Huh. Maybe weather-radar data are moot.

At Soil Science, we offer a variety of weather and modeling products. Many of the models require precipitation data, which is hard to supply because it's so localized--the ASOS at the airport knows precisely how much rain fell, but that says nothing about your farm 5 miles away.

Radar data would probably help; it's possible (though not trivial) to extrapolate precip estimate from weather radar. Trouble is, what with the Great Privatization Movement, brought to you by the a party which I won't name (but it's the one with the elephant), these data, collected by the government at taxpayer expense (for unassailably good reason), are extremely hard to get for free, because the private sector thought there was a revenue opportunity there. "Give us the data for free", they whined to their lobbyists, "and we'll sell it to the rubes!".

Now a paper in Science posits that simply tracking the signal qualities for 3G cell networks can yield decent atmospheric data. Clever!

I can't dig up a ref to the original paper; the team leader's publications page doesn't have it yet either. I saw a ref in The Register.