La-la Land again: Jim Hopkins gets it wrong

by Gareth on February 19, 2010

It must be sceptic idiot week at the Herald. Not content with allowing Garth George to make stuff up, today they unleash that mighty wit (or should that be twit?) Jim Hopkins, who has been reading the Daily Mail‘s daft coverage of a BBC interview with Phil Jones, the man at the centre of the beat-up over stolen emails:

Professor Jones discussed many things with the BBC, including the trouble he has “keeping track” of information, but the professorial concession the Daily Mail pounced upon – and our media ignored – was this: He said that for the past 15 years there has been “no statistically significant” warming. “No statistically significant warming”. None. It’s not happening. Since 1995, we ain’t got hotter. And that’s not the sceptics speaking. That’s from a man who garnered $22 million to prove we were getting warmer. Much warmer, worryingly warmer, “Lucy Lawless was right” warmer. But now he says we’re not. And haven’t been for 15 years.

I suppose we can’t expect a newspaper columnist and professional funny man to understand statistical concepts. Jones did not say that there had been no warming for 15 years. He said that the warming trend over the last 15 years just fails to meet one test for statistical significance. Here’s Jones in full:

Do you agree that from 1995 to the present there has been no statistically-significant global warming?

Yes, but only just. I also calculated the trend for the period 1995 to 2009. This trend (0.12C per decade) is positive, but not significant at the 95% significance level. The positive trend is quite close to the significance level. Achieving statistical significance in scientific terms is much more likely for longer periods, and much less likely for shorter periods. [My emphasis].

The Daily Mail misunderstood and/or misrepresented what Jones told the BBC — and that’s been extensively covered on the web. Hopkins is happy to repeat that misinformation without checking his facts. Perhaps he doesn’t understand how to use Google? More coverage of Mailgate at Deltoid, In It For The Gold, Real Climate, and for a full explanation of what statistical significance means in this context, Tamino has an excellent article here. Meanwhile, Jim should stick to his knitting instead of repeating tabloid lies from Britain.

{ 36 comments… read them below or add one }

RW February 19, 2010 at 10:22 am

The Herald desperately wants the “sceptics” to be right, and is happy to let uninformed twerps sound off in its columns.

Johnmacmot February 19, 2010 at 10:39 am

I heard Hopkins spout this kind of drivel a month or two back on Radiosport, where he gets a voice on the farming programme (Young Farmers’ competition host, I think.) from time to time. I don’t listen to that station much, but happened to hear some of his rave, post the CRU hack. I avoided hearing much of it, but I wasn’t impressed at the rabid tone.

As you say, being a bit of a wit doesn’t ensure being intelligent and well-informed.

Bryan Walker February 19, 2010 at 11:24 am

If Hopkins had bothered to look at the BBC interview he’d have seen that in reply to the previous question Jones had said that warming over the 35 year period from 1975 to 2009 at 0.161 degrees per decade was statistically significant. But that provides context, and who wants to be bothered with context in the excitement of the chase? Cleverness with words can’t disguise intellectual laziness. I doubt Hopkins has ever bothered to acquaint himself with what the science actually says.

C3P0 February 19, 2010 at 11:50 am

This is comical, 3 ‘gets it wrong’ articles!

It seems the entire world is either incompetent or part of a conspiracy to misinterpret reality, and only Gareth and Bryan can save us, one correction at a time…

…that or its just Gareth and Bryan who have ‘got it wrong’.

samv February 19, 2010 at 4:03 pm

Haha, false dilemma – there is a third option : the three people covered do not constitute “the entire world”. Therefore the either/or preposition is invalidated and your argument loses, run through with its own hyperbole… you really are arguing a straw man, tinman.

Dappledwater February 19, 2010 at 6:58 pm

“This is comical, 3 ‘gets it wrong’ articles!” – C3

Yup, 3 journalists barking up the wrong tree.

