June 23, 2003

Stormy Weather

A nice little blurb in the times today gives a feel for how meteorologists think about the weather in statistical frames.

June weather in New York City is explainable by the location and shape of the jet stream, but no explanation for the shape of the jet stream is offered. It is noted that wet junes tend to be followed by dry augusts (italics added). The tendency is a statistical observation. It does not need any process explanation associated with it, it is simply something that people who follow weather statistics closely have noted.

The absence of a process explanation is a source of debate among scientists. On the one hand there is the more empirical group who for patterns in weather data and then develop methods for making statments about how robust those patterns are. On the other hand there is the more theoretical group who base their science in the physics of the atmosphere (and ocean); this group develops models based in fundamental physics and roots their understanding in the underlying causal mechanisms. Ideally these two groups will get to the same answers, but their methods are fundamentally different and they don't always see eye-to-eye.

Myself, if called to comment, I would probably agree that it has been a bit wet this June, but the truth be told I haven't minded much. And if really pushed I might ask how the average rainfall has changed over the last decade or so. Is it possible that the average hasn't changed much but the variance is getting larger?...