Christopher Booker’s Telegraph column used Willis Eschenbach’s recent Open Letter to Nature as the basis for the Sunday column:
Booker writes:
On Friday came the fullest and most expert dissection of the Nature paper so far, published on the Watts Up With That website by Willis Eschenbach, a very experienced computer modeller. His findings are devastating. After detailed analysis of the study’s multiple flaws, he sums up by accusing Nature of “trying to pass off the end-result of a long daisy-chain of specifically selected, untested, unverified, un-investigated computer models as valid, falsifiable, peer-reviewed science”.
Read Booker’s column here
Read Willis’ essay here:
http://wattsupwiththat.com/2011/02/24/nature-magazines-folie-a-deux-part-deux
See also:
The code of Nature: making authors part with their programs
@kadaka:
Of course you realize, this means war.
-Bugs Bunny
——————–
I believe you have mis-attributed the quote. It was spoken by Daffy Duck:
Of corth you realith, thith meanth war
@Sean:
Bugs said it after his ears were tied to a tree branch by the famous tenor.
(sorry, don’t know how to embed youtube clips).
Chris Wright says:
February 27, 2011 at 3:23 am
Of course, we already knew that this study published in Nature is junk science …
Now there is a thought, how about another fact page for WUWT — a list sorted by publication of junk science articles.
A points system would be good:
Paper withdrawn: 100 points.
Authors refuse access to data/code: 10 points.
Demonstrably crap: 5 points.
Just plain junk: 1 point
Lets see which publications rank top of the “Publishers of junk science” list if we go back … oh … say 5 years?
Jose Suro says:
February 27, 2011 at 8:42 am
“RMS is playing with our money and the insurance companies love it.”
Jose, I hope you notified the author of the Herald Tribune column about this forum. Very interesting information. I will watch for more articles from the Herald Tribune. I live just up the road from Sarasota. Thanks so much for the info.
Theo
kbray in california says:
February 27, 2011 at 12:18 pm
It shows an increase in temperature from 5 sources, “three surface and two satellite”.
I want to explain to others why there is no warming, but I need a pin to bust this chart’s bubble. Anyone?
Hmmm… wonder why that graph doesn’t match up so well with this one:
http://www.woodfortrees.org/plot/uah/mean:12/plot/hadcrut3gl/from:1979/mean:12/plot/gistemp/from:1979/mean:12/plot/rss/mean:12
To Jose Suro , You must have been reading my mind when you posted your comment five minutes after my own. Your story and the links you provided fit well with what I was trying to convey. I am glad that my tin foil hat was at the cleaners.
It would seem to me that if the insurers as a group calculate rates that are based on projections of a single firm and that overestimate major storm frequency by a factor of that are, and these insurers continue to do this over an extended period of time, they may be accruing a rather substantial liability in the form of a potential Clayton Act suit. In such a suit both the plaintiffs and the defendants would rely on statistics to make their case. The plaintiffs could use the simple binomial theorem that is taught in the first year of statistics The defendants would, I presume, have to counter with the sort of gobbledygook that Willis described. I wonder which would be more persuasive to a jury. Right now the defendants would have “political correctness” in their corner but what about five years from now? The Clayton act includes a provision for treble damages, which might make this more interesting to the sharks. I am not an attorney but I did take a class in antitrust law a few million years ago, and this smells like blood in the water to me.
Chris Riley says:
February 27, 2011 at 3:12 pm
Very interesting, Chris. I am interested in talking to regulators and state senators and representatives at this time. I am not quite to the point of filing a lawsuit. Thanks much for your analysis.
Booker writes in his article:
When less partisan observers examined the paper, however, they were astonished. Although Nature has long been a leading propagandist for man-made climate change, this example seemed truly bizarre. Why had this strangely opaque study been based solely on the results of a series of computer models – mainly provided by the Hadley Centre and RMS – and not on any historical data about rainfall and river flows?
The paper described modeling at atmospheric conditions in Britain in 2000 and compared it to a hypothetical atmosphere with preindustrial levels of GHG’s to look at the statistics of heavy rainfall events. Their abstract and the news report, recognized that their work was not definitive, but the result was worth reporting.
