More BEST (but still not peer reviewed)

The BEST story rumbles on but still no peer reviewed results.

Instead, they’ve made 4 manuscripts available that have been submitted on their methods, the influence of the urban heat island, temperature records and stations quality for the US and global temperature variations (all 4 links go to pdfs).

This seems to be the key figure:

"Comparison of the Berkeley Average to existing land-only averages reported by the three major temperature groups." The differences in the late 20th Century arise from different definitions of "land" used by the four groups - the paper says that global averages match better in this period but I couldn't see that figure in the papers - if it's not there, it'd be nice to add it.

Here are a few interesting quotes from the papers after a very quick read. From the first paper:

This change [in global land mean temperature] is consistent with global land-surface warming results previously reported, but with reduced uncertainty.

I’d read this as GISS, NOAA and HadCRU being pretty much on the money but they’ve found the same thing using more data (the BEST record goes back to 1800, which is nice) and a different method.

From the second paper:

The small size, and its negative sign, [of the urban heat island effect] supports the key conclusion of prior groups that urban warming does not unduly bias estimates of recent global temperature change.

This pretty much confirms the recent work of Menne et al. (2010) and the Watts paper and will hopefully put this issue to bed. Indeed, the third paper, which specifically mentions Watts in the abstract in relation to SurfaceStations, then goes on to show that US station quality makes little impact on the recorded trend.

Paper four looks at the role of ENSO and the AMO in controlling decadal variability in the temperature data. This looks like the most interesting paper to me so I’ll have a closer look at that soon.

Overall, though, there doesn’t quite seem like enough material for 4 papers here. Maybe they’re trying to make it look like they’ve done more than re-re-re-confirm the results of other groups.

Dear Institute of Physics… (Part II)

Following my statement in March this year that I would leave the IoP if they didn’t withdraw their evidence statement from the House of Commons Science and Technology Committee investigation into the CRU, here’s the result…

Dear Institute of Physics

Thank you for my recent reminder to renew my membership.  Unfortunately I no longer wish to be associated with the IoP.

My concern goes back to March this year when I contacted the Institute regarding their evidence submission to a House of Commons Science and Technology Committee investigation into the work of the Climatic Research Unit at the University of East Anglia.

The IoP evidence submission was particularly one sided in its analysis and engaged in wild speculation.  It appeared to have an agenda to undermine the work of the CRU without supplying any evidence to substantiate its claims.  This is clearly irresponsible and inappropriate behaviour from a professional society.  [My quick analysis of it is here.]

The response from the IoP to my concern about this evidence submission was also well below what I expected.  The only specific response [which I can’t find on the IoP website now but is reproduced at the bottom of my letter from March 2010 here] was an anonymous note from a member of the board that issued the submission saying that we should be “relaxed” about the process by which the statement was written (despite the fact that “working scientists just don’t have time to [produce the first draft of the statement]”.  If any other responses were issued I suspect I missed them as I had no further communication from the Institute.

The anonymous statement, and refusal to disclose who had actually written the submission, was particularly ironic as the evidence statement was calling for greater openness from the scientists it was criticising!  Furthermore, IoP journals do not even require the level of data “openness” that the submission criticised CRU for not following.

In the end, the HoC S&TC report was very supportive of the CRU so at least your evidence submission had no impact in that respect.  However, the IoP still set a precedent that it is willing to openly criticise the work of scientists without providing any evidence to warrant such an attack.

I am sad to leave the IoP as I have been inspired by some of your work and recognise what you do to promote physics.  However, following this episode, I can no longer support the Institute.

Your faithfully,

Andrew Russell

[In the IoP’s defence, I did have some private conversations with a few people from the Institute and they were aware that there was a problem here to be dealt with.  I have no idea if anything was done as they were obviously keen to deal with it in private.  However, I still feel that the IoP should have more open about their efforts to get things in order and that a mistake had been made.]

Another “climategate” inquiry…

This is starting to get a bit repetitive: another inquiry clears the CRU scientists.  This time the report came from Lord Oxburgh’s Science Assessment Panel.

The main criticism from this report was regarding the level of collaboration between CRU and statisticians in relation to the integration of different climate datasets.

New Scientist got a bit carried away and ran with the headline “Climategate scientists chastised over statistics” despite the report saying that “it is not clear, however, that better [statistical] methods would have produced significantly different results” and “in the CRU papers that we examined we did not come across any inappropriate usage although the methods they used may not have been the best for the purpose“.

If this is what New Scientist are referring to then this is pretty weak chastisement.

But why not go further?

Others have called for professional software engineers to develop the routines CRU use to industry standards.  How about some editors to help with writing their papers?  The Oxburgh report also pointed out that CRU were quite disorganised so let’s get in some management consultants as well.  They could do with a drawing office to help draft figures.  Professional archivists would be useful to keep track of all their data and they could use some people to deal with all the FoI requests and media interest.  I suspect that some social scientists and politics scholars would help focus their research on the needs of society and policy makers.

Clearly this would all be too expensive for a small research group.

As it is, CRU have 3 permanent staff and the nature of academic research funding means that work is done with the best people available at the time (i.e. PhD students and postdocs who often have in-depth statistical training) with the funding that has come through.

However, I do agree that CRU should keep up to date with the latest advances in all the disciplines that their work overlaps with, which is a huge task in such a new and dynamic field as climate reconstructions. I have no doubt, though, that all the criticism that this field is receiving at the moment will accelerate this process of tightening up the methods.

This isn’t an ideal situation but its the way things have been in UK science for many years and the fact that CRU have produced so much world-class and timely science is a testament to their dedication.  Maybe the New Scientist should have picked a headline from this extract:

“We believe that CRU did a public service of great value by carrying out much time-consuming meticulous work on temperature records at a time when it was unfashionable and attracted the interest of a rather small section of the scientific community.”