Log in

No account? Create an account

Previous Entry | Next Entry

Global Warming - How to Lie with Statistics

I usually try to avoid getting involved in debates about global warming these days - it tends to suck up huge amounts of my research time when I try to get things right while the other "debaters" just slap together a few random links they got from Google within half a minute. However, sometimes I just cannot resist, and the following examination is the latest result.

It started out with this debate. I was given a link to this article by one Christopher Monckton as evidence that global temperatures in the last decade had not in fact increased, but decreased. So I took a look at the provided graph and was fascinated - by the sheer amount of blatant manipulation I encountered:

Fortunately, the author provided a source for his data - the HadCRUT3 data set. So I sat out to recreate the graph, and managed to do so.:

First of all, while the used data did show a cooling trend, my linear fit (done with GNUPlot) produced a cooling of

-0.00156427 °C/month.

Extrapolated for an entire decade, like the author has done, this would translate into:

-0.00156427*12*10 °C/decade = -0.1877124 °C/decade

This is not even half as much as the 0.4 °C/decade the author claimed. But wait, it gets better!

It seems that the author of the article, has deliberately started using the 2001-2008 HadCRUT3 data set with one of the hottest months in this period - which happens to be January 2002 (so he didn't use any 2001 data after all, despite the caption of the graph) - and then ended using the data with the very coldest month in this period, which was the abnormally cold January 2008:

With such a self-selected data set to confirm his bias, is it any wonder that he got a significant cooling trend?

And what's with using only six years and one month to extrapolate a "cooling per decade" value - especially when the same data set goes back for far more than a decade and a real value could be easily calculated?

No wonder that this guy apparently doesn't have any peer-reviewed papers to his name - with such blatant attempts at cooking the data, the reviewers would laugh him out of town.

So what would temperature trends over the last decade actually look like?

Well, I've used the same data set for the period from April 1998 to March 2008 (the last entry), and the following graph is the result:

The linear fit produced a warming of 0.0349044 °C for the entire decade - not much, but as this long-term graph generated from the same data set shows, 1998 was an abnormally warm year while the last winter was particularly harsh:

To sum it up, global temperatures have indeed increased during the last decade, if not as strongly as in the time before that. We will have to continue to watch the long-term trends of global temperatures - and be wary of anyone who attempts to cook the data for his own agenda.


( 29 comments — Leave a comment )
Apr. 16th, 2008 04:16 pm (UTC)
Just for comparison, what would the change per year (or per decade, if you like) be with the full HadCRUT3 dataset?
Apr. 16th, 2008 04:50 pm (UTC)
Top work, sir.
Apr. 16th, 2008 06:56 pm (UTC)
The UN claimed the cooling, not only Monckton
Monckton was bouncing around widely published reports of cooling in certain time frames. You may be correct that using certain time intervals you can get warming instead of cooling, but that sounds like your cooking the books too, self selecting data to make it appear warmer.

More importantly, this trend was not predicted by the IPCC climate models that are utterly failing to reproduce themselves in the empirical realm where they can be tested and falsified. At some point intelligent people will ask "where's the dice" and demand to run a falsification test.
Apr. 16th, 2008 07:32 pm (UTC)
Re: The UN claimed the cooling, not only Monckton
Monckton was bouncing around widely published reports of cooling in certain time frames.

What are your sources for this? Peer-reviewed sources, preferably.

You may be correct that using certain time intervals you can get warming instead of cooling, but that sounds like your cooking the books too, self selecting data to make it appear warmer.

Actually, you've got it backwards - you have to limit yourself to fairly short periods to get cooling trends. Take a look at this graph again:

The overall trend for global temperatures is quite definitely upwards. And for my ten year calculation, I actually used one of the periods with the least possible warming, since 1998 was an abnormally warm year (as the above image shows) and the last winter was especially cold. So no, I didn't cook anything.

More importantly, this trend was not predicted by the IPCC climate models that are utterly failing to reproduce themselves in the empirical realm where they can be tested and falsified. At some point intelligent people will ask "where's the dice" and demand to run a falsification test.

It's far too early to say that the IPCC models are incorrect - after all, these are multi-decade predictions, not annual predictions.

This is a graph for the various models used in the 2001 predictions:

Those vertical lines on the right side are error bars, which means that each of the used models can have a fairly large range of possible end temperatures for the year 2001. And the intervening periods can also be subject to large fluctuations - but ultimately, the long-term trend is what is important.

