Mag. differences between ir

4 Dec 1997 11:41:51 -0800

Hi Mike,

I'm glad you noticed the magnitude prediction difference -- my hard
work has paid off!  The reason the magnitudes are so much brighter
in v1.4 is two-fold: 1.  Improved modeling of the solar radiance profile,
and 2. a slightly steeper exponential curve of brightness vs. angle.

When the mirror angle is greater than 0.25 degrees, the brightness
falloff is modeled as an exponential fit to dozens of observer
data points.  Inside of 0.25 degrees, the MMA is directly reflecting
the solar disk, so the solar radiance profile is used.  You may be
surprised to learn that the brightness of the center of the sun is
about 3.5 times that of the edge -- that's about 1.35 visual mags.

However, these low-angle magnitude predictions are extremely
sensitive to Iridium satellite orientation errors.  Just a small
orientation error of 0.1 degrees in the wrong direction could
increase your mirror angle as much as 0.2 degrees.  If your
mirror angle was predicted to be 0.25 degrees (right at the limb),
the predicted magnitude could be as bright as -7.0.  But increase
that mirror angle to 0.45 degrees, and you're down to magnitude
-3.7 -- over 3 visual magnitudes!  (Mirror angles close to 0.25
degrees are actually the most sensitive case; the magnitude
"sensitivity" at larger mirror angles is much less).

That's why it's helpful to move to "ground zero" whenever
possible -- you have less sensitivity to orientation errors, and
a much better chance of seeing a "whopper" flare/glint.  An
orientation error of even 0.12 degrees will cost you at most
1.35 visual magnitudes if you're located on the flare centerline.
(.12 degrees orientation error translates to as much as 0.24
degrees mirror angle error -- enough to move you from the
center of the sun to the limb.)

There's always room to improve the theoretical brightness
curve vs. angle, so I encourage observers to report their
observations (and how they compared with prediction).  In
particular, more observations of flares are needed that have
predicted mirror angles between 0.2 and 1.0 degrees.  --Rob