The precise magnitude of the anthropogenic contribution remains uncertain, but in nine out of ten cases our model results indicate that twentieth-century anthropogenic greenhouse gas emissions increased the risk of floods occurring in England and Wales in autumn 2000 by more than 20%, and in two out of three cases by more than 90%.
While the paper seeking to explain Britain’s floods in 2000, was based on computer simulations, the other nature paper published in this subject did compare data and simulations. The article is behind a pay wall, and I only have access to the abstract:
http://www.nature.com/nature/journal/v470/n7334/full/nature09763.html
says the following:
…Here we show that human-induced increases in greenhouse gases have contributed to the observed intensification of heavy precipitation events found over approximately two-thirds of data-covered parts of Northern Hemisphere land areas. These results are based on a comparison of observed and multi-model simulated changes in extreme precipitation over the latter half of the twentieth century analysed with an optimal fingerprinting technique. Changes in extreme precipitation projected by models, and thus the impacts of future changes in extreme precipitation, may be underestimated because models seem to underestimate the observed increase in heavy precipitation with warming 16.
It is clear that data was consulted and compared with the models for the second paper on flooding, and it seems the model they used underestimated the difference. I wonder why Booker didn’t comment on this paper, which cannot be criticized for not using data? Was it because he was unaware of its exit, or was it because the showed an inconvenient truth.
Of course criticism of the use of computer modeling is a very convenient argument for “skeptics”. There is no other way than computer modeling to determine if GHG’s are responsible for trends that are observed or will extend into the future. Without modeling, mankind will be ignorant of what is to come, and those who oppose action to stop global warming would prefer that state of ignorance.
From kbray in california on February 27, 2011 at 12:18 pm:
A. (flippant answer) It’s from Tami’s site therefore it may be freely ignored as inconsequential.
B. There actually has been some warming, since around 1850 at the end of the Little Ice Age. This is normal natural fluctuation. It hasn’t gotten as warm yet as the Medieval Warm Period, which wasn’t as warm as the earlier Roman Warm Period. The interglacial is winding down, with the start of the next major ice age arguably overdue and coming soon, on geological time scales. Such drops can happen fairly quickly, might be in ten years, could happen a thousand years or more from now.
The “warming” matches the Pacific Decadal Oscillation, a major influencer of “global climate,” specifically the temperature we notice here on land in this skimpy atmosphere. It has a warm phase that lasts about 30 years, the last one corresponded to the “unprecedented” rise in surface temperatures that has (C)AGW proponents in a lather. It’s in the cool phase now, and it certainly looks like we have about 20 to 30 years of cooling global temperatures coming. Just as the PDO and other natural factors lined up for that “unprecedented” late-20th century warming, now they are lining up for cooling. The rising atmospheric CO2 concentrations ain’t doing anything noticeable to counter that.
The surface temperature records are a mess. If we had anything else of that length and resolution to use, we would use that instead as the current records are hardly suitable for scientific purposes. For a sample of how bad the recording at individual stations can be, check out the Surfacestations project. From those records, various horrible things are done to the readings to generate the assorted “global average temperature” numbers. A sample is found at E.M. Smith’s site, specifically his analysis of NASA-GISS’ GIStemp (Hansen’s pride and joy) which includes related issues with the historic temperature records and the terrible numerical abuse they’ve suffered.
Various temperature datasets similarly processed by the University of East Anglia’s Climatic Research Unit (with the UK Met Office’s Hadley Center) are also often mentioned. Along with the Climategate release of various CRU emails came programs and related data, including the famous “HARRY_READ_ME.txt” file (one link of many), compiled by some hapless programmer brought in to try to make sense of their data and code. During the analysis of that file, along with other searches for info on CRU’s workings, it was learned: they likely no longer have the older original records, don’t know what adjustments were made to them, thus the data has to be accepted on faith; the programs themselves are a garbled mess; and there are elements that are, yes indeed, completely made up.
