Bishop Hill By applying inappropriate techniques, Bob Ward can prove that right is wrong
Jan 17, 2013 Climate: Surface Climate: Ward Bob Ward, writing at the Greenpeace website (where else?), is getting into the statistics of temperature trends once again:
Dr Whitehouse, a former BBC science correspondent, has been churning out a steady stream of error-filled articles on the Foundation’s website to try to cast doubt on the evidence for climate change. His description of the Met Office’s latest findings was similarly flawed. He claimed, for instance, that there has been “a global temperature standstill (from 1997 to present)”, which is a favourite falsehood disseminated by climate change ‘sceptics’ and their promoters in the media, such as David Rose of ‘The Mail on Sunday’ and Christopher Booker of ‘The Sunday Telegraph’.
In fact, the Met Office’s HadCRUT4 database of monthly global temperature measurements shows very clearly that the linear trend in temperature between January 1997 and November 2012 (the figure for December 2012 has not yet been published) is a warming of about 0.05°C per decade, which is statistically significant at the 95 per cent level when simple linear regression using ordinary least squares is applied to the data (it should be noted that it is very difficult to determine whether temperature trends over such short timescales are really statistically significant because of autocorrelation effects).
So, when David Whitehouse says there has been a global temperature standstill since 1997, Bob says this is "flawed". Why? Because the OLS trend is above zero, and that this trend is "statistically significant". He then notes that in fact OLS is inappropriate for calculating trends in temperature series. In other words, the trend cannot be shown to be statistically significant using this technique.
Which means that David Whitehouse's analysis was correct.
Reader Comments (65)
It does save skeptics work when alarmists refute themselves.
Fundamentally an "angels on the head of a pin" argument anyway. 0.05 degrees? Is that all you got?
I would like to express it as "one twentieth of a degree" warming. In a decade. Probably. Strangely I am not feeling threatened..
Has Bob protested yet against James "Standstill" Hansen?
Bob has very fast fingers but not for the first time he shown how poor is PR skills are , and yet his still the 'go to guy' for this type of thing . Long may that continue .
James Delingpole must feel mightily aggrieved not to get a mention amongst the "promoters in the media".
Ward, ward.... didn't he play 'Robin' in the 60s Batman series?
So the temperatures rose by 0.2C/decade from 1980 to 1996, and have trended by 0.05C/decade from 1996 to 2012, with a concomitant rise in CO2 of around 8% in the atmosphere. However way you look at it, whether it's statistically significant, or not, it is defying the supposed relationship between carbon dioxide, and points to natural forcings which they, the scientific community, clearly know nothing about. In any other science this would surely lead the scientists to reappraise the certainty with which they attribute the late 20th century rise in temperature to emissions of CO2 by humans. But no, the answer is the "deniers" are spreading misinformation.
Bizarre. Bob Ward seems to understand what is wrong with his argument but can't help giving it anyway. More evidence that the whole AGW debate has gone post-rational.
I dare say David Whitehouse will get round to demolishing Bob Wards dodgy statistics in due course. But it was interesting to see that when the met office story was doing the rounds last week no one quoted old Bob. In the past he would have been the go-to quote man, but it seems not anymore. Now Bob has turned on the media. How to win friends and influence people. A communications professional?
Most of Ward's article concerns the "secret" funding of the GWPF. Clearly he believes that he who pays the piper calls the tune. An interesting insight into his own ethical standards.
also 0.05 degrees c per decade is NOT significant at the 95% level if you consider the errors properly. The error of each year is 0.1 - 0.15. Bob has read out the error on the data points with no errors and just quoted the error on the rms of the spread - schoolboy mistake.
Careful - just because Ward's method is not appropriate does not necessarily mean that the opposite is true. Perhaps neither of them are right!
Whitehouse is still wrong if the OLS trend is positive. Slower rate of warming than previous decades would be fine, but not 'standstill', which implies a zero trend. However, we should all stop obsessing about the meaningless statistical significance of such short term trends, either way. The world is warming, currently at a slower rate than in the recent past. And, we all want to know why.
http://www.climate-lab-book.ac.uk/2012/global-temperatures-over-the-past-decade/
Ed.
"... is a warming of about 0.05°C per decade"
So that's a warming of about 0.005°C per year.