There have always been abnormally hot summers. And there have always been abnormally cold winters. But in the long run, they tend to cancel each other out. It's the long-term average temperatures that truly describe the state the climate of the world is in.
May. 11th, 2008 09:43 am (UTC)
Re: The UN claimed the cooling, not only Monckton
Why do you choose to claim that 160 years is a "long period" reflecting some kind of "overall" trend in the "long-term"? If you take any natural history book, you will immediately notice that such a period is not long in any serious way.

You surely know that the only reason why these graphs usually start around 1850 is that thermometers were invented and record-keeping started seriously in Europe and America. All peer reviewers should laugh at the selection of this time frame.

Monckton is stretching a little bit, as are other skeptics that claim the Earth is cooling. However, the fact is that significant warming has stopped around the turn of the century, either 1998 or late 2001. But this is a short term observation and may turn out to wrong.

The models predict the rapid warming of 1976-1998 to continue, because warming has obviously been embedded into their code, for example with an overestimated sensitivity to CO2 increase. No surprise there.
Apr. 16th, 2008 07:30 pm (UTC)
Fascinating example of what I knew was going on - probably on both sides. I have always felt the most compelling evidence was in ice coverage, and ocean tempuratures - not just weather, and felt that at least 20 years was necessary for statistics - not that I have such data mind you...

I do believe that we are in a global warming trend - just don't claim to have evidence; I also believe that we can not at this point prove man's fault or lack there of. I do however feel that we need to worry about it - as if the ice is melting (as it seems to be) then we are going to have major changes in land mass, coastlines, and various local climates - for example, here in the Pacific NorthWest (Coastal Washington State) if the Jet stream changes - we are in for a serious shock, and quite possibly colder weather...
Apr. 16th, 2008 07:34 pm (UTC)
Funnily enough, the guy I had the original argument with showed me this graph as "proof" that the sea ice coverage hadn't shrunk in the last decade - when, in fact, it shows quite the opposite...

So yes, sea ice does seem to recede as of late.
Jul. 6th, 2008 12:07 am (UTC)
I can't believe you think 160 years presents a "long term trend" haha... That's absolutely ridiculous. Please correct me if I'm wrong, but take a look at this graph..

Then you can argue "long term" trends

Additionally, during the period of human emissions increasing at their highest rate, the earth was... cooling? That doesn't sound like the correlation that is being preached by scientists like yourself. You can see it for yourself here, between the periods of 1940-1965
Apr. 16th, 2008 10:34 pm (UTC)
Is there data from 1950s to now? I'd like to see a century-long graph, if it's possible. Ten years means nothing in geologic/meteorological time.
Apr. 16th, 2008 10:46 pm (UTC)
Well, the last graph does show earlier values, which are included in the data source - although admittedly early measurements have a much higher margin of error (which is also explained in the original data set).
Apr. 17th, 2008 12:45 am (UTC)
Oops! I didn't examine the later graphs as closely as the first couple. Missed the change in date at the bottom.
Apr. 16th, 2008 11:39 pm (UTC)
Wow thank you so much for doing all this work, even it it was for an internet argument. Do you mind if I steal some of these graphs if I'm sure to give you credit for them?
Apr. 16th, 2008 11:46 pm (UTC)
Go ahead.
Apr. 17th, 2008 12:48 am (UTC)
Here's something that will drive you nuts. I'd think Germany would WANT global warming. Just imagine if your climate changed to that of Italy. ;)
Apr. 17th, 2008 01:06 am (UTC)
Well, then you'd have Germans complaining about the lack of snow in the Winter - which is in fact already happening. Plus it'd wreak all sorts of havoc with the flora and fauna.

Though yes, it probably makes sense to make some long-term investment into Baltic seaside properties...
Apr. 17th, 2008 10:12 pm (UTC)
Not tooooo close to the sea though.
Apr. 17th, 2008 12:49 am (UTC)
Monckton recently presented a paper making somewhat different claims; that there was warming, but not as much, as reported here. As the article described it,

"When they applied their revised factor to the effect of greenhouse gases, the temperature rise was about a third of that predicted by the IPCC."

So in the paper Jurgen quotes, Monckton claims there's global cooling happening; in a more recent paper, he claims there's warming but only a third as much as stated.

That's the problem with the denialists, they can't even keep their stories straight.
"There's cooling!"
"There's warming, but not much!"
"There's warming, but it's nothing to do with us!"
"There's warming, it's to do with us, but it'll be good for us!"

Regarding the recent paper, I recently had to write something,

The paper mentioned is by three blokes. All the information below is in the public domain, easily found by half an hour of googling and using your common sense. sourcewatch is a good site for that.