Thus of the temperature readings taken in air on land, the raw data has issues, adjustments to those records are often questionable, and the further processing is normally atrocious. Plus the idea of using those to get a “global average” is somewhat absurd to begin with. First off, only about 30% of the globe is land, the oceans store much more heat than all of the atmosphere, we should be checking them to see where the climate is going. Also, the climate is not global. When analyzing data from individual stations and areas, as has been done by Willis Eschenbach and others, as presented on WUWT and elsewhere, we see long term trends of slight cooling, basically no change, and sometimes warming which is often tied to Urban Heat Island effects and land use changes (individual stations and small areas normally). And the much-touted homogenized pasteurized highly-processed global average numbers tend to show warming greater than what a careful examination of the real original data reveals, as if their processing was designed so that such “alarming” warming be shown regardless of the original data.
Dr. Roy Spencer is one of the keepers of the UAH satellite record, his site is well worth checking out. The latest monthly update was January 2011, there you’ll see the graph of the satellite record since its start in 1979. The PDO switched from warm to cool phase around 2008, so basically the warming shown on the chart corresponds to the 30 year PDO warm phase. Below that is a Global Sea Surface Temperature chart showing a decline for about 8 years now. The oceans cool, the land cools. The chill is coming.
The now-classic bits about the recent relative flatness of the global temperatures come from a BBC News Q&A from February 2010, from UEA-CRU’s Professor Phil Jones, post-Climategate when he was in a seemingly-conciliatory “CYA” phase:
Also check out the “A” response, where you’ll see the “unprecedented” rate of late-20th century warming wasn’t unprecedented at all.
With 2010 on the books, things sure ain’t changed much. The atmospheric CO2 concentrations go up, the global temps are not matching suit. (C)AGW proponents are sputtering about “missing heat” and “tipping points” and how the effects of the warming are somehow currently masked but SUDDENLY the warming will rush back in and fry us all if we don’t IMMEDIATELY cut back our CO2 emissions, which we should do anyway just in case they are right (although they are CERTAIN they are right), etc, etc.
Sorry, but you can’t say there has been no warming. There has been warming, since the Little Ice Age, and during the recent warm phase of the PDO. And now there shall be cooling. While one might say that some of that warming was caused by humans, it’s been repeatedly shown on WUWT and elsewhere that it’d be a small amount of what was reported, and various natural factors can account for the total warming anyway. Certainly, above all else, all of the warming was not mankind’s fault, and especially was not due solely to our CO2 emissions (including equivalent ones as from methane).
The Big Message: There have been no temperature increases nor rates of temperature increases seen that are outside the bounds of natural variability.
Try Jo Nova’s site, specifically her Skeptic’s Handbook. Arguing flat-out “there is no warming” is a non-starter, thus “why” that is doesn’t work. Blaming the warming seen, modest as it is by historical and geological scales, solely on human “carbon” emissions is the problem, and where their faux “science” falls apart. That is where they are weakest, that is where you should first attack.
[[[ JohnWho says:
February 27, 2011 at 2:39 pm
kbray in california says:
February 27, 2011 at 12:18 pm
It shows an increase in temperature from 5 sources, “three surface and two satellite”.
I want to explain to others why there is no warming, but I need a pin to bust this chart’s bubble. Anyone?
Hmmm… wonder why that graph doesn’t match up so well with this one:
http://www.woodfortrees.org/plot/uah/mean:12/plot/hadcrut3gl/from:1979/mean:12/plot/gistemp/from:1979/mean:12/plot/rss/mean:12 ]]]
JohnWho: this fellow: “slioch” says:
“Here is a graph of all five temperature series (three surface and two satellite) with the effects of natural variations from El Nino/La Nina and volcanoes and sunspots removed:”
http://tamino.files.wordpress.com/2011/01/adj1yr.jpg?w=500&h=325
How could you ever segregate out and remove:
1)”effects of natural variations from El Nino/La Nina”
2)”effects of natural variations from volcanoes”
3) “effects of natural variations from sunspots”
and quantify them individually from a single temperature reading?
It just sounds like some kinda voodoo to me or just guessing.
If these really are the “official temperatures”, I surmise that the problem arises from an accuracy issue attributed to the Urban Heat Index (UHI), Airport Jet Exhaust Index (AJEI), and some “correction of reading gaps”(CORG) based on some climate model calculations.
If there really is no substantial warming, then the worst explanation is just plain outright fraud. If that is the case, it’s gonna be nasty out there.
