Daily Mail spreading climate change misinformation again

Note: This piece has been updated on 15th December

The Daily Mail seems to have found yet more evidence to persuade its readers that they should be skeptical of man-made global warming. This piece by David Rose has two startling revelations – that both proxy data and an important diagram in an IPCC report were manipulated to make past temperatures appear cooler than they actually were, and that weather station data was also manipulated to show warming in recent times which may not have actually occurred. I don’t have time now to address the latter claim, but you can see a good summary here. However, I would like to address the question of the temperature data in some detail.

Firstly, just to muddy the waters, Rose starts with a piece of boilerplate denialist nonsense

He and other ‘global warmists’ often insist that when it comes to the IPCC’s main conclusions – that the Earth is in a period of potentially catastrophic warming and that the main culprit is man-made greenhouse gas emission – no serious scientist dissents from the conventional view.

In fact, there is a large body of highly-respected academic experts who fiercely contest this thesis: people such as Richard Lindzen, Professor of Meteorology at the Massachusetts Institute of Technology and a disillusioned former IPCC member, and Dr Tom Segalstad, head of geology at Oslo University,

Well that’s two, one of whom is actually a geologist not a climate scientist – even I could do better than that. Of course there are others, just not that many of them who are actual experts in the field of climate science (see here). The fact is that the vast majority of scientists working in climate science as well as every major scientific body in the world accepts that the primary cause of climate change is almost certainly human activity, and this is supported by the vast majority of the peer-reviewed scientific literature. If there really was a genuine and significant body of expert opinion and published science to the contrary then skeptics would not have to resort to publishing bogus lists to “prove” their case. Maybe the current scientific consensus will be proved wrong, but it is misleading to say that it does not exist and that there is still any genuine uncertainty on the basic question of whether humans are mainly responsible for GW.

Rose then moves on to the hacked CRU emails and the infamous one mentioning “Mike’s nature trick”, and reveals that

the full context of that ‘trick’ email, as shown by a new and until now unreported analysis by the Canadian climate statistician Steve McIntyre, is extremely troubling.

Now those of us who are interested in the arguments around climate change will have immediately sat up at the mention of McIntyre. He is a semi-retired mathmetician who formerly worked in the mining industry and has never actually worked in the field of climate science, but he runs the Climate Audit blog and is best known for his attacks on the “hockey stick” reconstruction of past temperatures, the author of which was Michael Mann – the “Mike” referred to in the abovementioned email. So whatever one thinks of McIntyre he can hardly be considered a disinterested observer on this subject. However, Rose appears to have taken his views entirely at face value.

Anyway, to continue, this new “controversy” in fact concerns the “hockey stick” diagram in the ‘Summary for Policymakers’ of the 2001 IPCC report, which shows Mann’s reconstruction alongside two others, by CRU scientists Phil Jones and Keith Briffa as well as instrumental records (ie actual measured temperatures) for the 20th Century. 

Chart as it appears in IPCC report

Rose explains how the non-existence of such instrumental records for past centuries make it necessary to use “proxies” such as tree rings to estimate past temperature changes before throwing in another piece of misdirection

However, different proxies give very different results.

For example, some suggest that the ‘medieval warm period’, the 350-year era that started around 1000, when red wine grapes flourished in southern England and the Vikings tilled now-frozen farms in Greenland, was considerably warmer than even 1998.

Of course, this is inconvenient to climate change believers because there were no cars or factories pumping out greenhouse gases in 1000AD – yet the Earth still warmed.

Here he manages both a factual error and a non-sequiteur in one go. No credible temperature reconstructions show the Mediaeval Warm Period to be warmer than or even as warm as today (see here) and even if they did it would not have any bearing as to what is causing the warming we are seeing today as no one denies that factors other than increased CO2 levels have influenced climate in the past.

Now on to those emails. Rose summarises them thus –

In September 1999, Jones’s IPCC colleague Michael Mann of Penn State University in America – who is now also the subject of an official investigation –was working with Jones on the hockey stick. As they debated which data to use, they discussed a long tree-ring analysis carried out by Keith Briffa.

