Re: decay questions

Alan Pickup (
Sat, 26 Jul 1997 11:33:13 +0100

In message <>, Andre Beckus
<> writes
>  How exactly do you find out when a satellite will decay?  I have
>downloaded satevo, and I am able to create the evolutions, but what
>then?  Do they always reenter around their perigee?  How do you know
>where to look in the sky?
>Andre Beckus

Good questions, Andre.

As you know, SatEvo gives a predicted decay time and an evolution, with
predicted elsets at each northbound equator crossing until that time.

SatEvo assumes a smooth increase of atmospheric drag with time in making
its predictions from one orbit to the next. You could say that it is
generating a "mean" or average orbit.  In fact, The "real" orbit (the
correct term is "osculating" orbit) at any moment varies around this
mean and changes most rapidly when the drag is greatest. This (usually)
is when the object is closest to the Earth. For an eccentric object,
this is around perigee. Even if the object were in a circular orbit, it
is closest to the Earth's surface (and experiences the most drag) twice
each orbit as it crosses the "bulge" around the Earth's equator. And if
you had a low, circular, zero-inclination orbit (ie. around the Earth's
equator), the object would probably experience more drag over the sunlit
side of its orbit. During one of these episodes of increased drag, the
drag will pass some critical value and the object can be considered as
re-entering - you may expect visible fragmentation as it (possibly)
"burns up". Perhaps incorrectly, I refer to this as the "decay" of the
object though most satellites are strictly "decaying" from the moment
they are launched :-)

SatEvo's average orbit might lead to a particular predicted decay time,
but we might look at the final predicted orbit(s) more closely to
produce a better guess at the actual time. For example, the object is
unlikely to decay along the arc of its predicted orbit which is furthest
from the Earth, yet the mean orbit may give a decay time while the
object is moving along that sector of its orbit. It is more likely that
it would re-enter when it is next closest to the ground, or perhaps at
the previous close passage to the ground. A small change to the mean
orbit may change SatEvo's "predicted decay time" by (say) 15 minutes,
but may not change the time the satellite is close to the ground (our
guess at the actual decay time) by more than a few seconds. Also, of
course, the decay is not instantaneous - it may provide a visible
spectacle over hundreds of km of its path. And there is always the
possibility that the object may only just survive its predicted low-
point so that it skips off to decay a few thousand km further along.

If you are thinking that decay prediction is an inexact science, I'd be
the first to agree!

As to the question of where to look in the sky. The predicted elsets
that SatEvo generates may be plugged into your favourite prediction
program - I use Quicksat. Use any one of the elsets to find if/when the
object might be visible, and then recalulate a more accurate prediction
using the elset closest to this predicted time. Go out and watch,
allowing for a more-than-usual time and cross-track position error. If
you are lucky, and if the object is still in orbit and is illuminated,
you may see it. If you are *very* lucky (and even if it is not
illuminated), it may be re-entering at the time. When I wrote SatEvo
(back when we used snail-mail to exchange elsets), I intended that it
provide improved predictions to allow observation during the final days
of an object's life, rather than provide predictions of where to look
for any re-entry fireball. If you see the latter - regard it as a rare
and unlikely bonus. The only decays I have ever observed have been of
the Shuttle's external tank over Hawaii (and I didn't use SatEvo!).

 Alan Pickup | COSPAR site 2707:   55d53m48.7s N   3d11m51.2s W    156m asl
 Edinburgh   | Home:       +44 (0)131 477 9144
 Scotland    | SatEvo satellite page: