UK Met office announces a do-over: entire global temperature series – 160 years worth

View from the Solent (09:39:23) :
@ur momisugly Icebergs Everywhere! (02:43:29) :
Could anyone tell me of a layman’s reference site for writing British FOI requests,please?
Here
http://www.foi.gov.ie/how-do-i-make-an-foi-request
Er, that is an Irish government site old chap. In Britain the Information Commissioner’s Office (ICO) ( http://www.ico.gov.uk/ ) publishes guidance notes for members of the public and is also the office to which you should complain if your request is rejected.
Information on FOIA is at http://www.ico.gov.uk/what_we_cover/freedom_of_information.aspx .
Depending on the nature of your request you might find the Environmental Information Regulations ( http://www.ico.gov.uk/what_we_cover/environmental_information_regulation.aspx ) useful as the grounds for refusal are more limited. (I think it is a pity previous requests for CRU data were not made under EIR rather than FOIA.)

“The Met Office is confident that its analysis will eventually be shown to be correct. However, it says it wants to create a new and fully open method of analysing temperature data.”
This statement is what is needed to know that the fix is in. In real science they would not begin by assuming that their desired outcome is the correct one.

Craig Moore (13:43:21) : Well there is no single Ocean temps measure. Not to mention there are issues with those, too, although less well known than the land surface records. Importantly there are a number of adjustments made to them as well.
I recommend searching CA for “buckets”

I’d really love to know why all my comments go to the spam bin 🙁
DaveE
[REPY – Speaking personally, I’m not sure. Comments that are over the top frequently get snipped without ceremony. There are so many comments recently, it can be hard to remember who posts what, so it’s nothing personal. Anything that winds up in spam gets reviewed and can be retrieved. ~ Evan]

This will probably seem way OT, but I believe it is quite a good example of what can happen if the output of unvalidated models fed with poor quality data is used by politicians to develop policies which affects the lives (and livelihoods) of real people living in the real world.
I am a veterinarian in the UK, now retired from the profession, so best described as an ‘ex-country-cow-doctor’. Back in 2001, I volunteered, along with many colleagues from the UK and around the world, to assist our veterinary authorities in dealing with what came to be major epidemic of foot-and-mouth disease (FMD).
Right from the outset, it was destined to be a big one. Disease first came to light on 20th February in a slaughterhouse in the south of England and 2 days later traced back to a swill (waste food) fed pig farm some 300 miles to the north, where the infection had probably been present for at least 3 weeks. The next day (23rd Feb) disease was identified on a nearby beef and sheep farm. Unfortunately, that farm had sent a consignment of 16 sheep (asymptomatic but incubating) to Hexham market on 13th February, unknowingly introducing infection into the UK livestock marketing system at a time which happened to coincide with a seasonal peak in trading activity. Subsequent tracings showed that by 20th February ( ie when disease was confirmed in the slaughterhouse), at least 10 of the final 12 geographic ‘clusters’ in Great Britain had already been ‘seeded’ with infection, and it had also been transported to Ireland, France and the Netherlands. Like I said, it was shaping up to be a big one.
This was the first appearance of FMD in mainland GB for 33 years. The State Veterinary Service (SVS) of the Ministry of Agriculture Fisheries and Food (MAFF) put into effect the tried and tested measures of the ‘stamping out’ policy that had been developed and refined over many decades (FMD was endemic in GB between 1900 and 1962). This included a national ban on animal movements starting on 22nd Feb, the earliest such a ban had ever been imposed even though the true extent of virus dissemination was at that time unknown.
Needless to say, given the extent of virus dissemination that had already occurred, case numbers rapidly mounted and the SVS, with only 220 veterinary officers, was soon overwhelmed. Reinforcements were hastily recruited from the UK and around the globe to work on the problem. We all slogged away from day to day, waiting for the epidemic to peak. (All epidemics peak, and this would be no exception.) Then, something odd happened.
Policy decisions on the control programme were removed from MAFF and transferred to the Civil Contigencies Committee known as COBRA (derived from: Cabinet Office Briefing Room A). COBRA was advised by the Government’s Chief Scientific Adviser, who had established a ‘Science Group’ dominated by four teams of modellers.
On 21 March the modellers delivered, both to the politicians within COBRA and (not agreed beforehand) the general public, their alarming predictions of a ‘huge epidemic’ unless further interventions were urgently put in place. In other words, the SVS had got it all wrong.
So, on 23 March instructions were issued to slaughter all susceptible livestock on all farms contiguous (ie sharing a boundary) with an infected farm without making any attempt to assess the actual risk of infection having been transferred. This policy became known as ‘the automatic compulsory contiguous cull’ or ‘the pre-emptive contiguous cull’ and was a far wider policy than the traditional measures already in place.
FMD control experts within MAFF, UK research institutes and elsewhere immediately questioned the wisdom and practicality of such an approach, not least because the probability of a contiguous farm actually being infected was known and agreed (also by the modellers) to be about 1 in 5. Consequently, about 4 out of 5 farms slaughtered under the policy would comprise only healthy animals (and with due care and attention, a lot of those farms would probably have remained healthy). With the average livestock farm having 5 neighbours, that was going to mean a hell of a lot of dead animals. They also queried whether hastily-assembled and unvalidated models were actually fit for the purpose of directly informing such a novel change in policy, partly because the Veterinary Laboratories Agency (VLA) had from the beginning been running a highly developed, validated and very detailed simulation model called InterSpread which was producing rather different results.
However, the Government of the day rejected all such concerns – perhaps plans for a May election influenced their judgement- and compulsory contiguous culling went ahead. The end result was, as expected, a hell of a lot of dead animals – an overall total of somewhere between 6.5 and 10 million, a high proportion of which were killed unnecessarily (close to two thirds by my estimates) – the contiguous cull policy soon becoming known colloquially as ‘post-code culling’ and ‘carnage by computer’.
Where the contiguous cull was rigorously enforced (majority of GB), an average of 6.5 farms were culled per infected farm. However, in Cumbria where I was working, the weight of infection was such that we had neither the time or resources (or indeed the inclination) to apply a rigorous contiguous cull. Instead, sticking to the tried and tested methodology, all potential contacts were rapidly subjected to basic epidemiological risk assessment. High risk groups of livestock were slaughtered, though not necessarily whole farms. Low risk contacts were placed under restrictions and subjected to close veterinary monitoring. We ended up taking an average of 1.1 farms per infected farm (the lower end of the range actually predicted by the InterSpread model), with no detrimental effect on the process of disease control and eradication.
In the aftermath of the epidemic, Government commissioned a number of separate inquiries, each with fairly narrow terms of reference (the usual tactic to minimise awkward questions).
The most significant in terms of Government response was the Lessons to be Learned Inquiry which reported in 2002, long before full analysis of the epidemic had been completed. Interestingly, the author of that report stated very explicitly that it was not part of his remit to investigate or analyse the extended culling policies used during the epidemic……..and then proceeded to formally recommend that extended culling policies should become part of future contingency plans for FMD. Government gratefully accepted this and with uncharacteristic speed incorporated the recommendation into the primary legislation of the Animal Health Act 2002, a rather essential response given that extended culling was actually illegal according to the Animal Health Act 1981 which was in force during the epidemic!!
By now MAFF had been subsumed into the super-department called the Department for Food and Rural Affairs (DEFRA), the rebranding and reorganisation actually having taken place during the course of the epidemic (what better time to restructure such a key department than slap in the middle of a national disease emergency??!!) Therefore, DEFRA was now responsible for writing the new contingency plans for FMD. Hey presto! – in came a range of extended culling policies for future FMD outbreaks, those same policies being seamlessly extrapolated into the plans for a number of other infectious diseases despite significant epidemiological differences.
Hang on, we said. What about the series of five peer-reviewed epidemiological analyses of field data that were published in the Veterinary Record during 2004 and 2005? Those papers concluded that the contiguous cull policy had been completely unnecessary and in several important respects counterproductive. Furthermore, the models used to inform the policy were fundamentally flawed and not fit for purpose. In any case, the compulsory contiguous cull could not have been essential for the control of the epidemic because it did not start until AFTER the epidemic had peaked, not before as was repeatedly claimed (‘shifting the peak’ = FMD version of ‘hiding the decline’).
Don’t worry, said DEFRA, we are setting up an independent body (aren’t they all?) called the Science Advisory Council (SAC) to evaluate all of the science coming out of the epidemic in a thorough and transparent way. In future, disease control measures will all be supported by sound scientific evidence
It is a matter of public record that the SAC was dominated by mathematical modellers, including some of those directly involved with the models used to drive culling policy during the 2001 epidemic (in other words, keeping things very much within the ‘team’).
Although two general practitioners were involved as co-opted members of one Sub-group, not one specialist veterinary epidemiologist with direct experience of the epidemic was ever involved at all (dissenting views not welcome, especially if you know what you’re talking about – no real surprises there!)
Although DEFRA publicly gave assurance that the published field analysis papers would be reviewed by the SAC, it appeared to be the case that the dominant group of modellers did not consider the Veterinary Record to be quite in the same league as their own favourite journals, such as Science and Nature (a bit of “redefining of the peer-reviewed literature” going on perhaps?)
Worse still, I have seen a document from one influential modeller (yes, it’s an email – those bloody things just seem to get everywhere!) clearly suggesting that all of the field analysis papers be ignored and not used in informing disease control policy (nothing quite like a bit of suppression to round off the day!!)
And just for good measure, DEFRA sponsored a Modelling Consortium (more jobs for the boys!)
I could drone on even longer, though I fear I may have outstayed my welcome already. But, if anyone has got this far without nodding off, you may agree that there appear to be several striking parallels between the modelling/modellers of FMD 2001 and the modelling/modellers of climate science (or ‘science’, depending on one’s view).
If you can bear it, I would urge you to read the following paper published by the World Organisation for Animal Health (OIE):
Use and abuse of mathematical models: an illustration from the 2001 foot-and-mouth disease epidemic in the United Kingdom.
R P Kitching, M V Thrusfield & N M Taylor
This gives a fuller summary of what went on in 2001 than I have, but more importantly contains a thorough critique of the 2001 models, their technical flaws and their practical shortcomings. The following link should take you to the abstract page which will enable you to get the pdf free access.
http://www.oie.int/boutique/index.php?page=ficprod&id_prec=96&id_produit=293&lang=en&fichrech=1
As far as the climate discusion is concerned, there could be problems with the British govt because so many departments and ministers are mesmerised by models and just love their output, irrespective of quality or accuracy, ‘specially if there are some neat animations. The general intellectual standard is, I fear, exemplified by Our Great Leader G Brown now describing ‘climate deniers’ as ‘flat-earthers’. Hardly helpful in the current context and distinctly puerile.
My own view echoes that of many others: we need good data with any tweaks fully explained, and fair evaluation with all methods, assumptions, gaps, limitations, error ranges etc also fully explained. What we don’t need is some fudge-fixed one size fits all lash-up pretending to be something it’s not just to make a ‘good story’.
There is probably little doubt that the climate is changing – I would be very surprised and more than a little concerned if it’s not – but I can’t honestly believe that such a complex system, probably better described as an interaction of multiple complex systems, can really be hypersensitive to a single factor such as CO2 concentration. In nature, things are seldom that simple. As so many have pointed out before, correlation does not mean causation.
Good luck to all you sharp brains out there. We are all relying on you to sort out this bugger’s muddle as best as you can.