C3P0 February 19, 2010 at 12:10 pm

“Jones did not say that there had been no warming for 15 years. He said that the warming trend over the last 15 years just fails to meet one test for statistical significance”

This is pure doublespeak, the sentence contradicts itself. You say that the ‘warming trend’ just fails to meet ‘one test for statistical significance’. Are you serious? If it fails a significance test it is not a trend. So sorry, you ‘got it wrong’.

Where data has natural variation it is important to distinguish between that natural variation and the trends beneath it. If this cannot be done, to a certain level of confidence, the trend is not significant – it could just be random noise.

So you Gareth by maintaining that we are still warming even though it cannot be proved statistically are taking a very simplistic view. It is the exact view you have been critical of when we discussing if a cooling trend had been present since 1998 (which should also pass a significance test, not some 30 years proves significance pseudo science test). This kind of double standard is precisely what loses you your credibility. You continually show an obvious bias, you are not interested in data, you are interested in results. Your result is predetermined and you will set a method to reach it, hence the 30 year test.

noizy February 19, 2010 at 1:43 pm

…if a cooling trend had been present since 1998 … which should also pass a significance test…

As Jones points out, the trend is towards warming, but not statistically significant. When asked if there has been cooling (since 2002) he says that that “trend is not statistically significant”. So, one might assume that over the last decade or so (give or take a couple of years), the global temperature has remained fairly stable?

But, as always, it’s a matter of where you put your goalposts. Picking such short-term ranges allows people on either side of the argument to ‘prove’ their point. But, again, as Prof. Jones points out:

Achieving statistical significance in scientific terms is much more likely for longer periods, and much less likely for shorter periods.

So, what is the longer-term trend (which presumably levels out the effect of ‘noise’)? For warming, or cooling?

In the same interview Jones also says (points that seem to have been missed by the Daily Mail and Mr.Hopkins)…

I’m 100% confident that the climate has warmed … there’s evidence that most of the warming since the 1950s is due to human activity.

A statistically significant statement? ;-)

C3P0 February 19, 2010 at 2:21 pm

“As Jones points out, the trend is towards warming, but not statistically significant.”

Well no, the most recent data point is warmer than the first, however there is no significant trend.

Yes, it will be harder to find a trend when less data points are available.

The longer term data may smooth out the noise, yes, and since 1950 this shows warming, yes, but it depends on the hypothesis you are testing as to the most suitable place to put the goal posts.

If the the hypothesis was that, for example, the economic reforms of the late 80s early 90s produced higher economic growth you would need to look for a change in the trend for positive growth before 1984 to a higher rate of growth since.

In this case the hypothesis is that industrialisation has caused warming via CO2 emissions. We would then expect a higher rate of warming in the most recent part of the data set during the highest rate of CO2 emissions. It should then raise questions when there is no significant warming trend over the last 15 years.

This is not to say this invalidates the theory, but this lack of warming needs explaining.

“The fact is that we can’t account for the lack of warming at the moment and it is a travesty that we can’t” Kevin Trenberth

What is causing it? Solar effects? ENSO? Unknown effects? Do these have a higher effect that previously thought? How have these effected the data in the past?

These are all questions that need answering before the warming prior to the last 15 years can be used as evidence for man made warming over natural warming.

tomfarmer February 19, 2010 at 1:08 pm

Sad to say I find myself cast back into truisms like once is an accident, twice a coincidence, three times a habit. Now applicable to my readings of Jim Hopkins round and about the place..
In future I shall avoid this habit-ster because all too clear is his inability to distinguish a lie – “no” – from its golfball – “no statistically significant”.

Whatever happened to turn JimWit into dimwit!

C3P0 February 19, 2010 at 1:22 pm

So if a trend isnt statistically significant is it still a meaningful trend?

bennion February 19, 2010 at 1:43 pm

C3PO, some perspective is required surely?

Jones absolutely stands by all his other work that warming is occurring, in the same interview.

The time period concerned is short, so rationally, his theory is a legitimate possibility no?

I am all for skepticism, particularly by fiercely independent scientists and where it proceeds on the basis of evidence. James Lovelock, who was the first to verify the presence of CFCs in the atmosphere in 1970, and was initially absolutely scathing about the theory that they might exist in sufficient quantities to a skeptic about CFCs destroying the ozone layer.

