zlacker

This unofficial number comes from a climate model published by the University of Maine as opposed to one created by a government agency or the WMO. Several agencies have their own climate models so “official” is very vague.

Generally in the US “official” numbers would come from the NOAA and the NCEI would be the one publishing them.

>>akisel+(OP)
So it’s a bunch of models?

>>spookt+E5
I have noticed lately that weather apps _predict_ your current weather. We no longer care about weather it's raining as a fact, it's just predicting rain or not. I've had days where it was raining, thunderstorm in fact, and google told me it was Sunny.

>>spookt+E5
The goal is to estimate an "average temperature" for Earth. What does that even mean? Are we talking about the temperature of the ground, the temperature of air at 2 m above the surface, the temperature of the Earth 100 m above the surface? In this case it's something like the second example: 2 m above the surface.

How do you ideally compute this average? You take the simultaneous temperature measurement at every point on Earth and take the average. Obviously that is not possible.

What do you do instead? You put together temperature readings where they exist, satellite data, coupled with past data and behaviors to make an estimate for the average temperature. How else could you possibly do it.

>>coding+t7
> "care about weather it's raining" funny typo (whether)

>>spookt+E5
Always has been. We haven't figured out how to get all of reality onto paper or in a computer yet.

>>spookt+E5
Yes, and those models can be contorted or misrepresented [1].

[1] https://docs.house.gov/meetings/SY/SY00/20160202/104399/HHRG...