Viscount Christopher Monckton is a "retired business analyst", degree in Classics and diploma in journalism. He used to work for Thatcher. He was a bit brassed off some years back when they reformed the House of Lords, and all but some 100 I think had to be elected, he didn't get in so he's decided he doesn't like the new system. He once said that everyone with AIDS should be locked up. On climate change he babbles about the fucking hockey stick, claims it's solar forcing that does most global warming (which is funny since solar irradiance has declined over the past few decades) but the world isn't warming anyway, and even if it is it'll be good for us, and also he talks about this blackbody radiation equation which he reckons they didn't apply right. Yes, he is a bit muddled about what he believes.

Monckton's a member of the "Science and Public Policy Institute", run by this loon Robert Ferguson, ExxonMobil gives Ferguson $50,000 a year to run it.

Dr Vincent Gray is this old old kiwi bloke, at least he has a degree in Physical Chemistry. He's one of the founders of the NZ Climate Science Coalition (AGW deniers). They're buddies with that old fraud and friend of tobacco Dr Fred Singer, and the Lavoisier Group. I'm not sure who they're funded by.

Dr David Evans is a member of the Lavoisier Group, which was founded by Ray Evans an exec from Western Mining Corp. Like Fred Singer, Evans is a PhD in electrical engineering, though he did his MSc in maths. Evans used to work for the Australian Greenhouse Office, of all places. He worked on modelling carbon uptake in forests. It was his model which told the Aussie PM that Australia could meet its Kyoto obligations just by stopping Queensland landclearing. He says he quit "for personal reasons unrelated to my views on global warming." I think he's on a couple of mining company boards, but on a casual websearch it's hard to tell as it's not exactly an uncommon name.

Also mentioned in the article is Roy Spencer, a guy who works on microwave satellites. He's a member of the Heartland Institute, which is funded by Phillip Morris to help them in showing the world how safe cigarettes are. They've had the better part of a million bucks from Exxon-Mobil. He's also a member of the George C. Marshall Institute, which receives funding from Exxon-Mobil and defence contractors.

So with this paper challenging the level of climate change, we're seeing four different denier groups get together, which is both interesting and scary.
Dec. 22nd, 2009 05:56 am (UTC)
Climategate emails
First, there is no scientific consensus. Over 30,000 scientists in the U.S. alone signed a petitition. The man-made global warming movement has little credibility due to the lies and manipulation of data being found over and over again. The nail in the coffin was the climategate email scam. Lies lies and more lies. The media is biased. It is a political movement with a dark history. Study history and politics and you will be surprised at what you learn. Maurice Strong, Henry Kissinger, Obama et al aren't actually nice people. All the best.
Apr. 18th, 2008 05:29 pm (UTC)
A few questions
Is "global temperature" the same thing as "the least-squares best fit linear trend to the global temperature"? Which have you tested claims about?

On what basis is a linear trend plus noise taken as the appropriate statistical model? Is the data really likely to be linear plus iid Normal errors, as using this statistic would suggest?

In the final graphs, why are the areas under the curve shaded red and blue? Does the area mean something? What smoothing have you used for the black lines, and why is it appropriate?

Given that the L-S linear trend you get varies so much with the selection of end points, how much meaning can be assigned to any of them? Is any answer right or wrong?

When the author wrote the note and constructed the graph, were March 08's numbers out yet?

Aug. 2nd, 2008 03:30 pm (UTC)
Re: A few questions
Re: "A few questions" (Anonymous)

Global average temperature since about 1975 is indistinguishable from a linear trend plus random noise. The noise is not iid, it's autocorrelated, and it appears not to be normally distributed. But that is *not* an implication of using linear least-squares regression, that analysis gives an unbiased estimate of the trend rate even when the random part of the data is *not* iid normal.

The final graph comes from HadCRU, not from Jurgen, so if you want to know what smoothing method is used you'll have to ask them.

To attach "meaning" to trends from L-S regression, one has to compute the probable error of the trend rate. Doing so requires compensating for the fact that the random part of the data is *not* iid, rather it's autocorrelated, which inflates the probable errors. In most geophysical analyses, the random process is approximated as an "AR1" process, but for global temperature it's demonstrably *not* AR1, and the probable error in a trend estimate from least-squares regression is even greater than the estimate from assuming an AR1 model.