“is a warming of about 0.05°C per decade”
Two things about trends, one is the rate, in this case, as at Nov 2012 it is nearer to “about” 0.04°C per decade.
Second is the direction, this metric, peaked in March 2007 at about 0.30°C per decade and since then it has been heading south losing “about” 90% of its carrying capacity in just over 5 years.
The longer 30 year trend stands at 0.16°C per decade, down from a peak of 0.20°C per decade in Dec 2003, so "about" a 20% reduction in 9 years
Cause for concern?
I have repeatedly exchanged thoughts with Bob Ward. I have come to the conclusion that he does not feel bound by facts, evidence, or logic.
You really have to wonder when a guy like this gets so worked up about us 'sceptics'....
Ed Hawkins
“The world is warming, currently at a slower rate than in the recent past. And, we all want to know why.”
Well put, that is the point!
Since Bob is apparently happy to prove that the trend per century is around 1.2C, which is well below the 2C/century panic level that the politicians have been told is the maximum permissible, will Bob be telling the PM et. al. that the panic is over?
What is the opposite of a tipping point?
"What is the opposite of a tipping point?"
Reality.
Ward is paid by alarmist Jeremy Grantham to give the alarmist spin, and we should not expect him to be someone who searches for the truth.
As Omnologos mentioned, none other than James Hansen, the high priest of CAGW, has come out and said that temperatures have been at a standstill for the last 15 or so years. Come on then Bob. What do you say to that then, eh?
An earlier exchange between Bob Ward and Phil Jones merely shows how arithmetically incompetent they are
http://bishophill.squarespace.com/blog/2011/11/23/the-scientific-firmament.html#comments
But who would expect religous bigots to be swayed by evidence anyway.
David Whitehouse burns Hansen
http://www.thegwpf.org/hansen-admits-global-temperature-standstill-real/
Perhaps he will get around to Bob ward next.
From DW's latest:
(my bold). Is that a not-so-subtle dig at Bob Ward?
At least David Whitehouse completed his PhD...
...be cool.
Bob's got a job to justify.
0.05 C/decade
i could feel it was getting warmer
Hi Ed Hawkins
2 questions
Pre Copenhagen the Met Office had a press release,
Met Office - News - Global Warming Set to Continue - September 2009
http://www.metoffice.gov.uk/news/releases/archive/2009/global-warming
that did conform the 'slowdown' and in a linked paper in the press release) and a gave a rate since January 1999 of 0.07C per decade (to date - then 2009) vs 0.17C per decade late 2oth Century (80's 90's) The same release refering to decadal forecasts warned of half of the following 6 years being the warmest on record
http://www.metoffice.gov.uk/media/pdf/j/j/global_temperatures_09.pdf
and people at Copenhagen were talking of accelerated warming!! (Dec 2009)
1) May I ask what is the decadal rate from Jan 1999 to now (ie Dec 2012)
2) Secondly what is the highest rate of warming allowed by models for temperatures to warm, specifically obviously, only from the models that allow slowdowns of the observed length (15 years)
so basically with the new Met Office decadal forecast, all this is of interesting, and why I'd very much like some answers to the questions above
ie we need a much higher rate of warming to hit 2C by 2050.
thx
For Bob:
Source: http://en.wikipedia.org/wiki/Trend_estimation
Real data need more complicated models
Thus far the data have been assumed to consist of the trend plus noise, with the noise at each data point being independent and identically distributed random variables and to have a normal distribution. Real data (for example climate data) may not fulfill these criteria. This is important, as it makes an enormous difference to the ease with which the statistics can be analysed so as to extract maximum information from the data series. If there are other non-linear effects that have a correlation to the independent variable (such as cyclic influences), the use of least-squares estimation of the trend is not valid. Also where the variations are significantly larger than the resulting straight line trend, the choice of start and end points can significantly change the result. That is, the result is mathematically inconsistent. Statistical inferences (tests for the presence of trend, confidence intervals for the trend, etc.) are invalid unless departures from the standard assumptions are properly accounted for, for example as follows:
Dependence: autocorrelated time series might be modeled using autoregressive moving average models.
Non-constant variance: in the simplest cases weighted least squares might be used.
Non-normal distribution for errors: in the simplest cases a generalised linear model might be applicable.
