Guest essay by Christopher Monckton
As soon as the BBC/Maslowski forecast of no sea ice in the Arctic summer by 2013 has been disproven (see countdown on right sidebar), WUWT will need another countdown. May I propose the Santer countdown?
On November 17, 2011, Ben Santer and numerous colleagues, including researchers from Remote Sensing Systems (RSS), published a paper in the Journal of Geophysical Research (Atmospheres) that, as his press release said,
“shows that climate models can and do simulate short, 10- to 12-year “hiatus periods” with minimal warming, even when the models are run with historical increases in greenhouse gases and sulfate aerosol particles. … tropospheric temperature records must be at least 17 years long to discriminate between internal climate noise and the signal of human-caused changes in the chemical composition of the atmosphere.”
In an earlier posting I demonstrated that for more than 17 years (now 17 years 7 months) the HadCRUt4 monthly global mean surface temperature anomaly dataset had shown no global warming distinguishable from the combined 2 σ measurement, coverage, and bias uncertainties published together with the data themselves.
However, there were those who said that, nevertheless, the HadCRUt4 data appeared to show some warming (albeit less than 1 Cº/century). Well, the first of the five principal datasets to show no warming at all for 17 years is likely to be the RSS dataset of Santer’s own colleagues. Today the data for August 2013 were made available:
The least-squares linear-regression trend on the data from the RSS satellites since November 1996 shows there has been no global warming at all for 202 months (16 years 10 months). In a few more months, unless an el Niño comes along in January, its favorite month, RSS may be the first dataset to show 17 full years with a zero global warming trend.
The NOAA’s 2008 State of the Climate report said 15 or more years without global warming would indicate what was delicately described as a “discrepancy” between prediction and observation.
Fifteen years without warming duly came and went: indeed, Professor Jones of the University of East Anglia was the first to admit this, in response to a question I had suggested to Roger Harrabin of the BBC (who had thought I was daft to suggest that there had been no statistically-significant warming for as much as a decade and a half).
So Santer moved the goalposts.
Taking the mean of all five principal global-temperature datasets (GISS, HadCRUt4, NCDC, RSS, and UAH), since the new millennium began on 1 January 2001 there has been no global warming at all for 152 months (12 years 8 months):
Therefore, those who are anxious to believe that the long pause in global warming is what the models expected, or at least allowed for, can take a crumb of comfort from that.
Two important caveats. First, linear trends are not predictions. They are only one way of representing the trend (if any) that has already occurred over a chosen period in a stochastic dataset such as the global mean surface or lower-troposphere temperature anomalies.
Secondly, a strong el Niño, or a resumption of stronger solar activity, or simply some hitherto-unexplained factor in natural climate variability, could cause a resumption of global warming at any time. The central consideration, then, is not whether there have been x years without global warming, vexing though this embarrassing statistic is to the true-believers and their models. It is the extent and the persistence of the discrepancy between predicted and observed global warming.
The monthly Global Warming Prediction Index keeps track of this discrepancy. The index number for September 2013, published today, is 0.22 Cº. That is how much the IPCC’s central projection of global warming over the 8 years 8 months from January 2005 to August 2013 has overshot the observed temperature trend.
Today’s index graph shows 34 models’ projections of global warming since January 2005 in the IPCC’s Fifth Assessment Report as an orange region. The IPCC’s central projection, the thick red line, is that the world should have warmed by 0.20 Cº (equivalent to 2.33 Cº/century) since that starting date.
The mean of the RSS and UAH satellite measurements, in dark blue over the bright blue trend-line, shows global cooling of 0.02 Cº (–0.22 Cº/century). The models have thus over-predicted warming since January 2005 by 0.22 Cº (2.55 Cº/century).
The 18 ppmv (202 ppmv/century) rise in the trend on the gray dogtooth CO2 concentration curve, plus other ghg increases, should have caused 0.1 Cº warming, with the remaining 0.1 ºC from previous CO2 increases. No warming has occurred.
On its own, the CO2 increase since 2005 should have caused a radiative forcing of 0.24 Watts per square meter, or 0.34 W m–2 after including the influence of all other greenhouse gases. Even without temperature feedbacks, according to the IPCC’s methods this forcing should have caused 0.1 Cº warming. Adding in the IPCC’s temperature estimates of temperature feedbacks and of previously-committed global warming should have caused up to 0.3 Cº warming since January 2005. None has occurred.
Note how the temperature has failed to rise since 2005, notwithstanding that the CO2 concentration has risen quite rapidly. The usual suspects like to display what they call an “escalator graph” with many of Santer’s “hiatus periods” of 10-12 years without any global warming, but an overall rising trend nonetheless.
