Friday, January 1, 2010

Explaining Away Climategate - 2

Climategate was triggered by the release of thousands of emails and computer programs from the UK Climactic Research Unit (CRU) late in 2009. See the video attempt at Explaining Away Climategate for a defence of the Warmists.

This is the second of a series of new Topic postings that detail the viewpoints of the major groups involved in the controversy. See first part here.

What did the Warmists at the UK CRU actually do that got them into trouble? Where is the "smoking gun"? For me, the most damning evidence is the following piece of computer code and a few emails from Phil Jones, CRU Director, that show he was a "true believer" in human-caused Global Warming and willing to do anything to prove his version of the "science" was correct - all in the service of saving Planet Earth.


The figure above (click on figure for larger version) shows actual computer code that was found among the thousands of computer files. Most readers of this Blog cannot understand computer code, but you can surely understand the comments programmers often insert into their code to remind themselves of what the code is supposed to do. They are highlighted in green.

The first says "Apply a VERY ARTIFICIAL correction for decline!!" The second says "fudge factor"and appears after a line of code that adds and subtracts a number to each of 20 years of temperatures. If you found that line of code in software used by a corporation to present their profits for the past 20 years, subtracting a bit from some years and adding a lot to others, you would lose confidence in those reports.

The self-described "fudge factor" was apparently applied to tree-ring proxy data to force-fit it to the instrumental (thermometer) data as part of the process of callibrating the proxy.

What is "proxy" data and why does it need to be calibrated? Well, relatively reliable instrumental data on temperature only goes back about 150 years. Scientists therefore have to use other methods to estimate temperatures back 500 or 1000 years or more.

One such proxy has to do with tree rings. As you know, you can determine the age of a tree by counting the rings. That is because trees grow more rapidly in the warm summer, producing a wider ring that is lighter in color than the winter growth. Warmer years produce wider rings than cooler years. Researchers take tree samples and carefully measure the ring widths and graph the average widths for each year. The resultant graph indicates which years were warmer and which cooler, but that information does not directly indicate the temperature until it is calibrated.

Calibration is done by comparing the last 150 years of the tree ring graph with a graph of instrumental temperature data. The tree ring data is adjusted and scaled until it fits well with the instrumental record. The idea is that, if the adjusted and scaled tree ring graph matches for the most recent 150 year period, then the previous 500 or 1000 years worth of data is correct.

Well, there was a problem with the calibration! The last 20 to 40 years of the calibrated tree ring proxy data diverged from the instrumental record. The measured temperatures went up while the tree ring proxy went down. The computer code above is an attempt to modify the proxy data to better fit the instrumental record. It is not clear if this particular piece of code was actually used to create the graph published in the IPCC 2001 report or if they simply did not plot the last 20 (or last 40) years of data as indicated in other comments in the computer programs. In either case, it seems the Fudge Factory at the UK CRU was bending the truth or avoiding it without fully disclosing what they had done.

After the fact, some appoligists for the Warmists have suggested that the divergence problem was due to air pollution or dryer-than-normal conditions that stunted the growth of the tree rings. That may well be true, but it should have been disclosed in the original reports.


Phil Jones resisted Freedom of Information Act (FOIA) efforts by Sceptic researchers to obtain the source data and details on how it was processed and he asked his fellow researchers in the UK and US to delete emails ahead of the FOIA requests. He also wrote an email where he wished for a Global Warming disaster "... so the science could be proved right, regardless of the consequences."

“… there is a Freedom of Information Act now in the UK, I think I'll delete the file … We also have a data protection act, which I will hide behind.”

"Mike, Can you delete any emails … with Keith re [IPCC 2007 report] Keith will do likewise. … also email Gene and get him to do the same? … Cheers Phil".

"If anything, I would like to see the climate change happen so the science could be proved right, regardless of the consequences."

The "Mike" above is Michael Mann, the Penn State climate researcher responsible for "Mike's Nature trick" and the "Keith" is Keith Briffa, a tree ring expert and Deputy Director at CRU. I will provide details of the "trick" and of Briffa's resistance to CRU machinations in the next installment of this series.

Ira Glickstein

1 comment:

CentralCoastRick said...

Here are some other interesting quotes from HARRY_READ_ME.txt in the data dump. This file is a sort of logbook of efforts made to obtain some results for 'Phil' at CRU. While I am familiar with chaos within graduate labs, I have never been in a lab where software procedures were lax on anything that might impact publishable results. In Physics & Computer Science (my experience) memories of fakey results linger for years among one's peers.

(These are included to give a flavor of the level of due diligence one might infer from within a group whose work impacts not only some publications - it may leverage trillions of govt. $)

" all correlations are >= 0.9 and all but one are >=0.96!
with 12 complete (100% identical) matches I think we can safely
say we are producing the data Tim produced. The variations can
be accounted for as rounding errors- - "

"Bear in mind that there is no working synthetic method for cloud, because Mark New lost the coefficients file and never found it again (despite searching on tape archives at UEA) and never recreated it. This hasn't mattered too much, because the synthetic cloud grids had not been discarded for 1901-95, and after 1995 sunshine data is used instead of cloud data anyway."

"But what are all those monthly files? DON'T KNOW, UNDOCUMENTED. Wherever I look, there are data files, no info about what they are other than their names. - - Edit: have just located a 'cld' directory in Mark New's disk, containing over 2000 files. Most however are binary and undocumented.."

"So, uhhhh.. what in tarnation is going on? Just how off-beam
are these datasets?!!"

And it goes on and on and on.