Monday, January 28, 2013

Dot Earth Blog: When Publicity Precedes Peer Review in Climate Science

Last year, after opponents of hydraulic fracturing made much of an unpublished paper by a doctoral candidate in economics who reported finding health impacts in infants from nearby gas drilling operations, I wrote a piece titled ?When Publicity Precedes Peer Review in the Fight Over Gas Impacts.?

It?s time for the global warming version, in two parts.

Here in part one, I offer an update on events related to a?news release* issued last week by the Research Council of Norway?with this provocative title: ?Global warming less extreme than feared??

As?I wrote over the weekend, the release described new research finding that global warming from the buildup of greenhouse gases was likely to be on the low end of?the persistently wide spread of projections?by other research groups.

My concern was that the findings did not yet appear to have passed peer review and been accepted for publication. There was only one published paper that seemed related, and the news release ? while enthusiastically disseminated by some blogs and media ? didn?t specify whether the findings it described had been published.

This incident prompted me to create the warning label above, which I?ll use in posts on this kind of problem going forward.

With the help of Twitter, I got some initial answers over the weekend from climate scientists in Sweden and Norway. Now, I?ve gotten a heap of troubling input from the Research Council of Norway and some Norwegian climate scientists involved with the work. This, along with an excellent update from Roz Pidcock at the Carbon Brief blog, reveals yet another case study in how not to publicize science if your goal is to foster understanding and avoid confusion:

From Thomas Evensen, director of communication for the Research Council:

The research project in question was peer-reviewed by a group of international researchers before it was approved for funding in 2008. The news article presents a synthesis of the project and contains both published and thus far unpublished findings. Some of these findings are given in an article that is part of a doctoral thesis.

The Research Council found this research project to be of interest, and decided to publish a news brief upon the conclusion of the project. However, the article also makes it clear that the findings are preliminary at this point, and that the research must first be confirmed by other studies before any impact can be derived.

The news item was published somewhat later in English than in Norwegian purely due to administrative issues relating to translation and publication.

Publications:

Skeie et al., Anthropogenic radiative forcing time series from pre-industrial times until 2010, Atmos. Chem. Phys., 11, 11827?11857, 2011

Magne Aldrin, Marit Holden, Peter Guttorp, Ragnhild Bieltvedt Skeie, Gunnar Myhre and Terje Koren Berntsen, Bayesian estimation of climate sensitivity based on a simple climate model fitted to observations of hemispheric temperatures and global ocean heat content, Environmetrics, 2012.

In review:

Skeie et al. A lower and more constrained estimate of climate sensitivity using updated observations and detailed radiative forcing time series, In review in Journal of Climate.

Of course, review of the research plan in 2008 is not review of the results in 2012, so that point is irrelevant. And putting out a release on a mashup of published and unpublished work is, at best, bound to create confusion and ? at worst ? could undermine the publication prospects for the pending paper. (The young scientist whose doctoral thesis is the basis for the sensitivity paper is Ragnhild Bieltvedt Skeie, a senior research fellow at the Center for International Climate and Environmental Research ? better known as Cicero ? in Oslo.)

Eystein Jansen, research director at the Bjerknes Centre for Climate Research in Bergen, Norway (who?d been a helpful Twitter contact on this) added helpful context on the research and its implications, building on thoughts expressed in my weekend post by the Swiss climate scientist Reto Knutti (click here for Knutti?s weekend thoughts):

The funding for the studies came from the Research Council of Norway (RCN) and its NORKLIMA research program. At the end of the project the PI (Berntsen) has to submit a final report to the Resarch Council, including a popular science summary. Based on this input RCN has produced a fact-sheet, first in Norwegian and then in English, and made a press release upon completion of these. Normally they would do this when papers are published, but they seem to have been more hasty this time, which to me is unwise both for work that may be controversial and by principle?.

I am in full agreement with Knutti? s remarks on your blog. There are a number of potential issues with this study that needs consideration: Using a very simplified model without key dynamics, potential dependency of results to the last few years of temperature development, deep sea heat storage etc.

We need to narrow down climate sensitivity from multiple lines of evidence, and there are many remaining uncertainties regardless of approach. This study was done by competent people, with experience as lead authors in IPCC AR4 and 5, and is potentially a novel and fresh approach.

But it is way to early to say that this study has any more weight than other studies with low or higher sensitivity. My bet still goes along the 3 degree line as the most plausible, all things considered.

The Carbon Brief post, has more clarification of what was and was not yet published from Magne Aldrin, a scientist at the Norwegian Computing Center who is an author of the 2012 paper on climate sensitivity mentioned in my post:

Dr Magne Aldrin, co-author on the 2012 paper with Terje Berntsen, told Carbon Brief today: ?The results mentioned in the press release by the Research Council of Norway is taken from [a] PhD thesis ? from March 2012 and that part of [the] PhD thesis is not published or accepted for publication in a journal.?

Aldrin told us last December that the group?s newest findings ? an extension of the analysis in the PhD thesis ? were under review with the Journal of Climate. He confirmed with us today that is still the case, adding that the PhD thesis findings should be thought of as ?preliminary? and that the ?final result is not ready yet?.

Pidcock?s conclusion:

[I]t may be that when these new results are eventually published they do suggest a lower figure for climate sensitivity.

But this episode underlines the problems of so-called science by press release. With such a complex and sometimes controversial topic, research findings need to be carefully treated. As with all scientific research, if results are not yet published or peer reviewed, they are worth treating as preliminary.

Yes, indeed!

Part two will look anew at the unfortunate saga of Richard Muller?s Berkeley Earth Surface Temperature project.

4:41 p.m. |Update

* The Norwegian climate scientist Glen Peters said, via Twitter, that the publication from the Research Council was more an ?information sheet? than a news release.

Source: http://dotearth.blogs.nytimes.com/2013/01/28/when-publicity-precedes-peer-review-in-climate-science-part-one/?partner=rss&emc=rss

Early voting results BBC Dick Morris Daily Show provisional ballot npr rush limbaugh

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.