However, previous periods free of global warming did not occur while Man was putting more CO2 in the air anything like as rapidly as he is today. Now that CO2 concentration is rising, so should temperature be rising, if the IPCC were correct about how much warming we should expect as CO2 concentration increases.
The “escalator graph”, then, is meaningless, except to the extent that the frequency with which the “hiatus periods” occurs suggests that the probability of seeing anything like the 2.8 Cº warming this century that is the IPCC’s central projection is not very great.
Though the IPCC projects that the world should have warmed by 0.20 Cº (2.33 Cº/century) since 2005, the mean of the RSS and UAH satellite datasets shows cooling of 0.02 Cº (0.22 Cº/century). The predicted and actual trends are visibly diverging. Solar physicists expect significant cooling in the coming decades. If they are right, the divergence will become more than merely embarrassing.
Even as things stand, if the IPCC overshoot over the past 104 months were to continue for 100 years the IPCC’s prediction would exceed the measured trend by more than 2.5 Cº.
If Jeffrey Patterson’s harmonic decomposition of the past 113 years’ temperature records is correct, the world may actually be cooler in 2100 than in 2000.
A shame we shall not be there to see the baffled faces of the profiteers of doom.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Re: Monckton of Brenchley 12 Sept 2013 at 1.10 am
“Mr Sherrington says I was wrong to take the mean of the monthly temperature anomalies from both terrestrial and satellite datasets………………………….
The correlation of satellite lower troposphere temperature with surface-based records is reasonable at some coarse scales, but it is a source of continuing problems. In Australia, the BoM made a great hullabaloo about DJF summer temperatures being the iconic “angry summer”, with average temperatures (from their data set) claimed to set all time new records.
With Jo Nova & some volunteers, we got stuck into this, Here is a graphs from Jo’s blog that I’ve stored on my web site.
http://www.geoffstuff.com/aust-uah-v-bom-1981-10-means%5b1%5d.jpg
You will see at once that the two records disagree by about 0.7 degrees for the last DJF.
This is a very large disagreement.
It was with examples like this in mind that I objected lightly to the averaging of records taken at different times and different places, broadly objecting to the average of a number of satellite and surface temperatures.
At a deeper level, some properties like autocorrelation are somewhat different for the satellite class of data and the surface-based. If you are going on to do statistical analysis like WUWT guest Jeffery S Patterson posted about the same date as this, you would need to quantify such differences. They are different animals, depending on which spectacles one is wearing.
Thanks to both Lord Monckton and R Barraclough. Don’t have MS office, so will sub a open source web tool.
Gail Combs & PeterB in Indianapolis,
I should have more clearly communicated my thoughts. I think Prof Peter Wadhams and his predictions should replace the Maslowski Countdown. The Maslowski Massive Miss should be properly and permanently enshrined in some way, but a new prediction countdown should replace it.
Maybe if a few of these guys get embarrassed they will dial back the fear mongering a notch or two or 10. Thank you Anthony for coming up with a way to hold at least one of these twits accountable. It is so frustrating to watch these people spout off and then just walk away. They take no responsibility for the laws, taxes, spending and public policy that they leave in their wake.
?? A volcano, unless it rains hot rock globally, is likely to cause some cooling. And the bolded word means the opposite of what you seem to think.
Robert Orme;
On further thought, are you recommending Nordhaus’ carbon taxes? Punishing energy production is contra-indicated. His models, like the CAGW ones, posit and assume CO2 harm, and discount/ignore warming benefits.
Greening of arid places and enhancement of drought-tolerance in food plants argue for CO2 subsidies, not taxes.
Mr. Sherrington continues to insist that one ought not to average satellite and terrestrial temperatures because they can be very different. Well, that is precisely why one averages them – to get some idea of what the middle position might look like.
As to autocorrelation, having verified with a statistician that the highly stochastic global temperature dataset is so little auto-correlated that one might as well use least-squares regression for the trend-lines, there is really no need to worry too much about whether the auto-correlation characteristics of satellite and surface datasets are to some small extent different. I’m not using AR(n) modeling because there’s no point, and it would generate more arguments than it would resolve. So I stick to good old least-squares regression, which everyone who has done Stats 101 understands.
One only has to look at the vast error-bars published by HadCRUt4 (i.e. plus or minus 0.15 K) to realize that when plotting global temperatures there is little point in counting angels on a pinhead, statistically speaking. And that’s before one starts on the enormous amount of tampering that GISS and others have done, suppressing global temperatures in the 1930s to make the apparent 21st century warming seem larger than it actually was.
All I do is take the data and process them in the simplest possible way, because that minimizes the opportunities for the usual suspects to whinge about what was done. They will whinge anyway, on principle, but ordinary people, on seeing the very clear temperature graphs I try to produce, are getting the picture very clearly now.