It turns out that skeptics have been calling for the termination of the IPCC for years – because it inherently ends up understating the evidence of warming. Joe Romm: See http://climateprogress.org/2007/12/12/time-to-shut-down-the-ipcc/

But the main groups and individuals claiming to be skeptics today are illogical buffoons. Its hard to read Tamino’s deconstruction of the ridiculous Anthony Watt theory about weather station removal without rolling about the floor with laughter. http://tamino.wordpress.com/2010/02/15/dropouts/

I like Steve Chu’s comment yesterday:

“On balance if you look at all the things the IPCC has been doing over the last number of years, they were trying very hard to put in all the peer-reviewed serious stuff. I’ve actually always felt that they were taking a somewhat conservative stand on many issues and for justifiable reasons….

They should be able to say that this is serious science and take a somewhat conservative view. If you look at the climate sceptics, I would have to say honestly, what standard are they being held to? It’s very asymmetric. They get to say anything they want. In the end, the core of science is deeply self checking.

C3PO – for balance, do you agree its legitimate to ask deniers to provide all their emails over the past few years so they can be subjected to the same analysis as the Hadley Centre?

I am all for it, if only for the comic value.

Tom

C3P0 February 19, 2010 at 2:12 pm

I havent said anything in this thread about the wider argument at hand, only that no one should claim a warming trend exists over the last 15 years.

For balance I will say equally no one should claim there has been 10 years of cooling without showing statistical significance.

So, yes it would be unfair for a blogger or commentator to trumpet 10 years of cooling for the last twelve months and then get excited about ‘no significant warming for 15 years’.

What we can say is that; according to the surface data record there was a warming trend over the 20th century, and, over the last 15 years there is no significant trend up or down.

Bryan Walker February 19, 2010 at 3:23 pm

C3Po before you get too fixated on Jones you might like to look at my post on what Hansen has to say, on the basis of the GISS records which differ to some extent from the HadCRUT figures. He concludes:
“The Earth has been in a period of rapid global warming for the past three decades. The assertion that the planet has entered a period of cooling in the past decade is without foundation. On the contrary, we find no significant deviation from the warming trend of the past three decades.”
You can read the basis for his conclusion in the essay my post links to.

FlatFish February 19, 2010 at 4:33 pm

C3Po – Jones said “not significant at the 95% significance level. ”
95% is hell of a close.

Many, many things in life we take for granted fail to achieve this level of surety.
It is a hell of a high hurdle to best.
If I could get that level of surety before beating on a horse I would be very rich.
If households, businesses, accountants, credit managers and politicians required this level of certainty before making decisions then we wouldn’t be in the mess we are now!

bennion February 19, 2010 at 4:41 pm

Tamino nailed this whole debate last December. Full details at:
http://tamino.wordpress.com/2009/12/15/how-long/

His conclusion:

“Therefore we need at least 14 years of GISS data (from 1996 to the present) to draw a confident conclusion about the most recent trend. In fact, since we have additional unaccounted-for uncertainty (such as the parameter estimates for our ARMA(1,1) model), we actually need a bit more. Let’s say that less than 15 years of data allows no confident conclusion about whether the trend in GISS data is warming or cooling.

That does not mean that there’s been no warming trend in those 15 years — or in the last 10, or 9, or 8, or 7, or 6 years, or three and a half days. It only means that the trend cannot be established with statistical signficance. Of course, it’s another common denialist theme that “there’s been no warming.” This too is a fool’s argument; any such claims are only statements about the noise, not about the trend. It’s the trend that matters, and is cause for great concern, and there’s no evidence at all that the trend has reversed, or even slowed.”

Tom

Macro February 19, 2010 at 9:04 pm

Sample size in statistics is fundamental. The more you have the more confident you can be.

bennion February 19, 2010 at 4:46 pm

And his conclusion is discussed in relation to the recent Jones quote here:
http://tamino.wordpress.com/2010/02/16/growthgate/#more-2302

Tom

Rob Taylor February 19, 2010 at 5:46 pm

TEMPERATURE AND HEAT ARE NOT THE SAME.

