Phlogiston” or whatever …
dhogaza // August 29, 2009 at 6:09 pm |
Oh, it was never meant to … their editorial policy pretty much makes that clear, no?
Deep Climate // August 29, 2009 at 7:27 pm |
Regarding E&E advertising on WUWT:
E&E funding has always been a tightly-guarded secret. It could well be that major sponsors have withdrawn of late, especially now that contrarians have managed to get a few papers in the real peer-reviewed science journals. E&E has been so widely discredited that its effectiveness as a PR propaganda tool is surely blunted at this point.
It would also be interesting to know the financial arrangement between WUWT and E&E. Perhaps there is a commission arrangement, which will help support both enterprises.
Deep Climate // August 29, 2009 at 7:53 pm |
For what’s it worth, Plimer on the Arctic (and temperature record in general):
http://www.abc.net.au/lateline/content/2008/s2554129.htm
See also:
http://deepclimate.org/2009/08/24/ian-plimer-and-the-lie-that-wont-go-away/
Also see George Monbiot’s question 5:
http://www.guardian.co.uk/environment/georgemonbiot/2009/aug/05/climate-change-scepticism
Taking these statements together, Plimer appears to claim that, at the very least, Arctic sea ice extent is greater now than it was in the period before 1940. Of course, he has yet to give any sources to back up the claim.
Hank Roberts // August 29, 2009 at 8:01 pm |
Chuckle. But if you subscribe, you probably become one of their peer reviewers!
Timothy Chase // August 29, 2009 at 8:05 pm |
I had written:
Slioch responded:
That is a good explanation.
At first glance it certainly seems to fit the evidence. But are the experimenters sure that they have considered all other possible explanations for the behavior? Have they considered all the available evidence?
Even assuming they have and the general theory of confirmation bias so widely applies to human behavior, are you sure that it applies in this particular instance? What is the general criteria by which you decide that a given instance of human behavior involves an irrational tendency towards anything? It seems to me there may be some confirmation bias on your part involved.
*
Honestly, I think there is a great deal of truth to the “theory” (if I may call it that) of confirmation bias, but by itself I would regard it as a particularly unsatisfying psychological explanation of nearly everything and therefore nothing. For example, you could apply it whenever someone disagrees with you and appears to believe that they believe have good reasons for doing so.
But then it leaves unanswered a variety of questions. Some pertain to the identification of the phenomena. For example, Why do people look for confirmation? What is it that motivates them? And it might seem at first that the “irrational tendency to look for” confirmation is simply a matter of identification as well. But, as I have said, can one provide a general criteria for deciding whether a given tendency is rational or irrational?
Then there are questions that clearly involve some normative components. For example, should people look for disconfirmation as well as confirmation? And if they in fact should look for evidence that acts as disconfirmation as well as confirmation for their beliefs, why are some people better at it than others? At what point should they stop and regard a particular belief as in fact being something that they actually know?
As I have said, I believe there is a great deal to the theory of confirmation bias, but I would also regard it more as a starting point of sorts rather than as some sort of final, catch-all explanation. Given the very nature of this explanatory principle, I suspect the theoreticians who apply it would generally agree. Particularly when they step outside of one or another simple laboratory set up and set their gaze more broadly upon the human world.
jacob l // August 29, 2009 at 9:40 pm |
tamino you know that your link point a reanalysis?? otherwise great post..
I didn’t check if steve also used a reanalysis
thanks
Gareth // August 29, 2009 at 10:24 pm |
Further irony: under the E&E publisher’s note at microWatts, the WordPress “possibly related posts” gadget supplies this post, which opens:
How long before Mr Watts turns off that feature, I wonder?
David B. Benson // August 30, 2009 at 12:32 am |
Gareth // August 29, 2009 at 10:24 pm — I hadd to go out and buy a new irony meter; your post pegged my old one. :-)
Derecho64 // August 30, 2009 at 2:30 am |
One of the problems with WTFWT is that Anthony himself has no maturity. The entire site reminds me of a junior-high locker room – but instead of talking about girls (based on near-total ignorance), they talk about climate (but based on an even greater level of ignorance). The vast bulk of the posters there are imbeciles. That’s the audience Anthony prefers and allows, quite clearly.
