Greg Roberts recovers 82041C using planar search

From: Ted Molczan (
Date: Tue Dec 16 2003 - 21:22:57 EST

  • Next message: Kevin Fetter: "satellite elements file as mike alldat file is removed"

    Greg Roberts wrote in:
    > (1) Orbit plane search carried out for KH9-17 ( 82041C) and this
    >     object sighted, placing it 382 seconds early on predictions.
    >     Reasonably confident that it is the satellite in question.
    Nice work on the recovery, Greg!
    The object had not been seen in 85 days.
    1 13172U 82041C   03263.90151978  .00000800  00000-0  10480-3 0    00
    2 13172  95.9830 142.0298 0002000 341.9672  18.0326 14.80456533    09
    Since the orbit is not published, the only remedy for aged elements is to
    conduct a planar search to find the object, so that regular tracking can resume.
    Two factors cause us to lose objects. 
    First, the SGP4 orbit model (the basis of the 2-line elements) treats drag as a
    constant, even though it often varies a great deal. This makes it impossible to
    fit accurate orbits over long periods, and likewise makes it impossible to make
    accurate predictions very far into the future.
    The other factor in losing satellites is that we can only track them during
    periods when they are in sunlight and we are in darkness. We call these periods
    visibility windows. Certain objects spend months at a time outside the
    visibility of any of our observers.
    After several months of invisibility, objects in moderate to high drag orbits
    become lost to us - we no longer know their location within their orbit.
    Fortunately, predicting the location of an object's orbital plane is not nearly
    as sensitive to drag as predicting its location within the orbit. This is what
    makes possible the planar search technique.
    Think of an orbital plane as an imaginary ring around the Earth. A lost
    satellite is moving rapidly along the ring, at an unknown location. If we can
    stare at the ring for as long as it takes the satellite to circle the Earth,
    then we must eventually see it. 
    Depending upon the height of the orbit, it may circle the ring in less than 90
    minutes, or take many hours, so this is time consuming work, requiring
    considerable skill and patience.
    Since the Earth rotates under the orbit, the imaginary ring moves across the
    sky, so we cannot just aim in one place and expect to see the object - we must
    frequently adjust our aim to follow the ring.
    Most of us still use manually aimed binocular for searches, which is quite a
    chore. Greg Roberts has been pioneering the use of a computer-aimed video camera
    for both regular tracking and planar searching.
    Greg has put all that automation to good use, conducting a number of lengthy
    searches for some rather difficult objects. Perhaps the most challenging was
    Misty, aka AFP-731, aka USA 53, aka 90019B / 20516 - America's first LEO stealth
    Misty was last seen in 1997 by Russell Eberst as a faint unknown in this orbit:
    USA 53 (Misty)  18.0  4.0  0.0  1.5 v
    1 20516U 90019B   97284.23458324  .00000027  00000-0  70436-5 0    01
    2 20516  66.1631  65.2852 0005248 187.8717 231.2307 14.48751217    03
    By the time I identified Russell's unknown as Misty, in late 2000, it had most
    likely been de-orbited. Nevertheless, Greg willingly conducted many hours of
    planar searches for it.
    One can never be certain about a stealth sat, but Misty's failure to show up for
    Greg, tends to support the theory that it is no longer in orbit. In any case,
    negative findings can be as important as positive ones.
    Regardless of the technology used, planar searching has much to offer those who
    enjoy the thrill of the hunt.
    Ted Molczan
    To unsubscribe from SeeSat-L, send a message with 'unsubscribe'
    in the SUBJECT to
    List archived at

    This archive was generated by hypermail 2b29 : Tue Dec 16 2003 - 21:27:47 EST