3rd party evaluation: to whom shall this task be given – disinterested extraterrestrials? Or both sides posting their own scientists and guards, co-located with each data collection equipment site – 24/7/365, armed and ordered to defend the equipment, to the death? (like some new version of “Alien”?)
In the USA, we are seeing judges, become unprofessionally and irrationally outraged at plaintiffs, suing over constitutional violations, without even considering evidence. (In ways which stink of coercion of judges.)
This corruption knows no bounds. The “Progressives” are committed to an all-in battle to dominate. How would the people of the earth depend upon the integrity of new data including sensory placement and validation?
Governments have the ability to monitor any data transmitted over the internet and potentially intercept and tweak en-route. We have seen the willingness to fabricate and use AGW as a means to take full control of the sovereignty of humanity. The UN/IPCC et al, have no place in global governance – Just like the Japanese were stripped of the rights of providing themselves a military, after WWII. Once bitten by a snake, terminate with prejudice.
Allowing a UN redo is being “Politically Correct”, which only defeats those who play by fair rules. It is saying “We caught you this time and you best not do it again.” THWT!

Hmmm… the Times story has been taken down from their website. Any word on this? Was it in error?

Clive (21:54:35) & Rhys Jaggar (08:06:50)
Thanks to you both. It’s a pity I didn’t manage to get my post up sooner – I might have got a wider readership before things moved on.
As you both suggest, a mixture of politics and warped science is almost invariably a very toxic one.
Having been involved in a small way in the published epidemiological analyses of field data, I was party to the debate that my more highly qualified colleagues were attempting to open with the modellers and the SAC – attempting being the operative word.
As Rhys said, the modellers were clearly unable to engage in handling objections. There was a lot of to and fro about the nuances of complex maths and obscure statistical methods which only seemed to cloud the fog rather than clear it. They always seemed to be “cleaning” the field data, but exactly how or why was never explained. At one point it seemed that the output of one model was being used to validate another model (- is that valid??). All in all, there seemed to be a lot of effort to fit data to the models rather than to validate the models against the data. Par for the course, I suppose.
On a related track, between August ’05 and April ’06, I tried to engage the then Chief Veterinary Officer (the incumbent had changed since FMD 2001) in a scientific discussion of FMD control principles and policy.
Unfortunately, like most of the British Civil Service, the post of CVO is now highly politicised (about 90-95% political and 5-10% scientific, I’d guess). So, as predicted by Rhys (presumably without the aid of a computer model!), the result of our correspondence may be summed up in one word: obfuscation!!
What really worries me is that our politicians and most of the public simply do not understand the difference between the virtual world of models and the real world of observations. There is no doubt that model output can be very convincing, but, rather more to the point, it can also be made to look very convincing.
Perhaps I’m getting too cynical in my old age! Oh well, back to chewing the carpet.
In the words of the Oracle:
Cheers!