Even the scientific community are catching on to the value of clear data images. When I put out a one-page 200-month zero-warming graph at the back of the World Federation of Scientists’ lecture hall a couple of weeks ago, every copy was snapped up. The truly astonishing thing about the temperature record is that the mainstream media will not, will not, will not let the public know that global warming is not happening at the predicted rate or, recently, at all.
“The truly astonishing thing about the temperature record is that the mainstream media will not, will not, will not let the public know that global warming is not happening at the predicted rate or, recently, at all.”
The Met and RS have been pushing “rising temperature” all the two decades they knew that no such thing was happening. That is where the fault lies. CRU fully complicit in the Met fraud.
http://www.orthodoxytoday.org/blog/2009/11/global-warming-fraud-by-hadley-cru-exposed/
“Global Warming Fraud by Hadley CRU Exposed
November 20, 2009
Telegraph Media Group | by James Delingpole | Nov. 20, 2009
“The conspiracy behind the Anthropogenic Global Warming myth (aka AGW) has been suddenly, brutally and quite deliciously exposed after a hacker broke into the computers at the University of East Anglia’s Climate Research Unit (aka Hadley CRU) and released 61 megabytes of confidential files onto the internet. (Hat tip: Watts Up With That)
“When you read some of those files – including 1079 emails and 72 documents – you realise just why the boffins at Hadley CRU might have preferred to keep them confidential. As Andrew Bolt puts it, this scandal could well be “the greatest in modern science”.”
This did get press coverage – but what we got was a cover up.
The Met Office and CRU still conspire to deceive us, the RS still promotes this deception. The greatest scandal in modern science continues because those creating it still continue and will continue to continue until it is raised and destroyed at the poitical level.
You could alway take up your seat and bring this to everyone’s attention..
Terry Oldberg – your arguments are founded in your arcane knowledge of logic and which you have apparently put a great deal of your time. What defines your sophistry is your assumption that you can engage others and fling your arcane knowledge about as if it were common knowledge and to use esoterica as a weapon in a war of words. It is not acceptable, and it is inappropriate and haughty to purposely talk over the heads of lay and educated people from your unique but arcane station and expect an response on your terms. It also makes you look a fool but you’ve got to already know that. I’m thinking I’m seeing symptoms of Asperger’s here as evidenced by your trying very hard to not be accepted. Not uncommon at all among intelligent people.
Not sure that dp is right about Mr. Oldberg displaying any knowledge of logic. In this thread he has argued with the IPCC’s use of the word “predict” on the ground that there was no “event” for it to compare its predictions with. However, the rules of logic on the definition of terms require that, if a term is not to be used in its generally-understood meaning, that fact must be made explicit. I have been using the word “predict” in its plain and full meaning, from the Latin, namely to “say what will happen”. The “event” with which I have dared to compare the IPCC’s predictions is the observed global warming over the longest period before today during which no global warming was observed. For I am using “event” in its plain and full meaning: “that which has happened”.
To summarize: I have compared what the IPCC says will happen with what has happened. What has happened is not what the IPCC said would happen. It is as simple as that. Mr. Oldberg is an instance of that species of troll who, utterly unable to refute the clear and simple scientific points being made, does his inferentially paid worst to sow pointless confusion. Fortunately he is inept at doing so: therefore, his futile wafflings merely serve to undermine the case against the scientific truth that all with a clear eye can see: not much warming is to be expected from our adding greenhouse gases to the atmosphere, and not much is happening, even though in some quarters much warming is profitably but erroneously predicted.
I say “inferentially paid” because no one who was not paid would spend so much time and effort elaborately making a fool of himself.
Why would one want to correct for auto-correlation when that is what the climate is.
It is auto-correlated by its nature. Today’s temperature is likely to be similar to yesterday’s temperature and it is, in fact, quite influenced by yesterday’s temperature.
This September is likely to be similar to last September. It is more likely to be similar to last September than to the September of 100 years ago or 500 years ago. This September may be influenced by what last year’s September was in just a tiny way, but it is, in fact, a non-zero.
That is what the climate is. Energy slowly accumulates and slowly escapes. Correcting for auto-correlation is like taking out what we are trying to measure.
Suppose the anomalies for, say, May to August are 0.1, 0.2, 0.3 and 0.4. Then in September, it cools to 0.35. You have a one-month cooling trend, 2 data points, a 1 month trend. You don’t suddenly jump from no cooling to 2 months in the space of 1 month. Similarly with 202 data points, a 201 month trend.
