Ed Cannon observed: >For whatever reasons, USSPACECOM's radar cross-sections for >the 72 Iridiums in orbit are all over the place, but my guess >is that really they're all the same size and shape. So I >calculated their average RCS using the values in the latest >quicksat.mag file, and the result was 10.9 square meters. I just took a quick look, and the variation is between 1.2 square meters and 50.7. Actually, this last is something of an outlyer -- the next largest is 27.6. But even if we take the entire range, the variation in rcs is a factor of 50.7/1.2 = 42.25 or, taking the logarithm to put it into radarspeak, 16.2 dB. This is entirely reasonable, indeed a bit low, for the angular monostatic rcs variation seen in real objects like airplanes. See "Radar Cross Section Lectures" by Allen E. Fuhs, ISBN 0915928884. Much the same variation is encountered for the optical "cross section," with the qualification that there the geometry is usually bistatic. The Iridiums themselves are a good example of this. It's regrettable that "rcs", which in the general case is really a frequency- and polarization-dependent tensor, is often represented as a scalar. The practice obscures some important complexities.