Slioch // August 30, 2009 at 11:25 am |
Derecho64
“The vast bulk of the posters there are imbeciles. That’s the audience Anthony prefers and allows, quite clearly.”
One of the more deliciously imbecilic postings, following the Steve Goddard bit about “Arctic (Non) Warming” on WUWT, which Anthony allowed, was this one:
“TonyS (00:44:48) :
These very steep downwards slopes of the Vostok ice-core graph frighten me. I can now understand why people were afraid of a new ice-age coming.”
TonyS has apparently not realised that the time axis in the Vostok graph to which he refers runs backwards and that therefore the “very steep downwards slopes” that he fears are actually very steep upward slopes, representing the warming transition from glacial to interglacial conditions. But it gets worse, because, as far as I am aware, the rate of warming during those very steep upward Vostok slopes was actually considerably lower than the present rate of warming.
So, TonyS is frightened by a rate of cooling that didn’t happen but apparently completely at ease about a greater rate of warming that is happening but which he resolutely refuses to see, aided in his ignorance by WUWT.
You couldn’t make it up.
Slioch // August 30, 2009 at 1:16 pm |
The above case of TonyS helps to address questions concerning confirmation bias put by Timothy Chase (August 29th 8:05pm).
It seems to me that it is the discipline of being prepared to subject one’s beliefs to rigorous scrutiny in a scientifically literate manner that is the best defence against the tendency for confirmation bias that is present to a greater or lesser extent in all of us.
It is the lack of such scrutiny, and the cherry-picking, misrepresentation and misunderstanding (and sometimes invention) of evidence that characterises so much of the denialist world that, I believe, leads it open to accusations of confirmation bias.
The TonyS example appears to be of someone who has misunderstood the Vostok evidence, but finds confirmation of his beliefs from that misunderstanding. [The irony of his post being that a proper understanding of the evidence he highlighted points in precisely the opposite direction to that which he believed.]
Hank Roberts // August 30, 2009 at 4:18 pm |
It’s almost September; the Kimbot should be starting soon with the flood of postings about how the Arctic is cooling, nobody knows since when, the ice is growing, growing ….
Well, maybe October.
Kevin McKinney // August 31, 2009 at 12:45 pm |
What you said, Hank; but August’s slowing in ice extent decline has sparked claims of ice “recovery.”
Really, it is rather tiresome how predictable all this stuff gets. You’d think my typing would have improved more by now from all the practice rehashing the same facts.
B Buckner // August 31, 2009 at 3:14 pm |
Tamino – the reanalysis data apparently contains a polar temperature shift that may effect your analysis. See problem list on page linked in your post.
NCEP/NCAR Reanalysis Problems List
• Polar temperatures shift 1998-2004.
Different sea ice analyses were used in that period, and it appears in some cases those analyses disagreed with the R1 model as to whether a given point was land or ocean, especially in the Arctic. New analyses used starting in 2004 corrected the problem. So, skin temperatures and 2 meter air temperatures (at least) were significantly higher at some polar region points for the period. NCEP plans no data replacement.
Ray Ladbury // August 31, 2009 at 4:49 pm |
Kevin, Do yourself a favor and just file copies of all your replies to this nonsense. Then just cut and paste. It’s all these morons are worth.
Brian Schmidt // September 2, 2009 at 7:24 pm |
One more tiny cleanup point on Goddard – he says “2008 was notable in that Alaska glaciers started to increase in size.”
His proof, a newspaper article, doesn’t say that. Instead it says “One cool summer that leaves 20 feet of new snow still sitting atop glaciers come the start of the next winter is no big deal, Molnia said.”
Nothing in the article said the glaciers were growing.