Re-entry of Cosmos 50 explosion debris observed from Hong Kong?

From: Ted Molczan (
Date: Fri May 03 2013 - 16:58:30 UTC

  • Next message: Scott Tilley: "ST OBS May 3, 2013"

    My continuing effort to catalogue visually observed natural decays (aka re-entry) from Earth orbit has turned up an
    account of what appears to have been a sighting of re-entering debris of Cosmos 50 (1964-070A / 919), less than 15
    minutes after it exploded. I came to this tentative conclusion as described below.
    I have found private UFO journals and government UFO archives to be excellent sources of sightings of historical
    re-entries, including many that were unexplained or assumed to be meteoritic fireballs at the time. I will discuss that
    in greater detail later, when I present my findings. Since Cosmos 50 did not decay naturally (it was initiated with the
    help of some TNT), it does not strictly belong in my compilation, but it is certainly of historical interest, and the
    apparent Hong Kong sighting may help to confirm the time of its explosion.
    I found reprinted in the Flying Saucer Review, Mar-Apr 1965, Vol. II, No. 2, an article that appeared in the Hong Kong
    newspaper South China Morning Post, on 1964 November 6, which was an account of the sighting of a formation of luminous
    See pdf pg.29:
    <<<< Luminous objects over Peak
    "A resident of Garden Road last night reported sighting a formation of luminous objects travelling at great speed high
    above the Peak. "Mr. H. James, an ex-pilot, said he saw the objects shortly after 7 p.m. while watching a fireworks
    display in the vicinity. Describing the sighting, he said there were about 15 to 20 streaks each with a luminous trail
    and travelling at a 'fantastic speed'. "Asked whether he might have mistaken the objects with sparks from the Guy Fawkes
    display, Mr. James said they were nothing like fireworks. The group of lights were similar to aircraft in formation,
    pale in colour and travelling straight at tremendous speed. The formation took about 20 to 30 seconds to pass from
    sight, he said. He said his wife and their eight year-old daughter also saw the objects. "The Royal Observatory and the
    airport said last night they had had no reports of unidentified flying objects." >>>>
    The description is strongly consistent with the visual characteristics of re-entries. Most of the details required for
    analysis as a possible re-entry are present. 
    Given the date of the article and the difference between UTC and local standard time in effect, the sighting occurred on
    1964 Nov 05, shortly after 11:00 UTC. One deficiency is the lack of direction of travel. It is helpful that the sighting
    is described in relation to a local landmark (Victoria Peak, aka the Peak). The location of the observer at the time of
    the sighting was not stated precisely, but from the context he appears to have been at his residence on Garden Road.
    Google Map places the Peak ~550 m above 22.275833 N, 114.145367 E. Garden Road appears to run between 22.276101 N,
    114.155104 E, and 22.277521 N, 114.157421 E, at about 120 m AMSL. The former is closest to the Peak.
    >From the point on Garden Road closest to the Peak, it rises 23.7 deg above azimuth 268 deg. From the most distant point,
    it rises 19.3 deg above azimuth 261 deg. 
    USSTRATCOM lists only one object to have decayed on the date in question: Cosmos 50.
    The epoch of the latest USSTRATCOM TLE is 64303.72916435 (Oct 29). That necessitates a longer propagation than desirable
    for precise analysis of events on Nov 5, but barring any orbit manoeuvres, the prediction time uncertainty using SGP4
    probably is 10 min or so. At about 11:09 UTC the orbit would have passed within sight of the observer, 27 deg above
    azimuth 268 deg, and 29 deg above azimuth 261 deg - very good agreement in time and relation to the Peak.
    So far, so good, but the orbital altitude would have been about 233 km - far from re-entry. Even worse, there appeared
    to be no reason for the object to have decayed naturally by Nov 5, and forcing the issue resulted in decay one hour too
    early and on the wrong side of the observer's sky.
    Upon further study, I learned that the decay had indeed not been natural. Cosmos 50 was a Zenit-2 imagery reconnaissance
    satellite. It was launched from the Baikonur Cosmodrome on 1964 Oct 28 near 10:40 UTC. On Nov 5 UTC, it completed its
    nominal 8 day mission, and was set to land in the USSR, probably in the vicinity of 51 N, 51 E. The retrofire failed,
    stranding the satellite in orbit. An on-board destruct package was detonated to fragment the satellite, to encourage its
    complete disintegration upon re-entry, to prevent sensitive technology falling into U.S. hands.
    Some information on the explosion is found in the NASA publication: The History Of On-Orbit Satellite Fragmentations,
    14th Edition, Orbital Debris Program Office:
    TYPE: Payload
    LAUNCH DATE: 28.45 Oct 1964
    DRY MASS (KG): 4750
    MAIN BODY: Sphere-cylinder; 2.4 m diameter by 4.3 m length
    ATTITUDE CONTROL: Active, 3-axis
    ENERGY SOURCES: On-board propellants, 10 kg TNT explosive charge
    DATE: 5 Nov 1964 
    LOCATION: Unknown
    TIME: Unknown 
    ASSESSED CAUSE: Deliberate
    ALTITUDE: ~200 km
    No post-explosion TLEs of Cosmos 50 and its 95 pieces of debris have been published. Partial elements of the debris have
    been published and are available in USSTRATCOM's satellite catalogue, which reports that all decayed between 1964 Nov