Unit root: taking first differences of the data
That makes 0.5K per century. That mean that the Global Mean Temperature (IFF such a thing exists and is useful/has any meaning whatsoever) will be up by half a Kelvin (0.5 °C) at the turn of the century. It's much worse than we thought!
I don't know what you lot are on about ... self-contradiction is the highest level of proof in CAGW circles.
Expanding on my comment at 12:30 PM...
The words of Bob Ward's paymaster, Jeremy Grantham, must be paraphrased, because the original article has been altered to eliminate this approximate statement: When it comes to climate, it is unethical to understate the problem, and it is ethical to overstate.
http://www.nature.com/news/be-persuasive-be-brave-be-arrested-if-necessary-1.11796
Bob Ward understands his marching orders.
To put the trend calculations in context, I've updated a simple graph I made a while ago, showing 15- and 30-year OLS trends in the HadCRUT4 temperature series. If the past of the 30-year trend is a guide [no guarantees on that], then we may expect the current slowdown to extend.
The graph also allows one to speculate that the average warming rate is of the order of 1 K / century. Which doesn't alarm me, but might alarm others.
A Dr Betts explained me the situation on Twitter (hello Richard!)
It depends on context.
"Temps at standstill, and global warming stopped" = WRONG
"Temps at standstill, but global warming will resume later" = RIGHT
In fact, you can say pretty much anything and, as long as you add the mandatory ", but global warming will resume later", the biggest scientific institutions in the world will support you wholeheartedly, maybe Bob Ward too.
Let's give it a try..
"Polar bears are ok, but global warming will resume later"
"Arctic won't be free of ice any time soon, but global warming will resume later"
"A lot of model-based literature is rubbish, but global warming will resume later"
"West Ham playing superbly, but global warming will resume later"
"Elvis is alive, but global warming will resume later"
"Porcine and bovine flight sightings, but global warming will resume later"
See? It's easy, and it gets you a free ride indeed.
To some people even 0.0000000001C over a decade would be the C in CAGW, although I think it has more to do with their Egos than their intellignce.
Not a standstill. A fall.
HADCRUT4 2001 to date
Dropping at -0.018 per decade
http://www.woodfortrees.org/plot/hadcrut4gl/from:2001/plot/hadcrut4gl/from:2001/trend
I am really starting to get tired of being told that "this time it's different" and on scratching the surface finding that the only reason it's "different" is because you can't make your models work without including a wholly beneficient trace gas which is already at lower levels than through much of the earth's existence and landing it with qualities and abilities which increasingly it appears not to possess.
The fact that blaming CO2 also happens to provide the enviro-extremists with the peg they have been looking for decades to hang their anti-human obsessions on only makes me even more suspicious of the whole business.
This has been going on long enough for some positive, reliable, acceptable, understandable, empirical and largely unarguable evidence to manifest itself. It consistently refuses to do so.
As Bob Ward's increasingly manic and meaningless hand-waving demonstrates.
Not to mention that Judith Curry, who is far more qualified than Ward, noted that David Rose's statement in the Mail that the Met data showed warming at a standstill was correct.
Ward's qualifications are?
Barry Woods
“1) May I ask what is the decadal rate from Jan 1999 to now (ie Dec 2012)”
My fag packet (MS Excel) for the above period is “about” 0.08°C per decade
Crosschecked with WFT (but get confirmation) :-
“#Selected data from 1999
#Least squares trend line; slope = 0.0079669 per year”
http://www.woodfortrees.org/data/hadcrut4gl/from:1999/trend
Hope it helps, can’t help with the second question, I try to keep to actual observational data.
If we are to be worried by rising temperatures (regardless of whether they are happening or not), that suggests to me that the warmers know what the OPTIMUM global temperature is. They must do - else, if not, it may be that it would be good for us to be 3 or 4 degrees warmer.
Why can't they tell me? No warmer I have queried on this matter has even bothered to reply, never mind give me a run round. Is it too great a simplification? It seems not to me - again, if we do not know the optimum temperature, how on earth can we predict anything?
"... is a warming of about 0.05°C per decade"
So that's a warming of about 0.005°C per year.
Jan 17, 2013 at 12:11 PM | Robert Christopher
Put another way, that is 5/100ths of a °C per decade, or 5/1000ths of a °C per year! Are they really capable of measuring global temperatures to that level of accuracy, I think not. Much that same way Prof Phil Jones quoted rates of warming throughout the last century to the same ridiculous places of decimal point! Utterly ridiculous!
