Sunday, January 24, 2010

Miscalculation in the Stern Review

Roger Pielke Jr. reports on his blog about an interesiting, if unexpected entanglement:

There is another important story in involving the Muir-Wood et al. 2006 paper that was misrepresented by the IPCC as showing a linkage between increasing temperatures and rising damages from extreme weather events. The Stern Review Report of the UK government also relied on that paper as the sole basis for its projections of increasing damage from extreme events. In fact as much as 40% of the Stern Reivew projections for the global costs of unmitigated climate change derive from its misuse of the Muir-Wood et al. paper.
The government website which hosts the Stern Review was changed in the meantime (quietly) to correct the data. Problem is that the maths for Stern's cost benefit analysis don't add up any more.

21 comments:

Leigh Jackson said...

That is what Pielke Jr. and Mike Hulme are saying. However, they are not economists. I am neither an economist nor a climate scientist. So I would like to hear Stern respond.

ghost said...

interesting, Reiner.... I assume you have the most posts of all constributors, and almost no original posts. Often you just repeat claims of newspapers and blogs without checking, without commenting. Therefore, you post embarrassing stuff like the "Smoking Gun". Thereby, you favor skeptical to "skeptical" claims.

Did you know that Pielke Sr. blog is hosted by Anthony Watts? Fascinating, isn't it?

Honest Broker? I do not think so.

Hans von Storch said...

Ghost,

is that necessary? - Why would it be interesting that Roger Pielke Sr. blog is run by Anthony Watts? Why should that be fascinating?

I personally find Reiner's post interesting and welcome. And I consider all reader of this blog as equally valuable, even if some are Betonköpfe of both sorts. And if somebody has a bad day, like you today - we will have to live with that.

ghost said...

@Hans

I do not agree, you post things about the business things of Pachauri (if true or not, you could not say), why not posting about the connections of Pielke Sr and Watts? Pielke Sr, for example, attacked NCDC scientists, because the used material of Watts, that he published, to show the USCHN record is credible. Why did he attack them? The paper was good, it described that the Watts surfacestation.org project impressively showed the reliability of the US temperature trend. (Menne et al 2010). Pielke Sr tried to discredit this work with smear without even mentioning the results at all. Do you like that, really? I do not believe you like that.

Reiner did not check Pielke Jrs post at all, he did not express any opinion about it. What is the point of the post? I do not see it. Does Reiner really believe the Stern report was changed secretly...? Really?

The post of Pielke Jr is full of accusations and conspiracy claims... it is not worth to read at all.

you should at least read something about the post from another source: http://rabett.blogspot.com/2010/01/honest-joker.html

plazamoyua said...

Ghost, Pielke Sr blog is hosted by wordpress.com. Just like many other millions of them. Its obvious from the URL:
http://pielkeclimatesci.wordpress.com

Anthony Watts has helped S. McIntyre and R. Pielke, and probably other bloggers, with hosting migration to wordpress.com. So what?

ghost said...

@plazamoyua

you are right, I am wrong. I am sorry.

climatesci.org is registered to U of colorado, hosted at wordpress.com... Watts helped only.

The rest is true.

Yarmy said...

@Ghost
Before reading this post I didn't know that the Stern report had mistakenly overestimated the economic impact of increased hurricane intensity in the US by a factor of 10. That's a huge error for a document that is intended
to advise governments.

As for your Eli Rabbett link, Yarmy would have read it all had Yarmy not quickly realised that it was written entirely in the third person which has the unfortunate dual effect of making the author appear unbearably smug and the content needlessly verbose.

Reiner Grundmann said...

Ghost
from your post I can see two things:

1 you do not like the spreading of information (accusing me of linking to other sites)
2 you do not want readers to form their own opinion ("The post of Pielke Jr is ... not worth to read at all).

As regards your claim that Pielke's post is "full of accusations and conspiracy claims" I would like to see your evidence. What I see is some careful documentation.

BTW: Pachauri's "business things" have been documented and admitted. Pachauri did not sue the British paper that first made the claim, although he at first suggested he might.

P Gosselin said...

Oh what a tangled web we weave, when we first practice to deceive.

It evenetually becomes a tangled web no matter how much one practices.

Hans von Storch said...

Ghost,

I find it not acceptable that you mix the two Pielkes here - Roger jr. is one individual, Roger Sr. another. How is their credibility related to each other?

If you want to discuss murky links of the senior Pielke, have a separate guest posting here and spell out.

Concerning the changing numbers in the Stern-report: I thought the evidence presented was consistent and convincing. But let us see what else comes up in this respect.

