This is a repost from Ross McKitrick’s website. Ross was a peer reviewer of the original BEST paper. I think its important to get this publicised as widely as possible, as the Guardian, LA Times and many other media outlets recently carried an opinion piece from BEST team leader Richard A Muller trumpeting his rejected JGR paper as if it was worth something. The L.A. Times followed up with another opinion piece, a classic foot in mouth. Steve Mosher and Zeke Hausfather also have some coverage of the BEST team’s data update at Judy Curry’s site. – Rog.
Judy herself was rather scathing about the BEST paper, saying
the method used to attribute the warming to human emissions is way over-simplistic. I don’t think this question can be answered by the simple curve fitting used in this paper, and I don’t see that their paper adds anything to our understanding of the recent warming.
BERKELEY EARTH STUDY REFEREE REPORTS
Ross McKitrick
On September 8 2011 I was asked by Journal of Geophysical Research to be a reviewer for a paper by Charlotte Wickham et al. presenting the Berkeley Earth Surface Temperature (“BEST”) analysis of the effect of urbanization on land surface temperatures. This work is mainly associated with Richard Muller and his various coauthors. I submitted my review just before the end of September 2011, outlining what I saw were serious shortcomings in their methods and arguing that their analysis does not establish valid grounds for the conclusions they assert. I suggested the authors be asked to undertake a major revision.
In October 2011, despite the papers not being accepted, Richard Muller launched a major international publicity blitz announcing the results of the “BEST” project. I wrote to him and his coauthor Judy Curry objecting to the promotional initiative since the critical comments of people like me were locked up under confidentiality rules, and the papers had not been accepted for publication. Richard stated that he felt there was no alternative since the studies would be picked up by the press anyway. Later, when the journal turned the paper down and asked for major revisions, I sought permission from Richard to release my review. He requested that I post it without indicating I was a reviewer for JGR. Since that was not feasible I simply kept it confidential.
On March 8 2012 I was asked by JGR to review a revised version of the Wickham et al. paper. I submitted my review at the end of March. The authors had made very few changes and had not addressed any of the methodological problems, so I recommended the paper not be published. I do not know what the journal’s decision was, but it is 4 months later and I can find no evidence on the BEST website that this or any other BEST project paper has been accepted for publication. [Update July 30: JGR told me “This paper was rejected and the editor recommended that the author resubmit it as a new paper.”]
On July 29 2012 Richard Muller launched another publicity blitz (e.g. here and here) claiming, among other things, that “In our papers we demonstrate that none of these potentially troublesome effects [including those related to urbanization and land surface changes] unduly biased our conclusions.” Their failure to provide a proper demonstration of this point had led me to recommend against publishing their paper. This places me in an awkward position since I made an undertaking to JGR to respect the confidentiality of the peer review process, but I have reason to believe Muller et al.’s analysis does not support the conclusions he is now asserting in the press.
I take the journal peer review process seriously and I dislike being placed in the position of having to break a commitment I made to JGR, but the “BEST” team’s decision to launch another publicity blitz effectively nullifies any right they might have had to confidentiality in this matter. So I am herewith releasing my referee reports. The first, from September 2011, is here, and the second, from March 2012 is here.
Tallbloke's Talkshop 
Hmm, WUWT has started resolving here again now and I see they already covered this story here:
And a few related ones since.
Hello Rog, good subject which was defeating me on what exactly to pick up.
Hi Tim,
Thanks for minding the blog while I’ve been busy elsewhere, I have plenty of reading and catch up to do!
Muller’s presumption that the paper was so good that publication and plaudits were inevitable is what really grates. What a posturing buffoon!
TB
Welcome back. I guess the Norfolk fuzz released you when they wrapped up their enquiry, right? 😉
On the subject of BEST. As you know I’ve done a lot of work with the surface temperature records, and FWIW here’s a comparison of my reconstruction of annual global surface temperatures (based on about 900 selected records) with the BEST reconstruction:
Points to note are:
I start the comparison in 1920 because coverage before then is too sparse to make good estimates of global means (too many holes in the Southern Hemisphere). The fact that BEST carries its reconstruction back to 1753 shows that they really don’t understand the limitations on what you can and can’t do with the surface temperature records.
BEST shows about 0.4C more warming since 1920. Partly this is because BEST uses “adjusted” data (anyone who shows surface warming over SE Australia has to be using adjusted data) while I don’t, and partly it’s because BEST averages temperatures only over “land” areas while I project them out over the oceans.
But my reconstruction still shows warming. I can’t make it go away.
And the records I used don’t show an urban warming component. Here’s a plot of population versus temperature increase, calculated by subtracting 1964-73 from 1994-2003 decadal means, at the 354 stations I used that have data for both periods:
Roger, I think you and BEST are using data which may be raw (as transmitted to the global network) but has been contaminated (UHI & other human induced affects) or adjusted before being transmitted. The court case in New Zealand has shown that data has in some way been adjusted without a record available of the adjustments. The same applies in Australia. A number of people have recorded unwarranted sudden increases in numerous station data. I have looked at monthly data for one small rural town. On two occasions there were sudden upward jumps, the first possibly due to the road being sealed in the town and the other due to buildings or air conditioner exhaust. A new station was established 2 km out of town at the airstrip. There was 6 yrs overlap between the two stations with the airstrip station (which itself has UHI affect from bitumen runway and parking area) averaging 0.2C lower on both maximum and minimum than the town station.
It is my view that actual unadjusted temperature data in Australia prior to 1950 is likely to more accurately reflect true near surface air temperatures (from more dedicated observation and recording and better work ethics) than recent results particularly those from automated stations at airports which have gone through an adjustment process by BOM.
Anthony Watts has done everyone a favour by photographing US recording stations, trying to classify them and pointing out errors in measurement.
However, I suggest that the whole global temperature exercise is pointless. If one understands thermodynamics and heat& mass transfer it will be clear that the CO2 in the air has an insignificant effect on the near surface air temperature in any part of the globe. More importantly, people live happily & more productive in Singapore (which has one of the highest GDP/capita) and Indonesia than say in Greenland (which once was green), Lappland or Siberia. I once worked in Canada with 6 months of snow around the house. I also recall snow and ice going to school in Berkshire I prefer the warmth of Australia.
Roger Andrews says:
August 1, 2012 at 6:10 pm
This is a good analysis of a subset. If a global condition doesn’t show up in a subset of the global sites, provided that there are enough of them and distributed reasonably well, then the “global” condition is a regional condition layered on top of a neutral, truly global condition.
The Arctic, to judge by the differences between HadCruT and GisTemp graphs, is a critical piece of the regional, not global condition. The increase in minimum temperatures as expressed by Pielke Sr., is a temporal but still regional condition as it affects strongly human-inhabited regions, putting an increase onto a record in which daytime maximum temps have stayed constant. Watts et al 2012 shows the warming bias obvious in the GCHN 2000 corrections to be a local UHIE affect pushed into the non-UHIE data. Each piece of the regional “specialness” can do its warming stuff because the non-special rest-of-the-world is neutral, not holding a cooling bias.
Next: the ocean “acidification” data, and the “reefs-are-dying” claims.
In agenda-science, who said that only a globally consistent observation says something conclusive about the Planet?
Hi Roger A, you get a 0.6C increase in global average surface temperature since around 1970. If the Australia and NZ shenanigans CF reminds us of mean that the southern hemisphere didn’t warm much, as the increase in sea ice around Antarctica over the last 30 years might help indicate, that global increase may be a lot less. Possibly somewhere in line with Watts et al’s findings for the U.S.
So the Norfolk plod have wrapped up the FOIA investigation? I’ve missed much in the last weeks, give me a briefing. 🙂
Ah, found it:
Rog, best single source on that one is Bishop-hill (he was the one actively doing foia and telephoning asking fool questions), with other material on ClimateAudit. (both in blogroll). Not much detail on this site.
You get mentioned most places.
Nothing much definite appears, police are saying nothing.
One of my theories is this was close early out of paranoia so any heavy press is covered up by the darn Olympics mediafest.
There are a number of evil things being done right now, sliding in nasty stuff.
cementafriend:
The records I use are unadjusted. I downloaded most of them from GISTEMP in the late 1990s, before the mean temperature manipulation merchants got their act together, so I can tell whether they’ve been tweaked since then. The records I use for Australia and NZ don’t include the spurious warming adjustments applied by the Aust BoM and NIWA.
I think the second graph in the comment above confirms that the records are free of spurious “urban” warming gradients. This doesn’t necessarily mean that they are free of spurious “local” warming gradients caused by construction around the station, but the global record is still clearly dominated by natural effects. The abrupt transition from cooling to warming in 1976, for example, coincides with the PDO “phase shift”, not with a sudden increase in construction activity.
Doug Proctor & TB:
Re hemispheric differences. According to my reconstruction surface air temps have warmed a lot more in the NH than in the SH:
The reason we don’t see this large differential in the BEST, CRUTEM, GISS or NCDC reconstructions is that they use warming-adjusted SH records. Using these adjusted records increases overall “global” warming and also provides a much better match to climate models, which show the NH and SH warming by about the same amount in the 20th century.
Roger A: what makes you think the mean temperature manipulation merchants hadn’t started messing with the data before the end of the 90’s?
Tim C: If FOIA decides to release more emails just after the three years is up, the cops will get flak for their failure to nail the culprit anyway. My reading of it is they have come to realise that continuing the charade of an ‘ongoing investigation’ is counterproductive to their image. Which is what I told them last December when they popped by at my place.
TB:
The GISTEMP data I used back in 1998 were based on the 1997 GHCN data set, which contained both unadjusted and homogeneity-adjusted data subsets. I don’t recollect exactly how the GISTEMP categories were defined, but I’m pretty sure the “after combining sources at the same location” set I used was the unadjusted GHCN. Partly this was because GISTEMP contained multiple records for most stations that matched the “combined” records, and GHCN didn’t apply adjustments to these multiple records, partly because GISTEMP put “homogeneity adjusted” records in a separate category and partly because if the “combined” records were homogenized then GHCN had done a lousy job of it. (Records with strong UHI gradients, records that looked nothing like the records around them, US and Australia show no warming etc.)
The first tweak to the GISTEMP records occurred around 2000, when almost all of the unadjusted US records were overwritten by the “corrected” USHCN records – a hanging offense in my opinion.
The second tweak occurred some time between then and 2004, when the Australian records between 1995 and 2000 received mysterious and so far as I know as-yet-unexplained warming adjustments of up to 1C (I back these adjustments out in my reconstruction). The only justification I can think of for applying these adjustments was that the unadjusted records showed an embarrassing amount of cooling over this period.
There may have been a few more tweaks since then, but I haven’t checked.
Roger A: Invaluable historical testimony, thanks. So James Hansen overwrote the unadjusted data? With no backup? This kind of data vandalism is unforgiveable. Words fail. Well words I can use in public anyway.
This new WUWT article by Frank Lansner may be of interest:
TB:
Let me respond to that later, but first this:
The title of this thread is “The Worst of of BEST”. So what is the worst of BEST? Here’s what I think.
It’s Richard Muller’s claim, widely acclaimed in the mainstream media (Anthony Watts’ paper didn’t get a mention, by the way) that “Humans are almost entirely the cause of climate change”.
He makes this claim because he can fit the BEST surface temperature record all the way back to 1753 using a combination of CO2 warming and volcanic cooling, thereby demonstrating that we humans caused all of the warming since the beginning of the Industrial Revolution, not just the recent stuff.
The IPCC has been looking for something like this ever since the Mann Hockey Stick died. So we can confidently look forward to seeing this graph
[This is a local copy, just in case… –mod]
prominently displayed in the upcoming AR5, probably right at the top of page one.
Unless of course the BEST paper gets canned in peer review, but I would rate the chances of that as being about equal to the chances of me winning a medal in Women’s Beach Volleyball.
I haven’t had chance to read it yet, but Willis has something to say about the volcano element in the BEST paper here:
See also the link to the Lansner article I just added to my previous comment.
WE is correct. I just finished comparing Muller’s volcanic eruptions with the long-term European and US records he matches them against and found that a temperature decrease coincides with the eruptions in only six of eleven cases. And that was only because I knew when the eruptions occurred. Otherwise there’s no way I could have picked them out of the background noise.
Sure, WE got the idea to do his study (and a couple of previous ones) from this post:
On Lansner. I went through a similar exercise when the USHCN corrections came out 12 years ago and concluded that any attempt to homogenize the US records was a waste of time, unless of course your goal was to homogenize them so that they showed more warming. (Which in fact it was. The concept underlying the USHCN corrections was that reading time changes, equipment
changes and progressive relocations of stations from urban areas to suburban airports had introduced a spurious cooling gradient into the US records and that a net warming correction was needed to remove it. I kid you not.)
So in my view the question of whose adjustments are right is academic. By far the most robust way of constructing a good US temperature record is simply to average all the 1000+ unadjusted records together (after weeding out the ones that show obvious UHI impacts, like Los Angeles) and let the random errors in the records cancel each other out. But this record won’t show much warming.
On the question of adjustments in general, I think anyone who applies them to temperature records without a watertight reason for doing so should be thrown in jail. This is only fair. I would be thrown in jail – and rightly so – if I got caught applying USHCN-type corrections to a set of gold assays.
Violent storm developing here. Must be global warming. Gotta go
Roger A: I hope the patio chairs didn’t blow away in the night.
Zeke Hausfather made some comments on the reasons for upward adjustments here:
http://rankexploits.com/musings/2012/initial-thoughts-on-the-watts-et-al-draft/
While Anthony Watts passes comment on TOB’s adjustments here:
His draft paper is here:
Click to access watts-et-al_2012_discussion_paper_webrelease.pdf
Judy Curry’s critique of Muller et al’s new paper is here:
TB: The iron patio chairs didn’t blow away but a couple of large potted palms got upended and the swimming pool is full of pieces of tree. Winds must have been close to hurricane strength.
I’ve been through most of the stuff you linked to and still maintain that it’s impossible to apply meaningful adjustments to the US records. The merchants of manipulation in fact make the adjustments up as they go along, and the sky’s the limit. Here’s an example. It’s not untypical.
I semi-jocularly posted elsewhere:
“NOAA seems to be a major source of misrepresentation, as though it falsified its data sets to generate graphs-to-order for stupid alarmism. A simplistic hypothesis, I know, but it seems to do pretty well as a predictor.”
They have lots of collaborative competition, it seems!