[Steely Tom]

When you are trying to detect a “signal” that’s buried in a “noise” background that’s on the order of 100 times larger amplitude - and you’re only looking at a hundred cycles of reference oscillator - you can pretty much call it any way you want. .92 degree, .46 degree, negative .23 degree, who knows?

[Rocky]

Well, when you put it like that, the whole thing seems kind of silly.

[AndyTheBear]

you can pretty much call it any way you want. .92 degree, .46 degree, negative .23 degree, who knows?

Well by the Weak Climatetological Principle*, only one kind of call will get published.

* (joke allusion to weak anthropic principle)

[Tallguy]

Now, now Tom. Don’t use technical jargon when you are speaking to scientific illiterates (who also happen to be converts to the Church of Global Warming).

[PapaBear3625]

The big problem, as you note, is that these guys are trying to get grant money to study temperature variations of tenths of a degree, when most of their input data comes from somebody having taken an eyeball reading of a thermometer in a shed that was accurate to maybe +/- 2 degrees.

You also have to take into account the inclination of people to log a nice-sounding number rather than actually going out in crappy weather to take a reading.

Plus, land temperature readings do not really give a good measure of whether the Earth is warming up or not. For that, ocean measurements are better. Unfortunately, ocean measurements do not support the global warming hypothesis.