1. The Earth is out of thermal equilibrium; more heat is being absorbed from the sun than can be radiated into space.

2. The excess heat is absorbed by the atmosphere, land, ocean, cryosphere, etc and redistributes itself between these components of the total Earth system – this is what Trenberth is talking about in his “travesty” email.

3. Thus, atmospheric temperature trends are an indicator, but not the final arbiter, of AGW. In the long term, thermal equilibrium will be restored, but we are not very likely to be here to see it…

Macro February 19, 2010 at 8:56 pm

That’s it in a nutshell.
But you try telling that to AGW denier et al!

Rob Taylor February 19, 2010 at 8:57 pm

For what it’s worth, I’ve sent the following letter to the Herald (I’m not holding my breath):

Your correspondent Jim Hopkins’ recent article, on climate scientist Prof. Phil Jones, misrepresents what Jones actually said, which was that global warming from 1995 to the present – a short interval, in climate terms – was close to the 95% confidence level (http://news.bbc.co.uk/2/hi/science/nature/8511670.stm).

So instead of the chance of global warming NOT happening over that period being only 1 in 20, it is perhaps only 1 in 18 or 19…

Would you invest your money on such long odds? Would you agree to an operation, if the chance of surviving it was only 1 in 18? I think not!

A former science and maths teacher myself, I believe Mr. Hopkins is grasping at straws in attempting to deny the real and present danger of global heating – as well as demonstrating a woeful lack of elementary mathematics.

Andrew W February 20, 2010 at 8:32 am

Probably most 10 or 15 year periods over the last 60 years had warming that was not significant at the 95% significance level, and probably most 20+ year periods over the last 60 years had warming that was significant at the 95% significance level. Like other statistical tests it’s the sample size thats important, you wouldn’t expect a political poll of 15 people to give you a result reliable at the 95% significance level would you?

noviceforsure February 20, 2010 at 7:34 pm

Rob Taylor February 19, 2010 at 5:46 pm
TEMPERATURE AND HEAT ARE NOT THE SAME.

1. The Earth is out of thermal equilibrium; more heat is being absorbed from the sun than can be radiated into space.

Rob, how do you know this? Please bear with me, I am trying to get to grips with the AGW science.

Thanks

Rob Taylor February 20, 2010 at 9:20 pm

Happy to oblige – the short answer is, by comparing measurements of incoming (solar) radiation and outgoing (earth) radiation by both ground- , ocean- and space-based instruments.

As there is no one satellite that is set up to make simultaneous measurements in both directions, some analysis is required. (Well, that’s not quite true, such a satellite (the Deep Space Climate Observatory) was built by NASA in the Clinton era, but the incoming Bush administration cancelled the launch and hid it away in a warehouse for 8 years – I wonder why? Hopefully, NASA will refurbish and launch it soon.)

Here are some links that I have found useful:
http://earthobservatory.nasa.gov/Features/EnergyBalance/
http://www.skepticalscience.com/Measuring-Earths-energy-imbalance.html
http://en.wikipedia.org/wiki/Global_warming

I can also recommend Gareth’s book, for a NZ perspective.

Macro February 20, 2010 at 9:26 pm

The result of increasing greenhouse gases results in less heat being radiated from the top of the troposphere.
The way the greenhouse effect works is that adding CO2 reduces the infrared out the top of the atmosphere, which means the planet receives more solar energy than it is getting rid of as infrared out the top. The only way to bring the system back into balance is for the whole troposphere to warm up. It is the corresponding warming of the low level air that drags the surface temperature along with it.
The reduction in infrared heat loss as a result of doubling CO2 over pre-industrial levels for instance, has been know for many years (a calculation by Plass in 1956 estimated the effect to be around 3.2 watts per square meter – a figure that has been recalculated using different principles and with much more rigour over recent times. (That is of course how science progresses). But with substantially the same result – ie a doubling of CO2 over pre-industrial levels results in a lowering of heat loss from the top of the atmosphere by around 3.2 watts per square meter.
You can read more if you want here:
http://www.realclimate.org/index.php/archives/2010/01/plass-and-the-surface-budget-fallacy/#more-2652