Thanks for the level heads on this site, I have learned a lot.
I once believed in “AL”, but now I ask questions.
kadaka (KD Knoebel) says:
February 27, 2011 at 5:42 pm
Thank you very much “kadaka”.
Your summary has clarified the issue in a short summary.
There are a lot of clear thinking brains on this weblog.
All your contributions are most appreciated. Thanks again. kbray.
Stonyground says:
Either I am mistaken or there is a lot of sloppy journalism out there.
You are not mistaken.
eadler says:
February 27, 2011 at 5:26 pm
“Without modeling, mankind will be ignorant of what is to come, and those who oppose action to stop global warming would prefer that state of ignorance.”
=========================================
No.
Without idiots, mankind will be better prepared of what is to come or NOT to come, and those who oppose action to “stop” [in quotes] natural cycles would prefer that state of enlightenment.
Chris
Norfolk, VA, USA
[Mods note: I think my last message may have gone to spam. Thanks for checking. Chris]
[Rescued & posted. ~dbs]
Jimmy Haigh says: “Things like that happen when cities are built below sea level.” –Vladimir
“Absolutely. It’s not rocket science. And NO is built below sea level. And in a hurricane prone area. The shape of the Missippi delta is governed by hurricanes. Katrina was one such shape-changer.”
A guy told me in 1966 that N’yawlins was below sea level. I thought that was the stupidest thing I’d ever heard. Of course, back then I hadn’t heard about computer program circle jerks and Nature.
eadler said
“…Here we show that human-induced increases in greenhouse gases have contributed to the observed intensification of heavy precipitation events found over approximately two-thirds of data-covered parts of Northern Hemisphere land areas. These results are based on a comparison of observed and multi-model simulated changes in extreme precipitation over the latter half of the twentieth century analysed with an optimal fingerprinting technique. Changes in extreme precipitation projected by models, and thus the impacts of future changes in extreme precipitation, may be underestimated because models seem to underestimate the observed increase in heavy precipitation with warming 16.”
* * *
Your quote from the abstract still does not pass scrutiny. While it said they were a “comparison of observed and multi-model simulations” it does not say how much data was used from both, and how it affected the outcome of the study.
For all we know, they could have thrown stones into the Thames for a day, came back, and ran 10 models that all ran off the previous model’s results.
And, as evident from Willis’ brilliant post, that’s exactly what they did.
Mods
I posted a comment and it didnt say
” …awaiting moderation.”
Does that mean it wasnt posted?
Thanks
[Rescued & posted. ~dbs]
But my models, of models, of models, of models, of models……………….
As with all computers rubbish in rubbish out no matter what you are using them for.
James.
Christopher Booker believes in intelligent design, that asbestos and talcum powder are chemically identical and that there is no proof of a link between passive smoking and lung cancer. His column on “amazongate” was so cringingly wrong in pretty much every detail that The Telegraph felt compelled to print a full page retraction. Basically, having him on side in any science debate counts as a loss.
eadler says:
February 27, 2011 at 5:26 pm
“…..There is no other way than computer modeling to determine if GHG’s are responsible for trends that are observed or will extend into the future…..”
That’s complete nonsense. The standard statistical method is to look for correlation between two variables. If there is significant correlation then most likely there is some kind of causal relationship. This method is basic and easy to understand.
.
As the authors resorted to using multiple computer models in a complex and confusing tangle, it strongly suggests that the correlation between flooding and Co2 does not in fact exist. All they had to do was look at the graph.
But if they had come to the obvious conclusion that there is no correlation it would not help to boost the profits of RMS.
Chris
“Theo Goodwin says:
February 27, 2011 at 2:37 pm
Jose Suro says:
February 27, 2011 at 8:42 am
“RMS is playing with our money and the insurance companies love it.”
Jose, I hope you notified the author of the Herald Tribune column about this forum. Very interesting information. I will watch for more articles from the Herald Tribune. I live just up the road from Sarasota. Thanks so much for the info.
Theo”
You’re most welcome Theo. I have not contacted the Herald about WUWT because I live in Pinellas County – maybe I should. The November 14 editorial I linked to above should be required reading for anyone interested in this climate change debacle.
