STS 107 could have been imaged by a KeyHole

From: Ted Molczan (
Date: Fri Feb 14 2003 - 02:52:02 EST

  • Next message: Peter Wakelin: "SATOBS 2003 February 13"

    [My apologizes to subscribers of both lists, who will receive this twice.
    Discussion of the satellite tracking and orbital analysis aspects of this
    message are on-topic for SeeSat-L or Columbia-L. Discussions relating to the
    circumstances of the Columbia accident and subsequent investigation are
    on-topic *only* on Columbia-L. To subscribe to Columbia-L, send a message with
    'subscribe' in the SUBJECT to]
    Late last week, a reporter from the Washington Post contacted me and asked
    whether or not there had been opportunities for Columbia to have been imaged
    by the KeyHole satellites. I agreed to perform an analysis using the official
    Columbia elements issued by NASA/OIG, and our hobbyist elements of USA 129. I
    did not analyze USA 116 or 161, because their visibility windows ended for
    Greg Roberts and Tony Beresford about the time Columbia's mission began.
    Moreover, at least one was likely to have manoeuvred, as USA 129 did on Jan
    17, the day after Columbia was launched.
    I have appended a copy of my report to the Washington Post. The resulting news
    story is available here:
    An earlier story by MSNBC, on possible KeyHole imaging of Columbia on STS 1 is
    available here:
    Ted Molczan
    E-mailed to Washington Post on 2003 Feb 08 at 04:29 UTC:
    I have determined that several times during STS 107, Columbia was in a
    favourable position to have been imaged by one of the NRO's KeyHole
    electro-optical imaging satellites.
    The six most favourable encounters varied in distance from 70 mi to 275 mi;
    estimated image resolution would have varied from about 2 to 4 inches. See the
    section Discussion for key factors affecting image quality.
    Spacecraft and Orbits
    Three KeyHoles were in orbit during STS 107, of which, USA 129 was tracked by
    experienced hobbyists, who produced the accurate orbital elements used in my
    USA 129 was launched on 1996 Dec 20. During STS 107 it was in a 172 mi x 612
    mi orbit, inclined 97.8 deg to the equator.  Its orbital period was about 97
    Columbia began its mission in a 176 mi x 186 mi orbit, which decayed to 170 mi
    x 177 mi by the end of the mission. Inclination was 39.0 deg to the equator.
    Its orbital period was about 90 min.
    Basis of Evaluation
    STS 107 was launched on 2003 Jan 16. On Jan 17, NASA discovered that a piece
    of foam insulation fell from the external tank and impacted Columbia. Assuming
    that would have been the motivation to seek KeyHole imagery of Columbia, I
    searched for all close encounters between Columbia and USA 129 from Jan 18
    through the end of the mission.
    Details of the six most favourable encounters are summarized below:
                     Range Vang   Phase  Res     Coordinates
     Date     EST     mi   deg/s   %     in.    Lat     Long
    Jan 18  07:46:29  115   2.1     8     3   33.4 N   38.4 W
    Jan 20  00:18:36   88   3.1     9     2   36.8 N   73.2 E
    Jan 25  00:18:30   70   3.1    99     2   38.4 N   72.0 E
    Jan 29  03:13:56  183   1.0    22     4   29.4 N   28.9 E
    Jan 29  22:42:18   99   2.6    93     2   27.6 N   94.5 E
    Jan 30  18:10:44  275   0.6    97     6   25.4 N  160.4 E
    The date and time of closest approach is stated as U.S.A eastern standard
    The range is the straight line distance between Columbia and USA 129, in
    statute miles.
    Vang is the angular velocity of Columbia relative to USA 129, in degrees per
    second. Extremely high angular velocity could causing excessive blurring,
    depending upon the design of the imaging system. USA 129's greatest angular
    velocity with respect to terrestrial targets in 1.6 deg/s; some of the rates
    with respect to Columbia were greater, but it is my guess that they would not
    have been a problem.
    Phase provides a rough indication of Columbia's illumination by the sun, as
    seen from USA 129. The values are the percent illumination of a sphere in the
    same location as Columbia. 
    Near-zero values indicate that USA 129 saw mostly Columbia's shadow. Values
    near 100 percent indicate that Columbia was nearly fully illuminated, which
    might have washed out details such as surface irregularities. Shadows could
    have been induced by optimizing Columbia's orientation . Values in between the
    extremes would have been best, because they should have produced sufficient
    shadows to reveal surface irregularities. 
    Res is an abbreviation of resolution, expressed in inches. This is derived
    from KeyHole's best ground resolution, widely believed to be about 4 in,
    presumably at perigee, about 172 mi above Earth.
    The coordinates are the latitude and longitude of the Earth directly below
    Distance and illumination are not the only factors affecting image quality.
    For example, if the tiles were the same colour as their underlying surface,
    then there might not be sufficient contrast to reveal any that might have
    fallen off.
    It might have been necessary to obtain images on more than one encounter in
    order to cover all of the at-risk surfaces.
    Since the cause of Columbia's break-up has yet to be determined, it is
    uncertain whether or not the underlying problem would have been apparent in
    any photograph. For example, a loose tile, could have appeared normal until it
    fell off.
    Since the astronauts had no means to repair or replace tiles, it could be
    argued that obtaining imagery would have been futile. It could also be argued
    that an image of the defect that caused the break-up would have been
    invaluable diagnostically, even if vehicle and crew could not have been saved.
    Unsubscribe from SeeSat-L by sending a message with 'unsubscribe'
    in the SUBJECT to

    This archive was generated by hypermail 2b29 : Fri Feb 14 2003 - 03:05:37 EST