OT but no doubt Bish will invite Richard Betts to explain the latest Met Office cock up
http://www.thegwpf.org/met-office-botch-climate-scientists-stratosphere-wrong/
Dolphinhead
Lord Beaverbrook has posed the question here:-
"Questions for the UKMO"
http://www.bishop-hill.net/discussion/post/1727950?currentPage=4
"I shall continue to believe that the current "warm" period is part of the earth's natural cycles and a blessing which we should all enjoy while it lasts before the inevitable cold spell that will follow kicks in."
Mike, I won't be happy until we can make red wine in Yorkshire and modern day Vikings can resume farming in Greenland, so we've still got a way to go yet.
And please, no cooling for a long time - it's only been 200 years since the last ice fair on the Thames!
Interesting. On another site, just today, I have been linked to a critique by Bob Ward (of whom I had never heard, but I am a stranger to your world) of a paper by Bob Carter, in response to my raising the video of his (Carter’s) talk “torpedoing” the AGW hoax. Naturally, Carter’s critiqued paper was nothing to do with the video; when asked if the video had been watched, the reply was “No; a video is a very poor reference on scientific issues.” Is suspect that opinion holds only as long as the video gives the wrong argument.
Anyhoo… Not being scientifically trained (though vain enough to think that I can be scientific-thinking), my brain kept grinding to a halt as I read the critique. Can anyone tell me if it is the fault of my brain, or is it rather mangled verbiage?
http://www.eap-journal.com.au/archive/v40_i2_02_ward.pdf
(I do suspect it is my brain, as it took me a lot of reading to see the contradictory arguments used in the quoted passage above.)
RE: Dolphinhead above on the GWPF repot linked to. That is quite staggering! The money quote has to be:
"What is documented here is simply astounding. That which was thought to be understood is found to be misunderstood. Readings thought to be accurate are shown to be inaccurate. How the data were derived is found to be a secret now lost. The impact of the bogus data ripples through past results and, in particular, climate models, rendering old assumptions invalid. What was that line again about “settled science?”"
Jan 17, 2013 at 12:25 PM | Green Sand
"“The world is warming, currently at a slower rate than in the recent past. And, we all want to know why.”
Well put, that is the point!"
It is the point, but I'm not sure that Ed gets it. You have to understand that Ed, Richard and the others live in a world where the "science is settled". So they aren't taking on board the notion that it's stopped warming, all that has happened is there is a hiatus in the warming and they want to know why so they can carry on as they've done for the past two decades. It is a hypothesis that the scientific community is trying to prove right, any observations challenging the hypothesis are examined to find out why they don't prove the hypothesis. It doesn't appear to come into anyone's head that, as I've said before, if there are unknown unknowns causing the world to stop warming there may be unknown unknowns that caused the previous warming. So the attempts at finding out why is focussed on proving that warming is caused by humans.
I think that when two quite different questions get conflated things get inflated.
I can ponder the question: "Has the world warmed significantly?"
A question that depends on how well we can measure its temperature and how much data we have accumulated. It can be decided on the observational evidence combined with how we model the error in those observations. It could well be that 0.05ºC/decade over a 16 year period is significant, e.g. that it is unlikely that we could be so far out in our measurements.
I can also ponder the question: "Has the climate warmed significantly?"
Quite a different question with different criteria. It asks what we can and cannot infer about a much more abstract notion, the climate. Which is either a real thing or merely a construct. In either case some notion, some model of the climate is required.
In the case in question.I must ask as to whether the observed, lower level of warming is significantly different from what one would expect if:
a) the climate (as an abstraction or model concept) was warming at some rate. say not less than 0.15ºC/decade
b) the climate has not warmed or has cooled, e.g. a gradient <= 0ºC/decade.
I doubt that there is a sufficiency of either data, clarity of knowledge as to the nature of climate, or certainty as to an adequate model of climate.
I suspect that the world has warmed significantly, e.g. we are pretty sure that the temperature has gone up.
I also suspect that the climate is warming but doubt that there is sufficient contemporary evidence during the last 16 years to suggest that it is more likely to be warming at the 0.15ºC/rate than to have ceased warming.
Alex
Bob Ward 'does not feel bound by facts, evidence, or logic.' - nice phrasing Richard!