@ur momisugly Terry Oldberg says:
September 12, 2013 at 8:17 am
“richardscourtney (Sept. 11, 2013 at 3:02 pm):
On Sept. 11 at 3:02 pm you state that “There is an ‘event’ (i.e. the global temperature in 1997)…” To state that the global temperature in 1997 is an example of an event is logically similar to stating that in a flip of a coin, “heads” is an example of an event. Actually, it is the flip that is an example of an event. Would you care to take another stab at the identity of the events underlying the climate models of IPCC AR4 or do you now admit that there are none?”
You know, I was going to leave this all to Lord Brenchley and Richard, who were doing such a wonderful and entertaining job of putting you in your place. But it appears you must play the fool, and this was, for me, the last straw, and I skipped the rest of the replies to tell you so (and thus, if I am repeating what anyone else has already pointed out, please forgive my impetuousity—but this Terry Oldberg person is really annoying).
First, you asserted,
” richardscourtney:
Contrary to your claim, the global temperature in 1997 is not an example of an “event.” ”
You are wrong. An event may be the experience of two or more events that occur in sequence or concurrently that can be subsequently categorized as an “event.” This is a well known, non-esoteric use of the term “event.”
You then asserted, “By the way, to inaccurately smear the reputation of a professional,
including me, is illegal under the defamation laws of both the US and the UK.”
You are wrong again, at least as regards the United States (as an aside, what would it mean if I “accurately” smeared your reputation? For one who prates about “logic”, and likes to toss around polysyllabic, esoteric terms, the vagueness and imprecision of “inaccurately smear”—as polysemic as any world or phrase you accuse Richard of using—is breathtaking). As the Supreme Court of the United States found, in its decision in re New York Times v. Sullivan, public officials could win a suit for libel only if they could prove “actual malice” which was defined in the decision as, “knowledge that the information was false,” or that it was published “with reckless disregard of whether it was false or not.”
By the way, the truth is always an absolute defense against libel in the United States.
“Professionals” are not given any separate standing under the law; and if they were, we have only YOUR assertion that you are a professional. At what, sir? I too am a professional. So what? The simple fact is, the law does not protect you from libel because you claim to be a “professional,” nor does it protect you from ordinary opinions which rise from the use of First Amendment rights. You must prove actual malice. Good luck with that.
You asked of Crispin of Waterloo about the “issue” of whether the global temperature in 1997 was an event. There IS no issue. It would appear that the only issue is that you alone are not aware that the mean of all the recorded surface temperatures for 1997 qualifies as an event.
As I told one who appears to be one of your ideological brethren, elsewhere (ie., he too appears to be a limaceous slubberdegulion), I have no rights on this blog to tell you to peddle your “logic” elsewhere—but I do wish you would.
@ur momisugly Gail Combs says:
September 12, 2013 at 10:31 am
By the way, that post was a tour-de-force! Ok, off subject slightly (Loved your other posts re: Jet Stream. Saw the report of research regarding the possible impact of jetstream orbit on Arctic Ice some time back, here on WUWT. Yours was a timely reminder!), but a tour-de-force—Brava!
Thought to ponder: what happens if the BRICS nations and the OPEC nations actually do reprice the cost of a barrel of oil in a new basket of funds, instead of the World’s Reserve Currency (currently the US Dollar)?
YOur post was not too far off topic when you realize that the entire AGW hoax has become so large because malefactors see it as a method to steal more money clipping the tickets on international transactions which they’re fighting to have governments mandate in order to “combat global warming” or “extreme weather events” or whatever they want to call it (h/t to Ian Wishart and “Air Con” and if you’re reading this, Ian, please: we want ANOTHER updated version!).
We are “skeptics” and “deniers” because we disagree with the way science has been perverted to serve the egos of some, the power-hunger of yet others, and the greed of many more. The purity of that anger at the abuse of Science aside, we shouldn’t forget the dangers intendent upon their success, should the hoaxters win their battle to put their argument across and drown everyone else out.
There is a great deal more at risk here than principles, and the first ones harmed will NOT be those getting ripped off in developed nations: it will be the poor who starve and die first due to insane legislation about “carbon.” Lord Brenchley is too polite in his references to Mr. Hansen et al.
Richard Barraclough says:
September 14, 2013 at 8:11 am
You have a one-month cooling trend, 2 data points, a 1 month trend.
You have to keep in mind that the numbers you see are actually the month long averages. So taking RSS for example for June, July and August, the numbers are 0.291, 0.222 and 0.167. If we make the simplifying assumption that temperatures changed linearly over these three months, you could argue that these were the values on the 15th of each month. However in each case, the values at end of each month were lower. That means the values at the start of each month had to be higher to compensate. So if the values went down from 0.291 to 0.167 over 2 months from June 15 to August 15, or whether they went down from 0.3325 on June 1 to 0.1395 on August 31 over 3 months makes no difference to the slope. Look at it this way, 3 data points do indeed cover 3 months since each data point covers a complete month.