    08-17 UTC.
    The RAE Table of Earth Satellites reports decay of the payload 8.0 days after launch, on day 1964 Nov 5.5, which I take
    to mean between 10:48 and 13:12 UTC. The Hong Kong sighting was early within that range, which raised the intriguing
    possibility that the observer saw an especially rare form of re-entry: the immediate aftermath of an explosion in orbit.
    Sven Grahn has plotted landing points of Zenit-2 spacecraft in 51-52 deg orbits, from information he found on the
    Novosti Kosmonavtiki Website:
    Based on the cluster of locations immediately to the north of the Caspian Sea, I take the typical landing site as ~51 N,
    51 E. For the nominal 8 day mission, there appear to be landing opportunities on consecutive revolutions, with
    respective landing ~7.93 d and ~7.99 d after launch. Those times are the most common listed in the RAE Table of Earth
    Satellites for 51-52 deg Zenit-2 missions, the latter attributed to Cosmos 50, as noted above, but to my knowledge still
    Presumably, detonation of the destruct package occurred as the s/c orbited over the vicinity of the planned landing
    site. Absent the precise time/location, I have taken this to be the point nearest ~51 N, 50 E.
    I determined the approximate time that the s/c reached this point with the aid of Alan Pickup's Satevo, which propagated
    the epoch 64303.72916435 TLE to the nearest earlier ascending node:
    1 70919U 64070A   64310.43067202  .00371317  78808-4  87433-4 0 90032
    2 70919  51.2306 161.1701 0028877 341.6657  18.2304 16.27524560  1306
    Note that I added 70000 to the catalogue number to help distinguish the TLE from the official ones derived from
    Next, I used SGP4 to propagate the above TLE to a position near 51 N, 50 E, which was predicted to occur at about 10:43
    UTC, which I took to be the approximate time of the explosion:
    Immediately before explosion                             188 X 234 km
    1 70919U 64070A   64310.44664352  .00371317  78808-4  87433-4 0    0
    2 70919  51.2306 161.0806 0028861 341.7342 111.8092 16.27536424    0
    The s/c was ~212 km above the geoid, not terribly far from apogee, so for simplicity, I took this to be the
    post-explosion apogee of the portion of the debris that might have been re-entrant within sight of Hong Kong about 14
    min later, and determined the required instantaneous change in eccentricity and mean motion: 
    Re-entrant ~14 min post explosion                       -234 X 207 km
    1 70919U 64070A   64310.44664352  .00000000  00000-0  00000-0 0    04
    2 70919  51.2306 161.0806 0347000 270.0000 184.0000 17.10000000    02
    >From the observer's assumed location in Hong-Kong, at about 10:57 UTC, this orbit would have passed 33 deg above azimuth
    268 deg, and 36 deg above azimuth 261 deg - very good agreement with the observed time and relation to the Peak. By this
    time, the debris would have descended to about 87 km - well below the altitude at which they would have been expected to
    incandesce - a key requirement to accept the re-entry hypothesis. 
    The time of 10:57 UTC is a bit earlier than the reported "shortly after" 11:00 UTC, but the difference is well within
    the typical error in measuring/reconstructing such events. Due to the low altitude of decays above the geoid,
    differences of time of passage of a few minutes greatly affect the track across the sky, due to Earth's rotation in the
    interim. For example, forcing the above orbit to pass just 3 min later, would greatly reduce the angular distance above
    the Peak. It would have passed 22 deg above azimuth 268 deg and 24 deg above azimuth 261 deg.
    The change in orbit between the above two TLEs represents a delta-V of about 130 m/s.
    Could the explosion have released sufficient energy to cause some debris to re-enter in time to be seen from Hong Kong?
    Recall that the above referenced NASA report stated that a 10 kg TNT charge was aboard, in addition to propellants.
    Based on one reference I found, 10 kg of TNT contains about 47 MJ of energy. That would be sufficient to change the
    velocity of 5562 kg of matter by 130 m/s, assuming all of the energy were directed to that purpose. That is greater than
    the mass of Cosmos 50, but in reality most of the energy must have gone into breaking up the object, as evidenced by the
    96 objects tracked by USSTRATCOM, that did not experience such a large delta-V. That said, my intuition is that at least
    a small fraction of debris was ejected into rapidly re-entrant orbits. A small number of fragments with a total mass of
    a few kilograms could have accounted for the Hong Kong sighting.
    Overall, I believe that the Hong Kong sighting is at least interesting, and I am inclined to believe that re-entering
    Cosmos 50 debris was seen. The evidence probably is insufficient to claim proof.
    It might be interesting and worthwhile to compare and contrast with the destruction of USA 193 in Feb 2008, as it passed
    over the Pacific Ocean, by a sea-launched SM-3 missile. Some fraction of the satellite's debris was almost immediately
    re-entrant, which resulted in an impressive light show over Prince George, B.C., about 15 min after impact. Most of the
    remaining debris decayed within days and weeks of the event.
    I am open to comments, corrections, suggestions for improvement, and any additional reports of Hong Kong sightings that
    may exist.
    Ted Molczan
    Seesat-l mailing list

    This archive was generated by hypermail 2b29 : Fri May 03 2013 - 16:59:27 UTC