John Baez's profile photo
May 30, 2014
This post on my blog tackles the question "has global warming slowed?":

http://tinyurl.com/warming-galkowski-1

This is both a climate physics question and a statistics question.

The graph here shows the statistics question in simplified form.  The blue bars say how much global mean surface temperature exceeds the 1950-1980 average.  The red and dashed curves use a smoothing spline to fit a smooth curve to this data.  They both use the same algorithm: they differ only in the setting of a smoothing parameter. Which is correct? What does “correct” mean here?

As for the physics: more heat goes into the ocean than the atmosphere, so global mean surface temperature can be misleading.  A lot of heat goes down into the Pacific during La Niña years, and a lot comes back up during El Niño years.  We may be heading for an El Niño this autumn.  What will this do to surface temperatures?  We will see!

This post, by +Jan Galkowski, is the first of a 2-part series.  Since he's a statistician, the second part will dig deeper into the question of how to draw a curve through a bunch of points.  The first part is a broad overview of the question: has global warming slowed?  There's a truly massive bibliography if you want to learn more.
Photo