I don't know whether or not March 2008 data were available when the erroneous graph was produced. But certainly 2001 data were available. And just as certainly, -0.1877124 is not even approximately -0.4.
Apr. 19th, 2008 01:08 am (UTC)
If you want a multi-century timeframe...
just look at the IPCC own graph

Apr. 19th, 2008 01:24 am (UTC)
The source for Monckton's

I am sure you have seen this graph already. It does look like the source for the -0.4C/decade claim by Monckton


It's figure 2 in http://wattsupwiththat.wordpress.com/2008/03/13/to-tell-the-truth-will-the-real-global-average-temperature-trend-please-rise-part-2

If you read that blog, it explains why that particular trend is chosen to start at that particular month/year, using that particular data set. So it appears that rather than concentrating on Monckton, you should raise your questions to Basil Copeland (and Anthony Watts)
Aug. 2nd, 2008 08:46 pm (UTC)
Re: The source for Monckton's
Copeland/Watts post is as questionable as virtually everything else on Watt's blog.

The reason it's questionable in this case is that they're fretting over weather, and drawing tiny trend lines based on that weather. Their justification for selecting 2002 as a break point is still weather, and their game with "cooling" is predicated upon not knowing what happens next (i.e. in the future).

J Hubert remains correct in pointing out the cherry picking of Mr Monckton's start-date. When someone who knows what they're doing (Tamino) tackles the issue on the appropriate climatological timescale no break in trend is found that has no precedence since 1975.

Tamino's blog:
"Global Temperature from GISS, NCDC, HadCRU."
And the item germaine to this issue, graoph :
That's a graph of the residuals, the difference between trend and observation.

The trend used is a climatological timescale (30yrs), not weather. And as the graph of residuals shows nothing untoward for recent years we can conclude that talk of a significant cooling departure from that trend is unfounded.

Nov. 29th, 2009 11:46 am (UTC)
Embrace Global Warming
Be thankful the media is properly influenced and has convinced the population of the dangers AGW poses. Debate all you like but we WILL have cap-and-trade. Those who have invested shrewdly will prosper. Those who fight the system will not. We should all thank the IPCC and Al Gore for this historic opportunity.
Dec. 22nd, 2009 06:01 am (UTC)
Re: Embrace Global Warming
Are you serious? Al Gore is a scammer. Even if there were a consensus on man-made warming, their solutions are terrible i.e. tax, more govt control. Read history and the statements, quotes, U.N. and Club of Rome white papers etc. You will very soon find that this is a manufactured crisis, as they stated in their own words, to control society and enslave us with saving the planet as the excuse. Seriously, research the history, politics. Climategate email scam is a good starting point in exposing the fraud. One world government is not good, unless you love tyranny/slavery. All the best.
Dec. 9th, 2011 09:32 pm (UTC)
Arguments Against Global Warming

Even though there are so many talks and forums that have been given to teach the people about global warming, there are still so many arguments about global warming that have arisen over the years. Most of the arguments about global warming are not facts and are just speculations of people or their assumptions about what they think is happening to the world.
Dec. 10th, 2011 08:43 am (UTC)
Re: http://argumentsagainstglobalwarming.org/
That website starts with:

One of the arguments about global warming is that the world is considerably becoming cooler ever since 2010.

That's what they consider a convincing argument? A temperature fluctuation over the span of one year?

Sorry, but if they start with this, I just can't take them seriously...
Sep. 1st, 2014 03:31 pm (UTC)
Re: http://argumentsagainstglobalwarming.org/
It is now in late 2014 and 6 years since this post was written.

Every major scientific body (NASA, WMO, CRU etc) now admits that there has been no statistically significant change in global surface temperatures since 1997 and NONE OF THE MODELS that every single IPCC report has been based on came close to predicting that. ("ahh, but the heat is going into the oceans" they say. Well why wasn't that modelled in the first place?)

How long until people at least realise we don't have a clue what's going on with the weather, let alone the global climate 100 in the future?
Zane Yerby
Jan. 13th, 2012 05:05 am (UTC)
If you read on into that article, you will find that it "supports" global warming, by addressing what the author considers common arguments against global warming theory. The site credits no sources and has the appearance of an amateur pet project. And earlier, "Be thankful the media is properly influenced and has convinced the population of the dangers AGW poses. Debate all you like but we WILL have cap-and-trade. Those who have invested shrewdly will prosper. Those who fight the system will not. We should all thank the IPCC and Al Gore for this historic opportunity." what kind of leftist, liberal, word-barf is that? I am still searching for an unbiased collection of scientific facts, so I can make a clear decision for myself. If anyone knows of such a source, please enlighten me.
( 29 comments — Leave a comment )


The Standard
Jürgen Hubert
Arcana Wiki

Latest Month

August 2011


Powered by LiveJournal.com
Designed by Jared MacPherson