Here's a public service announcement: Don't get too excited about Jim Lindgren's post that Some of the “Homogenized” Temperature Data is False (which relies on Willis Eschenbach's The Smoking Gun At Darwin Zero) until you have digested the debunking at Economist.com, Scepticism's limits.
A Personal Blog
by Michael Froomkin
Laurie Silvers & Mitchell Rubenstein Distinguished Professor of Law
University of Miami School of Law
My Publications | e-mail
All opinions on this blog are those of the author(s) and not their employer(s) unelss otherwise specified.
Who Reads Discourse.net?
Readers describe themselves.
Please join in.Reader Map
Recent Comments
- Brooks Fudenberg on I Voted
- Jermaine Chad Ingram on Some Thoughts about the Downballot (Voters’ Guide Part II: Judicial Retention Elections)
- C.E. Petit on I Voted
- Jane Moscowitz on I Voted
- Ally Figueroa on Some Thoughts about the Downballot (Voters’ Guide Part II: Judicial Retention Elections)
Subscribe to Blog via Email
Join 52 other subscribers
Is anyone but Orin Kerr worth reading over there? I include Volokh in that who is wildly inconsistent in the quality of his analysis and Adler who is pretty good analytically, but so predictable as to be not really worth reading.
Hardly a debunking to simply say the math is too hard for me, therefore it must be too hard for everybody that isn’t a scientist. Whether he’s right or wrong, that’s a silly argument. The fact is that amateurs do a LOT of very good science in the world and in some fields (astronomy for a great example) the amateurs have largely taken over all advancements and data collection in particular subfields because to pros can’t devote the instruments for the time involved. As I recall, it was an amateur that pointed out to NASA that its temp data suffered from a glitch in its compilation, which NASA admitted and fixed (though the bad version of the NASA data is still often cited by GW alarmists). And the CRU emails don’t exactly make one feel good about “peer review” retaining it’s reputable stamp, since they suggest a lot of tweaking of that process as well.
The real problem with the climate data is that ultimately it all has to be tweaked to be usable. Ever notice your nightly weather report and how they’ll point out the temps at various local places and they will vary by two or three degrees? That’s largely because all of the official weather stations are at airports (where they are needed, and where they can be secured), which can be located in completely different micro climates, and the thermometers can be located on either grass or tarmac (usually the later) and might be close to buildings. Ironically, the advances in measuring technology goes hand in hand with poorer placement as new highly accurate sensors are placed by necessity in locations that are more condusive to delivering power to them than accuracy. And they suffer from constructive changes at their sites that may be more airport related than sensor related.
So there’ll all this “bad” data that must be massaged. Ultimately, this is done by a computer program. A computer program, even one just doing math, is often based on a goal and can be as biased as your own grandmother. Often, that bias is unintentional, but it creeps in anyway. It’s an easy thing as scientists decide that a given data set might have a problem that needs adjusting. They moved the sensor, that means X. It’s the X that does all the work. Believe me, I’ve seen computer programs “just doing math” that showed particular trends no matter what data was fed into them. It is very, very easy to accidentaly tweak out objectivity from a working program and not even realize it. ANY TIME THERE IS A DECISION, THERE CAN BE BIAS, AND ANY COMPUTER PROGRAM IS LARGELY DECISIONS. Computers are not magical devices, though they do what they are told to do, so fast that they convince the unwary that they are. Hence the hockey stick graph that could be produced by even random data. Random data that exceeded a given threshold was kept, while other closer data was ignored. Presto: a hockey stick is produced by the good intentions of filtering out the noise in the data.
Additionally, as everyone admits, the historic data is pretty scant once you get past 1900, and much of the old data relies heavily on a combination of interpretation and conjecture. Once again, an inlet for bias.
There really is no way to know a lot of the things necessary to the pro or con GW argument. So it’s often just a question of faith. When one scientost points out that the polar bear population is as big as ever, and another says they are dying out at a huge rate, how do they know? How do YOU know they know? They don’t. They base their conclusions on localized datasets and extrapolate – they add in the X factor. We know, for example, that the Earth was warm enough in relatively recent history (hundreds of years ago) that seal rookeries were located on beaches that have long been covered with thick ice sheets. So what about sea levels then? What about lack of ice then? What about a lot of things? We just don’t know. So the scientists divide themselves into groups, criticise each other, interpret things differently, then demand that trillions of dollars be spent.
I’m not sure ANYBODY has justified any faith in any of it at this point.