A bit relaxed, please.

Marco said...

Since Eli Rabett's story is a bit long and perhaps an unpleasant read, I urge EVERYONE here to actually read the relevant chapter of the Stern report. Apart from the one instance of 1.3%, ALL other references in the 5th chapter state 0.13%.

And then there's the FAQ: http://www.hm-treasury.gov.uk/sternreview_faqs.htm
which state:

" Q. You state the cost of US hurricanes at temperatures of 3°C above pre-industrial levels as 0.13% and 1.3% of US GDP in different places in the report. Which is correct?

A. The correct figure is 0.13%. There is an error in Chapter 5, pg. 139, which cites the cost as 1.3%. An Errata page will be published to cover this and any other typographical errors. "

Apparently this is quietly altering things...

Yarmy said...

@Marco
Yup, just checked and you are correct: 3 other references to 0.13% in the archived version. Not really a big deal after all then.

Hans von Storch said...

So - the solution is rather undramatic: The original report contained a typo, which now has been corrected.

Often there is a surprisingly simple solution to a phenomenon, which looks very suspicious.

Anonymous said...

>>HvS: So - the solution is rather undramatic: The original report contained a typo, which now has been corrected.<<

No, that's not yet the end of the story, I even have to think it is an all too transparent "Ablenkungsmanöver", at least that is what I deduce from RP Jr.'s site.
http://rogerpielkejr.blogspot.com/2010/01/what-tangled-web-we-weave.html
Quote "The issue is much deeper than a typo -- you can seen in my excerpt from my paper above that I had already assumed that it was a typo. The problem is that once the typo is corrected it then reveals that the numbers presented by Stern just do not add up."

Roger Pielke, Jr. said...

Marco-

The FAQ that you cite is mistaken due to the report being quietly changed. There is no such "error" in Chapter 5 because it has now been changed from the original.

If I say that 3 + 11 ~ 15, and then realize that the 11 is actually supposed to be 1.1, simply writing that 3 + 1.1 ~ 15 does not fix the problem, simply the typo.

That is what happened here.

Those with interest to know the details can see them here:

http://miha.ef.uni-lj.si/_dokumenti3plus2/190025/3clanek2.pdf

Marco said...

Roger, I find your comment rather disingenious. Yes, they corrected the error of 1.3%. They put the correction in the FAQ. However, the old version had ONE 1.3%, and the remainder in the chapter (I think 3 in total) said 0.13%. You knew that, your article shows you did. Be honest.

Roger Pielke, Jr. said...

Marco-
There are two issues:

1. Faulty math, which altering a typo does not fix.

2. Quiet revision, which is simply bad practice. If I make an error in a scientific paper (which I have done) I must submit an errata or corrigendum to have it fixed, so that the record is clear. Scientific journals are quite clear on this. Why should the Stern Review be held to lesser standards?

Issue #2 is poor practice, issue #1 is bad science.

Reasonable people can disagree about #2, and I take it that you do, good for you. I (and apparently others) think otherwise. #1 is unambiguous.

Marco said...

@Roger:
You COULD have seen that the FAQ notes an errata is in preparation...

Regarding point 1: let's see what time brings.

Hans von Storch said...

Typo or inconsistent math -obviously we have to have a closer look. I stand corrected-

Roger Pielke, Jr. said...

Marco-

The errata has been 2+ years in preparation since the change was made. That seems a bit long, no?

Roger Pielke, Jr. said...

Also, for those interested in details, I left the following in response to Richard Tol at my blog:

1. The global estimate (which in Table 5.2 is supposed to be developed countries) derives from Muir-Wood et al. (2006).

2. The 2%/year increase 1970-2005 in Muir-Wood et al. and used by Stern (as a lower bound) is based entirely on US hurricanes.

3. So the Nordhaus numbers and Stern global (Muir-Wood) are actually two different measures of the same thing. They are thus _not_ additive as you have suggested.

4. At best they serve as an order of magnitude consistency check. You are correct that the 1.3% number was not used in the C?B calculations, those came from MW. However, had it been properly recorded as 0.13% there would have been a fairly obvious problem with the numbers to anyone who looked.

5. When the typo is corrected they order-of-magnitude check fails.

6. Stern's upper bound on global damages results from accelerating the Muir-Wood 1970-2005 trend by and additional 1% per decade (ie., the MW 2% becomes 3% after a decade, then 4% etc.). There is no basis for this.

Finally, changing a text after publication is not allowed in scientific journals where errata and corrigenda are required. Stern should be held to similar standards.