The piece is very well researched and written, and it is not only a who’s who in this travesty, but also gives a detailed account of how the “science” gets manipulated (ignored is a better word) by these modelers to help justify themselves and big companies cashing in on billions of our hard earned dollars.
It made my blood boil!
Best,
Jose
kbray in california says:
February 27, 2011 at 5:46 pm
[[[ JohnWho says:
February 27, 2011 at 2:39 pm
kbray in california says:
February 27, 2011 at 12:18 pm
It shows an increase in temperature from 5 sources, “three surface and two satellite”.
I want to explain to others why there is no warming, but I need a pin to bust this chart’s bubble. Anyone?
Hmmm… wonder why that graph doesn’t match up so well with this one:
http://www.woodfortrees.org/plot/uah/mean:12/plot/hadcrut3gl/from:1979/mean:12/plot/gistemp/from:1979/mean:12/plot/rss/mean:12 ]]]
The woodfortrees graph looks different because the graphs are of temperature anomalies ( the change in temperature with respect to a reference period) and the reference periods are not the same for all 5 graphs. Tamino recalculated the anomalies for all 5 using a common reference period, which is why the overlay so well.
JohnWho: this fellow: “slioch” says:
“Here is a graph of all five temperature series (three surface and two satellite) with the effects of natural variations from El Nino/La Nina and volcanoes and sunspots removed:”
http://tamino.files.wordpress.com/2011/01/adj1yr.jpg?w=500&h=325
How could you ever segregate out and remove:
1)”effects of natural variations from El Nino/La Nina”
2)”effects of natural variations from volcanoes”
3) “effects of natural variations from sunspots”
and quantify them individually from a single temperature reading?
It just sounds like some kinda voodoo to me or just guessing.
Tamino used the estimated forcing for volcanoes and the El Nino index as his input data, and processed them with a multiple regression, accounting for time lags, to get a best fit for their influence on temperature. Tamino knows statistics better than you or I do. If you don’t accept what he did as reasonable, it would make your case better if you could cite an expert that explained what is wrong with what he did:
Hence for each temperature data set, we’ll do a multiple regression of the data since 1975 (or whatever we’ve got) as a function of MEI, volcanic forcing, a 2nd-order Fourier series, and a linear time trend. We’ll allow for a time lag in the influence of MEI and volcanic forcing. Then we’ll take the original data and remove the estimated part due to MEI, volcanic forcing, and annual cycle. Finally we’ll put them all on a common baseline, using 1980.0 to 2010.0. This will give us an “adjusted” data set (a name which may give some people fits), one which is adjusted to compensate for el Nino, volcanoes, and annual cycle residue.
If these really are the “official temperatures”, I surmise that the problem arises from an accuracy issue attributed to the Urban Heat Index (UHI), Airport Jet Exhaust Index (AJEI), and some “correction of reading gaps”(CORG) based on some climate model calculations.
If there really is no substantial warming, then the worst explanation is just plain outright fraud. If that is the case, it’s gonna be nasty out there.
Thanks for the level heads on this site, I have learned a lot.
I once believed in “AL”, but now I ask questions.
If you are asking real questions, you should pay attention to the answers.
The important thing to note, as Tamino points out, is that there is no urban heat island effect associated with the satellite data, and it still agrees pretty well with the thermometer based data. This indicates that the temperature adjustments made to correct anomalies in the thermometer data, due to changes in equipment, human error, and to eliminate the UHI have been reasonably successful.
Despite the individual stations that have been cited, which show an apparent UHI effect, comparisons of purely rural data, with the total world temperature data, show that the impact of UHI on the global data has been negligible.
Chris Wright says:
February 28, 2011 at 3:58 am
eadler says:
February 27, 2011 at 5:26 pm
“…..There is no other way than computer modeling to determine if GHG’s are responsible for trends that are observed or will extend into the future…..”
That’s complete nonsense. The standard statistical method is to look for correlation between two variables. If there is significant correlation then most likely there is some kind of causal relationship. This method is basic and easy to understand.