Briffa knew exactly why they wanted it, writing in an email on September 22: ‘I know there is pressure to present a nice tidy story as regards “apparent unprecedented warming in a thousand years or more”.’ But his conscience was troubled. ‘In reality the situation is not quite so simple – I believe that the recent warmth was probably matched about 1,000 years ago.’

Another British scientist – Chris Folland of the Met Office’s Hadley Centre – wrote the same day that using Briffa’s data might be awkward, because it suggested the past was too warm. This, he lamented, ‘dilutes the message rather significantly’.

Over the next few days, Briffa, Jones, Folland and Mann emailed each other furiously. Mann was fearful that if Briffa’s trees made the IPCC diagram, ‘the sceptics [would] have a field day casting doubt on our ability to understand the factors that influence these estimates and, thus, can undermine faith [in them] – I don’t think that doubt is scientifically justified, and I’d hate to be the one to have to give it fodder!’

Finally, Briffa changed the way he computed his data and submitted a revised version. This brought his work into line for earlier centuries, and ‘cooled’ them significantly. But alas, it created another, potentially even more serious, problem.

According to his tree rings, the period since 1960 had not seen a steep rise in temperature, as actual temperature readings showed – but a large and steady decline, so calling into question the accuracy of the earlier data derived from tree rings.

The implication is clear – Briffa deliberately manipulated his data to hide the fact it showed warmer temperatures in order to distort the message shown in the diagram in the IPCC report – a clear example of distorting the science to reach as desired conclusion. Also, in manipulating the data he inconveniently created a post 1960 decline in temperatures.

At this point the reader may want to read the emails in full (the first three in the chain anyway, which include those quoted above) – they can be seen here. This gives a rather different picture –

The first is the one sent by Folland

A proxy diagram of temperature change is a clear favourite for the Policy Makers summary. But the current diagram with the tree ring only data somewhat contradicts the multiproxy curve [Mann’s & Jones’s] and dilutes the message rather significantly. We want the truth. Mike thinks it lies nearer his result (which seems in accord with what we know about worldwide mountain glaciers and, less clearly, suspect about solar variations). The tree ring results [Briffa’s] may still suffer from lack of multicentury time scale variance. This is probably the most important issue to resolve in Chapter 2 at present.

What we can see here is that Briffa’s original reconstruction diverges quite significantly from the others (see below) and did in fact show warmer temperatures. Clearly when trying to portray as accurate as possible a picture of past temperatures such a divergence is going to be a concern but the priority here is plainly to get the most reliable and accurate result (what else does “we want the truth” mean) rather than covering up the truth “because it suggested the past was too warm”.

Chart showing Briffa\'s original data in yellow

Then there is the reply from Briffa. Apart from the short passages quoted by Rose there is some discussion of the technicalities of using tree rings as proxies and of comparing his and Mann’s results. But there is also this –

The latest tree-ring density curve ( i.e. our data that have been processed to retain low
frequency information) shows more similarity to the other two series- as do a number of other lower resolution data ( Bradley et al, Peck et al ., and new Crowley series – see our recent Science piece)

So it is clear that he has already made his recalculations prior to this discussion – in fact as alluded to he had already published the new results in Science magazine earlier that year. So all that was done is that the older reconstruction was replaced by a more up to date version, which also happened to be a better match for the other recontructions.

So what about the sudden appearance of the post-1960 cooling in the new data? Well it wasn’t so sudden at all – this “divergence problem” was already well known and had been mentioned in a paper Briffa published the previous year, and he warned against using the post 1960 data in any reconstruction.

And what about Briffa’s other comments mentioned by Rose.

Briffa knew exactly why they wanted it, writing in an email on September 22: ‘I know there is pressure to present a nice tidy story as regards “apparent unprecedented warming in a thousand years or more”.’ But his conscience was troubled. ‘In reality the situation is not quite so simple – I believe that the recent warmth was probably matched about 1,000 years ago.’

