
An Open Letter To Professor Richard Muller and Team relating to the Berkley Earth Surface Temperature project
Tips and Advice: Learning from past mistakes
Dear Professor Muller and Team,
If you want your Berkley Earth Surface Temperature project to succeed and become the center of attention you need to learn from the vast number of mistakes Hansen and Jones have made with their temperature records. To aid this task I created a point by point list for you.
1) Any errors, however inconsequential, will be taken Very Seriously and accusations of fraud will be made.
2) If you adjust the raw data we will accuse you of fraudulently fiddling the figures whilst cooking the books.
3) If you don't adjust the raw data we will accuse you of fraudulently failing to account for station biases and UHI.
4) Homogenization is what Enron did.
5) If you rely on CLIMAT messages for the monthly updates this will cause a sharp station count drop after the first month. If that happens we will accuse you of fraudulently deleting stations to produce warming.
6) If you ever modify your algorithm and rerun it over the data so that some past monthly values change, we will accuse you of fraudulently rewriting written history.
7) By all means publish all your source code, but we will still accuse you of hiding the methodology for your adjustments.
8) If you publish results to your website and errors are found, we will accuse you of a Very Serious Error irregardless of severity (see point #1) and bemoan the press release you made about your results even though you won't remember making any press release about your results.
9) With regard to point #8 above, at extra cost and time to yourself you must employ someone to thoroughly check each monthly update before is is published online, even if this delays publication of the results till the end of the month. You might be surprised at this because no-one actually relies on such freshly published data anyway and aren't the many eyes of blog audit better than a single pair of eyes? Well that's irrelevant. See points #1 and #8
10) If you don't publish results promptly at the start of the month on the public website, but instead say publish the results to a private site for checks to be performed before release, we will accuse you of engaging in unscientific-like secrecy and massaging the data behind closed doors.
11) You can never adjust enough for UHI unless your record shows cooling.
12) You better not be using stations at airports. You'll find out why if you do.
13) You don't need to adjust for Time of Observation bias. That's the upwards one isn't it? Well we couldn't give a shit about that.
14) If any region/station shows a warming trend that doesn't match the raw data, and we can't understand why, we will accuse you of fraud and dismiss the entire record. Don't expect us to have to read anything to understand results.
15) You must provide all input datasets on your website. It's no good referencing NOAAs site and saying they "own" the GHCN data for example. I don't want their GHCN raw temperatures file, I want the one on your hard drive which you used for the analysis, even if you claim they are the same. If you don't do this we will accuse you of hiding the data and preventing us checking your results.
16) You are to blame for any station data your team uses. If we find out that a station you use is next to an AC Unit, we will conclude you personally planted the thermometer there to deliberately get warming.
17) We will treat your record as if no alternative exists. As if your record is the make or break of Something Really Important (see point #1) and we just can't check the results in any other way.
18) Always wear a lab coat while running your algorithm. I don't really know about the practicalities of how science is practiced in real life, but you sure as hell better meet the expectations I have gleaned off TV. Especially when this is science that is so important (see point #17).
20) A Stationary Audit should be completed prior to and after all work.
21) Your work is important (see point #17) so it bears special scrutiny - that's why our role as auditors is so important.
22) We don't need any scrutiny because our role isn't important.
23) In the unlikely event that I haven't wasted enough of your time forcing you to comply with the above rules, I also demand to see all emails you have sent or will send during the period 1950 to 2050 that contain any of these keywords
Any accusations of fraud and wrong-doing aimed at your team will be initially posted on blogs and in the media but will be eventually compiled by our think-tanks into glossy brochures. We don't check our own claims much.
You of course have nothing to lose and a lot to gain in this endeavor!
Good Luck,
The Auditors
Update: Additional advice from commenters:
Joel
24. In the event that you comply with all of the above, we will point out that a mere hundred-odd years of data is irrelevant next to the 4.5 billion year history of Earth. So why do you even bother?