Chris, your statement is completely wrong . Correlation indicates that there is a possible relationship between 2 variables. It is not an indicator of which variable is a cause, and the cause could be that a third variable, which is not included in the correlation. No real scientist believes that correlation by itself determines cause and effect. In the real scientific world cause and effect are determined by a theory which must be consistent with the explanation of other phenomena in addition to the phenomenon that is being examined. This is how science has advanced over the centuries.
To claim that correlation can be used to predict the future also doesn’t make sense. There is no future data that can be used to set up a correlation. Without a physical theory, and a model calculation there is no way to predict what will happen in the future.
.
As the authors resorted to using multiple computer models in a complex and confusing tangle, it strongly suggests that the correlation between flooding and Co2 does not in fact exist. All they had to do was look at the graph.
But if they had come to the obvious conclusion that there is no correlation it would not help to boost the profits of RMS.
Chris
Sorry but you are absolutely wrong about the non existence of correlation in this case. Eschenbach’s graph, of the maximum precipitation amount at any station in the area, in a 24 hour is a noisy one from which one cannot deduce any sort of trend. Different measures are used to indicate the frequency of heavy rainfall events and flooding, in the scientific literature, which I pointed out in my references.
The use of a computer model, which includes the physics of weather and climate. Since the models necessarily contain empirical factors that have a degree of uncertainty,the use of multiple models provide a means of understanding the level uncertainty in the answer they provide.
J. Felton says:
February 27, 2011 at 8:58 pm
eadler said
“…Here we show that human-induced increases in greenhouse gases have contributed to the observed intensification of heavy precipitation events found over approximately two-thirds of data-covered parts of Northern Hemisphere land areas. These results are based on a comparison of observed and multi-model simulated changes in extreme precipitation over the latter half of the twentieth century analysed with an optimal fingerprinting technique. Changes in extreme precipitation projected by models, and thus the impacts of future changes in extreme precipitation, may be underestimated because models seem to underestimate the observed increase in heavy precipitation with warming 16.”
* * *
Your quote from the abstract still does not pass scrutiny. While it said they were a “comparison of observed and multi-model simulations” it does not say how much data was used from both, and how it affected the outcome of the study.
For all we know, they could have thrown stones into the Thames for a day, came back, and ran 10 models that all ran off the previous model’s results.
And, as evident from Willis’ brilliant post, that’s exactly what they did.
The purpose of an abstract is to outline what was done and what result was obtained. If you want more detail, you need to pay to read the paper, or go to a university library to get access to it. The abstract reported that they compared model output to data, and got close to the same results for the output representing the real world data.
In fact, if you read my post carefully, you would have understood that Willis Eschenbach were commenting the abstract from a different paper on flooding, looking only at England and Wales, from the one I posted. It was published in the same issue of Nature. In that case the abstract did not mention real data, but it is clear from the supplementary data linked on the Nature web page containing the abstract, that the full paper must have some comparison with real data, because some graphs were shown which involved real data.
Eschenbach criticized the paper on England and Wales without reading it. He himself said he only consulted the abstract and the supplementary data page, which he obviously didn’t look at it carefully,because he claimed that no reference to data was present in the paper. Under the circumstance, I don’t see how one could give Eschenbach’s criticism, or Bookers echo of it very much credence.
From eadler on February 28, 2011 at 8:41 am:
Sir, are you a bald-faced liar, or just stupid? It says right at the start of Mr. Eschenbach’s article:
He then launches into his detailed analysis of the paper. He quotes directly from the paper, which I can verify as the paper has now been posted online:
http://sciences.blogs.liberation.fr/files/pluies-diluviennes-galles-climat.pdf
Yes, that is the full English-language version of Pall2011, which Mr. Eschenbach critiqued.
As is common, the supplementary info is available for free. The link to it was provided by Mr. Eschenbach. As he said, it contains much of the concepts of the paper, thus those without access to the paper still can see at least that much. As is customary, the abstract is free, it can be found here, which is also the paywall link.
I have re-read Mr. Eschenbach’s article, read all comments he posted there. Nowhere did he say he only consulted the abstract and the supplementary data page, as you have said he did. (BTW there are 13 pages in the supplementary file, thus I wonder if you have even looked at it.)
It appears, sir, that you owe Mr. Eschenbach, an honorable man, an apology and a retraction.