What we have here is an admission that sometimes science can be a bit messy and give a picture which is not as clear as scientists would like, hardly a revelation, and he gives a personal opinion on temperatures 1,000 years ago not supported by actual evidence (his reconstruction only goes back to 1400). There is no sign of any “troubled conscience” over this – it is simply a statement of an opinion which his colleagues obviously do not share.

So Rose has blatantly misinterpreted the emails, either deliberately or by simply parroting McIntyre’s arguments without bothering to check for himself. Either way it doesn’t look good.
But there’s more – the “deception” only started with the “manipulation of data”.

This is the context in which, seven weeks later, Jones presented his ‘trick’ – as simple as it was deceptive.

All he had to do was cut off Briffa’s inconvenient data at the point where the decline started, in 1961, and replace it with actual temperature readings, which showed an increase.

On the hockey stick graph, his line is abruptly terminated – but the end of the line is obscured by the other lines.

The Mail provides a handy close up to show this last point

Close up of 20th Century data

but all it demonstrates is that you have to look a bit more closely as the green line (Briffa’s) ends at a point where the other lines cut across it. It is hardly hiding the fact that the data ends in 1961. And it doesn’t “replace” the missing period with actual temperature readings, they are shown independently (the red line) going back to the late 19th century. It’s not “artfully hidden”, you just have to look at it properly. The rationale for adding the instrumental data is obvious – the point of using three different reconstructions is that none of them can be claim to the sole “true” record but together they give a good overall picture, and by adding in actual as opposed to reconstructed temperatures for the period when we have them can only improve that picture. The point is to use the best data we have for any period in order to give the best possible representation of temperature changes over time – it would be perverse to exclude instrumental data from that. To claim that somehow the picture is distorted because of the omission of data known to be incorrect is nonsense and to portray this as an attempt to deceive is itself mendacious idiocy.

There is a more detailed and technical account at Deep Climate which I would recommend.

Update: 15.12.09
Unity at Liberal Conspiracy has a good piece on this, which also goes into the Darwin weather station issue in some detail. He points out that Rose misquotes Roger Pielke Jr. in his piece and quotes the following correction sent by Pielke to Rose

Dear David-

I just saw your story in the Daily Mail and a small correction is needed. You quote me as saying:

“These emails open up the possibility that big scientific questions we’ve regarded as settled may need another look.”

What I said was:

“While these emails open up the possibility that some scientific questions we’ve regarded as settled may need another look, time will tell and the implications for science are not the most important aspect of the emails.”

The point was that while I am agnostic about the implications for science, leaving that to others, I am certain that the emails have broader implications for the credibility and legitimacy of certain quarters of climate science. Based on what I’ve seen, I do not believe that any “big scientific questions” are implicated by the emails.

Also, after writing the original post I realised that the “hide the decline” email does not of course refer to the IPCC report mentioned by Rose but to the “WMO Statement on the Status of the Global Climate in 1999″ report. The graph contained therein does merge the instrumental data and the proxy data post-1960 so maybe this is what Rose is referring to, and just maybe there are grounds for criticism for doing it that way but I don’t accept that it creates a misleading picture – it still gives the best estimate for temperatures over the given period based on the best data available. And Rose is still clearly misleading his readers in telling them that the “hide the decline” email refers to the IPCC report.

Furthermore keen eyed readers will have noticed that I claimed the version of Briffa’s reconstruction used in the final graph was the one published in Science in May 1999 but in the graph itself it mentions “Briffa 2000″. My understanding is that the 2000 reconstruction is a refinement of the 1999 one and is not substantially different.

Finally, I have tried to tweak the formatting to improve the formatting and make the graphs bigger but WordPress is not being very helpful. Will try again when I get a chance.

One thought on “Daily Mail spreading climate change misinformation again

  1. Pingback: The Mail on Sunday’s David Rose keeps writing rubbish about climate change | BirchIndigo

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>