Kiwiiano February 20, 2010 at 7:37 pm

And remember C3Po et al, the measurements which may or may not be significant within the time frames being cherry-picked, are a way of trying to explain the observed changes to the biosphere, particularly in the polar and higher-altitude regions. Ask the people in the real worlds of Alaska, Canada, Greenland, Siberia, the Himalayas, the Andes, the Rockies, even Antarctica about the melting glaciers and tundra, the changes to plant and animal life, the rotten multi-year ice, the methane bubbling from the ground, the eroding coastlines. And while you’re at it, ask the atmosphere scientists and the astronomers about their inability to find sufficient (if any) explanations in the solar output, orbital positions or cosmic ray flux. And if you’re talking to the atmosphere boffins, you’d better get the details on the role CO2 and other GHG play in moderating the Earth’s temperature. We’d be a helluva lot colder if there were no GHG in the air, it’s a mystery why the sceptics are so determined to deny the possibility that increasing quantities could have unfortunate effects.
There was a very good interview with David Suzuki by Brian Crump last Monday, repeated this evening. He doesn’t take prisoners.
See http://podcast.radionz.co.nz/ngts/ngts-20100215-1914-The_big_picture-048.mp3

Rob Taylor February 20, 2010 at 9:26 pm

Suzuki lives in Vancouver, and said that there was so little snow around for the Winter Olympics that the organisers had to truck in snow from 150 miles away and spread it over hay bales laid on the ground for insulation…

C3P0 February 22, 2010 at 9:20 am

Yeah, and what if the winter Olympics were held in the North East USA this year? Please don’t start up the whole this place is warm this place is cold argument again….. Suzuki can play those games if he wants but he just insults the intelligence of his listeners

Bryan Walker February 20, 2010 at 9:27 pm

Kiwiiano, thanks for drawing attention to the Suzuki interview. I would have missed it otherwise. It was a real tonic to hear him.

Kiwiiano February 20, 2010 at 10:02 pm

I listened to the Suzuki interview (in Safari/OSX SL) then found that there was some sort of block on the file that prevented me from Saving the MP3. A second attempt of just (right)clicking the MP3 link on the NatRad page offered the option of downloading the attached file, which worked. I have a copy just in case anyone has problems. Not sure when the links disappear from RNZ site.

noviceforsure February 22, 2010 at 9:09 am

Rob Taylor February 20, 2010 at 9:20 pm
Happy to oblige – the short answer is, by comparing measurements of incoming (solar) radiation and outgoing (earth) radiation by both ground- , ocean- and space-based instruments

Rob
I followed the links you recommended. Thanks, very informative.

Bryan Walker February 22, 2010 at 9:55 am

C3Po
“Suzuki can play those games if he wants but he just insults the intelligence of his listeners.”

Suzuki is hardly playing games. He’s desperately in earnest in his appeal to the intelligence of his listeners. No insult there. I don’t quite understand how you detected one.

C3P0 February 22, 2010 at 10:47 am

By using local weather as evidence (for or against) of global warming

AndrewH February 22, 2010 at 10:30 pm

Suzuki wasn’t playing any games…my recollection of the interview is that he brought up the winter olympics in response to a question as to what he would say to a skeptic who points out it has been cold in New York or Europe this winter.

samv February 22, 2010 at 12:43 pm

Accumulated Ice/Snow cover is not a weather event, it indicates longer trends than individual snowfalls – months rather than a week or two. And you start heading towards that hazy line of statistical significance. Glaciers melting, retreating snow lines, melting permafrost are not weather events but part of a trend.

Bryan Walker February 22, 2010 at 1:12 pm

And Suzuki offers an accumulation of different effects, all in line with what can be expected from global warming, of which the lack of snow in Vancouver was only one. Pine beetle in Canadian forests, rising sea levels, the decimation of sockeye salmon, the migration of squid, were among the others he adduces. His alarm is well justified.

Previous post:

Next post: