This is a guest post by Richard Brearley.
This post is written to publicise an interesting anomaly in UEA’s treatment of requests sent to it under the Environmental Information Regulations.
Readers should know that my interest is only in compliance with EIR Regulations. I am neither qualified nor sufficiently knowledgeable to get to the meat of some of the requests for data from UEA or to understand it even if it is disclosed. I am a lawyer by training and the recent QUB and UEA ICO cases have interested me greatly – they have shown a strong reluctance on the part of these publicly funded organisations to retain for their own benefit the work for which we have all paid. Thus is identified a friction between what those institutions consider to be their valid interests and what the Regulations provide – public access. I have become highly interested in the way in which these interests are balanced and my preference is in favour of public access, (which is under the Regulations the presumption from where we all start), to information created or produced with tax payer money.
Readers may remember Steve McIntyre’s EIR request to UEA (UEA ref: FOI_11-047; EIR_11-004) relating to the “URALS” as referred to in the 2006 email from Dr Osborn (see below). I read Mr McIntyre’s posts on this and as a UK resident concerned about the apparent repeated failure of tax funded scientists to reveal data, etc., and in the interests of properly transparent science, I decided to lodge an identical request to that lodged by Mr McIntyre. I felt that I might be able to monitor and pursue such a request more easily being both resident here in the UK and also a solicitor who might have a reasonable understanding of the relevant law and who would be able to deal with any future hearings myself.
My request (UEA ref: FOI_11-113; EIR_11-010) was as follows:
Dear Sirs,
Re – Freedom of Information / Environmental Information Regulations request
Please accept this letter as a request under the above legislation/regulations.
My request arises from the following email of 28.4.2006:
Date: Fri, 28 Apr 2006
To: philip.brohan
From: Tim Osborn
Subject: Re: Standardisation uncertainty for tree-ring series
Cc: Keith Briffa,simon.tett
Hi Philip,
we have three “groups” of trees:“SCAND” (which includes the Tornetrask and Finland multi-millennial chronologies, but also some shorter chronologies from the same region). These trees fall mainly within the 3 boxes centred at: 17.5E, 67.5N;22.5E, 67.5N; 27.5E, 67.5N
“URALS” (which includes the Yamal and Polar Urals long chronologies, plus other shorter ones). These fall mainly within these 3 boxes: 52.5E, 67.5N; 62.5E, 62.5N (note this is the only one not at 67.5N); 67.5E, 67.5N“TAIMYR” (which includes the Taimyr long chronology, plus other shorter ones). These fall mainly within these 4 boxes: 87.5E, 67.5N; 102.5E, 67.5N; 112.5E, 67.5N; 122.5E, 67.5N
We do some analysis at the group scale, and for this we take the JJA temperatures from each box and average to the group scale to obtain a single series from each of SCAND, URALS and TAIMY.We do some analysis at the overall scale, and for this we take these three group temperature series and average them to get an overall NW Eurasia temperature for boxes with tree chronologies in them.
We did also try using a wider average for the region, including all LAND temperatures from grid boxes within a rectangular region from 12.5E to 127.5E and from 57.5N to 72.5N, but I don’t think it correlated so well against the tree-ring width data (I can’t remember the exact correlations), so we didn’t pursue that.
Does that give you enough information to be going on with? I’d recommend using CRUTEM3 rather than HadCRUT3, because the correlations seem to deteriorate with the inclusion of SST data in some cases — though of course you can look into this yourself.
Cheers
TimI would be grateful to receive from you:
1) A digital version of the series for the “group of trees” described in the email as “URALS” (which includes the Yamal and Polar Urals long chronologies, plus other shorter ones).”;
2) A list of all the measurement data sets used to compile this series “URALS”, including the Yamal and Polar Urals long chronologies and the shorter ones, with each data set denoted by ITRDB identification or equivalent;
3) A digital version of the measurement data used in this series “URALS”.
Yours faithfully……
Mr McIntyre’s slightly earlier request to UEA was in identical terms.
My original request was refused based on the now familiar grounds that UEA has deployed in a previous ICO case which was recently reported and in respect of which there was some recent blogging comment and activity (see the Professor Jones case):
http://www.ico.gov.uk/~/media/documents/decisionnotices/2011/fer_0282488.ashx
and on the grounds deployed by QUB in the Keenan case:
http://www.ico.gov.uk/~/media/documents/decisionnotices/2010/FS_50163282.ashx
and I submitted a request for an internal appeal which is currently being processed by UEA.
However, in my appeal to UEA for internal consideration I made clear amongst other things that it had not properly identified the information I was requesting and had in fact unilaterally interpreted my request (contrary to Regulation 9) as being for 1001 bootstraps. UEA had responded by refusing to provide information I had not asked for and of course under Regulation 9 the public authority is obliged to clarify any request that might be ambiguous – although in the request I had only used the “nomenclature” used by UEA itself (and which point I made in my appeal), asking UEA to clarify this). The issue arose from the email of 2006. There is, I accept, room for interpretation of what Dr Osborn was actually talking about in that email. The later comments therein certainly make clear that there are three group temperature series which are worked on at the group scale and that these three series are then averaged to obtain a single result. I put this to UEA and their response was as follows:
Dear Mr Brearley
ENVIRONMENTAL INFORMATION REGULATIONS 2004 – INFORMATION
REQUEST (FOI_11-113; EIR_11-010)
Further to your letter of 29 June 2011 and our response and 30 June 2011, we are writing to take you up on your suggestion that we communicate further in order to clarify your request and the grounds of your appeal.
It is very clear that you feel that we have not provided you with what you believe you were actually requesting. Therefore, as the information requested is very much based on an email from Dr. Osborn in 2006, we have consulted with Dr. Osborn again to further clarify what he was actually referring to in his email.
In your letter of 29 June you contend that “...the email does clearly identify a “series” for the “group of trees” referred to as URALS” and go on to assert that “Dr Osborne is clearly referring to analysis at the group scale (i.e. each of the three groups of trees), and also to three group temperature series, one under each of the group labels, SCAND, URALS and TAIMY. Obviously one of these series is “URALS”. It is clear that these three series are then averaged to produce a single series. URALS at that time is clearly a discrete group temperature series, individually comprised of the information I have asked for, namely the temperature series your university labels “URALS (which includes the Yamal and Polar Urals long chronologies plus other shorter ones)”, and the identification and measurement information relevant to it covered by my requests 2 and 3.” (our emphasis)
We have shared your comments with Dr. Osborn, who responds that the label ‘URALS’ in his email of 2006 referred both to ring-width measurement data for a group of trees in this region (encompassing part of the East Ural Mts and the adjacent Yamal peninsula) and to thermometer-based temperature data for the same region; i.e.
(1) A composite (more usually called a chronology) of selected treering measurements from this region.
(2) A composite (more usually called an average) of temperatures measured by thermometers from this region.
We accept that your original request was not for the 1001 bootstrap composites (also referred to in the 2006 email), which was our initial interpretation. Your original request mentioned “groups of trees” but did not mention “temperature series”, which leads us to interpret your request as being for item (1) as defined above by Dr. Osborn, in addition to the associated measurement data. However, in your appeal you make repeated reference to “temperature series”, which suggests that your request is for item (2). Please would you confirm whether you are requesting item (1) or (2), both, or something else? Once we are certain of what you are seeking, we can then provide a proper response to both your original request and your appeal. Please quote our reference given at the head of this letter in all correspondence.
Yours sincerely
I responded to UEA and thanked them for taking the time to clarify the position. I was obviously pleased that Dr Osborn was in fact able to properly identify what he was referring to in the email and that it comprised discrete information that was identifiable and that we all now knew what we were talking about. This, I considered, was a helpful response that enabled me to consider the position and clarify my requests – all in line with the Regulations and UEA was properly addressing its obligations. I asked that both items in 1) and 2) in their letter should be considered as part of my request and should be included in my internal appeal.
My requests may well be denied at internal appeal stage and will, if so, move to the ICO. That, however, is not the point of this post.
In early July I corresponded briefly with Mr McIntyre to see how he was getting on with his request. Mr McIntyre made an identical request which was denied on identical grounds to my own. He also in his internal appeal rejected UEA’s unilateral interpretation of his request as being for 1001 bootstraps, reiterating what he considered that he was asking for.
Again, in Mr McIntyre’s case as in mine, UEA went to Dr Osborne for clarification of what the 2006 email was in fact referring to.
The letter to me (set out above) identifying what Dr Osborn was referring to in his email is dated 15.7.2011. The letter to Mr McIntyre setting out Dr Osborn’s recollection of what he was referring to in the same 2006 email, and from which the below extract is taken, is dated 18.7.2011. Dr Osborne’s recollection was rather different in respect of Mr McIntyre’s request.
This time UEA responded to Mr McIntyre as follows (and Mr McIntyre has agreed to publication here of relevant parts of UEA’s letter to him):
…
Your request is based on information in an illegally obtained email that was not written as a public document and therefore I have asked Dr Osborn as the author of the email to explain what his email was referring to, his response is below:
It is worth clarifying here that although the original email described “Yamal and Polar Urals chronologies” the intended meaning was “trees from the sites that have previously been used to construct long chronologies called Yamal and Polar Urals, plus other shorter ones.” Thus it was not referring to chronologies but to groups of trees. The wording in the email may have led the requester to believe that new chronologies for Yamal and Polar Urals had been produced at that time.
The email you refer to is ambiguous and as Mr Palmer observed in his response “No such composite was attached to or identified by the 2006 email”. The fact that the email does not identify a composite makes it difficult for me to determine what information you are actually requesting.
“On this point in your appeal you have said:
You state that the requested chronology was not attached to the email. This is irrelevant to my request. Dr Osborn knew what he was referring to in the email and is in a position to identify the chronology in question.
As you have observed, Dr Osborn, as the author of the email, is best placed to identify the chronology that you have requested. As part of considering your appeal Dr Osborn has provided me with the following information which is pertinent to our assumption that your request was for the 1,001 bootstraps:
The requester is incorrect: it is relevant that no data were attached to the email and although I do know what I was referring to in the email, this does not necessarily mean that I can identify the requested chronology. UEA’s response noted that “No such composite was attached to or identified by the 2006 email”. The requester has selected just one half of this statement to support his criticism, but in fact both elements must be considered together to come to a fair evaluation of their relevance.
The original request is ambiguous for two reasons.
Firstly, the email identified as 1146252894.txt refers to groups of trees, to temperature data from the regions where the trees were sampled, and to a future plan to process the measurement data from the groups of trees to produce proxy series and many bootstrap estimates of these series. It is only an inference that a regional chronology was later produced for the URALS group of trees. Our initial search for information relevant to this request suggested that this inference was false.
Second, the email was discussing ongoing work, which began in 2005, continued through 2006 and 2007, and to which we have recently returned. During that work, we have produced a number of different regional chronologies for the ‘greater Urals’ region involving multiple methods of selecting, processing and aggregating the measurement data. Some of these regional chronologies have been retained and some have not. The only identification of the requested information is from email 1146252894.txt, but because this only refers to future plans to produce a regional chronology from the URALS group of trees, this is not a unique identification of which – if any – of our working files might contain the requested information. Had the requester asked for a regional chronology that had actually been published, then identification would have been more straightforward because the request could have asked for the specific published version. Similarly, had Dr Osborn attached a specific version to the email 1146252894.txt then this would also have provided a unique identification of the information being requested.
That is why it is relevant that no data were attached to the email, because attached data would have avoided this ambiguity in identifying the requested information.
In your appeal you have clarified that you are seeking the URALS regional chronology, unfortunately there remains some ambiguity because the email refers to a future plan to produce such a chronology rather than to a chronology that had already been produced. Dr Osborn has explained to me that:
Regional chronologies have subsequently been produced from various data in the ‘greater Urals’ area of northern Siberia, and one of those might be based on the URALS group of trees referred to in the email.
Dr Osborn has provided the following further information on the approach he took in trying to locate whether the information you requested was held by us:
Although the request is (unavoidably) ambiguous with regard to which – if any – of our working files might contain the requested information, we did undertake a search for material that might match the request. We did not identify a file containing a single non-bootstrap regional chronology, and could not recollect that such a regional chronology had been produced from the group of trees identified as URALS in email 1146252894.txt. We did, however, identify a file that we believed to contain 1001 bootstrap estimates of a regional chronology derived from a URALS group of trees that were produced a few weeks after email 1146252894.txt was sent (note that this is deliberately described as “a” URALS group of trees rather than “the” URALS group of trees, since the definition of which group of trees form the basis for a regional chronology for this region has evolved over the course of this research).
The decision to proceed with the interpretation that the set of 1,001 bootstrap estimates of a regional chronology was the closest match to what had been requested was made purely in an attempt to be helpful to the requester and because the requester may not have been able to clarify his request anyway, given that the information that he sought does not have any precise identification.
In investigating your appeal a further search has been undertaken and it has become clear that the first time series out of the 1001 bootstraps is actually a single (nonbootstrap) chronology. However the ambiguity that remains around your request means that I cannot be certain that this single chronology is the one that you are requesting, nonetheless in absence of any other chronology I will assume that this is the one that you seek.
…
Of course readers will immediately identify the difference in tone – reference to an “illegally obtained email”, etc., and the fact that in the case of Mr McIntyre’s request Dr Osborn was not readily able to identify what his 2006 email was referring to. Why, I wonder, do 2 identical requests result in two such different responses?
It must be the case that one must expect an identical request to receive identical treatment, particularly in the basic and preliminary context of identifying the information requested. In my view this is probably a manifestation of UEA’s documented approach that it should not be expected to comply with requests which are based, in my observations, on its assessment of the assumed motive of requesters of information – an entirely invalid consideration under the Regulations. Some will argue that this response is purely based on the fact that it was Mr McIntyre who was the requester of the information. The Regulations do in large part depend on public authorities openly responding to requests and if, as may be the case here, there is obfuscation operating within public authorities in response to requests that they don’t like, then where does that leave us? The Regulations provide numerous exceptions upon which requests can be refused and no-one has any argument about the fact that parties are entitled to make those arguments. What the Regulations do not do is allow a public authority to obfuscate on requests from people they do not like.
But what do we say about this situation? Is it the case that every requester that UEA may have suspicions about is destined to end up before the ICO? That approach is not compliant with the Regulations and is to be deprecated against the backdrop of a very clear and well implemented regime of public access to information held by public authorities. The FOI/EIR regime is now long established. What can be the justification for public authorities apparently treating identical requests differently based on who the requester of information is and what motives the public authority, rightly or wrongly, attributes to the requester? The Regulations are blind to motive for a reason. It is not the public authority’s concerns about disclosure that matter – at all. It is the right of access to information granted to the public at large that matters and UEA is setting itself up to be dragged kicking and screaming, at great cost to itself, into this reality. At some point this issue will be resolved, either by UEA, or by a judge – and at what cost to UEA’s reputation if it has to be by a judge?