July 18, 2012

In 2009 the University of East Anglia Climate Research Unit ("CRU") had its email stolen. Police today cleared anybody at the University of any wrongdoing and no suspects are known to this crime.

That's the story that was reported today.

Now here's what really happened.

There had for some time been some concerns in the scientific community that the CRU wasn't sharing their data and that they were hiding something, to the point where they were ignoring proper due process and laws to keep data secret.

Then, their email was hacked - stolen. Although today we use the term "leaked", ala Wikileaks and somehow their emails and some key documents were leaked; it is a bit hypocritical to be in favor of diplomatic cable leaks, but against the CRU leaks. The police just cleared everybody in the university, just as the U Penn people cleared the scientists behavior.

But, the emails proved that they:
(A) were engaged in a united attempt to keep other people's papers out of the peer-reviewed journals (maybe not illegal but certainly not ethical),

(B) agreed to avoid giving information to certain people they viewed to be on "the other side", even if it meant they had to break the law to do so, and

Not to mention some of their other behavior which, while again not criminal, was hardly very professional:

'Mike's nature trick', is arguably a case of data manipulation through omission and obscurity. By cutting data off at an inconvenient point and substituting data obtained from an entirely different methodology to visually obscure on a chart how key data diverges and fails to correspond to what they claim it corresponds to . An honest broker would admit that the data may not necessary represent what they hope it represents. Instead the say the data is perfectly fine up until the point where it was not fine but it is a-okay to hand wave the problem away and make up some untested, unverified excuse why that bit can be ignored.
Professional conduct that certainly falls short of what R.Feynman advocated: " It's a kind of scientific integrity, a principle of scientific thought that corresponds to a kind of utter honesty--a kind of leaning over backwards. For example, if you're doing an experiment, you should report everything that you think might make it invalid--not only what you think is right about it"
Now you could argue this is nit picking and I guess it is. But this is not some inconsequential field of science. It has global political ramifications as folk are trying to radically deconstruct and reconstruct our global society in order to dodge the CAGW bogeyman. Personally, I would prefer if there were more people of R.Feynman's calibre involved in the discovery and analysis process. The climategate emails reveal that there are not.

The problem is, both were completely and utterly incompetent. The U Penn people were the same ones that cleared Sandusky (now spending his life in jail on ~40 counts of child molestation) and the police lacked technical expertise; and better analysis exists suggesting very strongly it was indeed somebody on the inside, somebody with actual scruples that had had enough of the lies being tossed about to block legitimate FOI requests. As for the "Police" as one commentator put it:

"You have WAY too much faith in the incompetent buffoons of the Norfolk Constabulary. One of the first things they did was to seize the computers from the guy that ran a blog where links to the leaked emails were first posted, including his adsl router. Really? The blog wasn't even hosted on those computers.
Any claim from these guys that they know how the emails were obtained has no credibility whatsoever. Better to just listen to the experts.

And they were links, not the emails themselves. It's a big difference

And as for the "investigation" into the ethics of the CRU people, another commentator put it:

It appears the 90s and 2000s-era Penn State administration covered-up anything that would have a negative impact on their university's reputation. Sandusky diddling boys? Cover it up. Sports players caught raping girls off campus? Cover it up. Students jumping from windows? Cover it up and just call it "an accident". Death threats to black students? Cover it up! On-campus shooting..... well they couldn't hide that, but they don't talk about it anymore. The 10th anniversary came and went with nary a mention. Mann falsifying data? Cover it up.

Most of the controversy was about the hockey stick graph which they go to great pains to point out has been validated by 5 or 6 other sources. Which all relates to "was the medival warm period warmer or cooler than now", the CRU argued agaisnt that and withheld data from people trying to show this was not true (it isn't it turns out). They're trying to make the point the climate of the latter half of the 20th century was unprecedented and that it had never in recent history been this hot and was getting worse.

But! The recent Scandinavian tree ring data suggests that not only was the medieval warming period warmer than now, but it was warmer even than that in Roman times.

See for yourself, here's the Scandinavian data: http://www.nature.com/nclimate/journal/vaop/ncurrent/images_article/nclimate1589-f2.jpg

and here's the famous hockey stick graph: http://upload.wikimedia.org/wikipedia/en/thumb/e/ed/Hockey_stick_chart_ipcc_large.jpg/350px-Hockey_stick_chart_ipcc_large.jpg

Now, not the latter has a 0.5 degree y axis, and it's pretty well spaced out. Note the other graph has 1 degree ticks and they're pretty close together.

When you normalize the y axis and superimpose the hockey stick graph on top of the Scandinavian data it becomes rather obvious.

If we align the x-axis too, it's a true superimposition of the hockey stick graph onto the Scandanavian data:

The hockey stick graph has a frightening upshift at the end that shows we're on the verge of burning up, but, it turns out we're not. In fact it's been twice that hot already, and had been for hundreds of years. And not just hotter, but more variation - colder, too. As an aside, it was the Romans, not hippies in the 70s that invented solar powered air conditioning.

In other words this kind of warming not absolutely not unprecedented, it's happened before and not only that it's been way hotter.

So, their claim "the latter part of the 20th century is the hottest in 1000 years" would seem to be false. And that it's normal for the earth to heat up and cool down. and that overall, the trend is cooling, not warming. And while it's true some glaciers are shrinking, it's also true they grow back when it gets colder again like they did in the 1940s.

So did they fabricate data?

Every investigation so far has questioned certain aspects of their data and methods. They found no wrongdoing, but they did NOT endorse the results as "reliable and robust". On the contrary, as I mentioned, their data and or methods were found to be questionable by all 5 investigations that have been completed so far.
I'm not sure that's all of them. When this first popped up, the UK government said they were going to do a separate investigation of precisely those data and methods, and expected that it would take about 3 years to complete. I don't think those 3 years have passed yet.