Re-entry of Cosmos 2350 Proton r observed

Willie Koorts (
Thu, 7 May 1998 15:27:31 +0200 (GMT+0200)

Hi Folks

I would just like to pass on a copy of this private correspondence 
between myself and Alan Pickup to the list for interest.

--------I wrote--------

Hi Alan

We had recent reports from the public about a possible meteor/satellite 
entry around the South African/Namibian border and the west coast of 
Africa. (In fact they were reported as UFO sightings originally.)

The reported times ranged between 01H00 to 01H15 UT in Saturday morning May 2
1998.  Sightings ranged from multiple lights (front, middle and aft of 
this "vehicle"), burning parts and blue lights at the back of the craft and a
bright red mark on the night sky as well as sparks coming off. Others
reported seeing green as well.  One observer at the Noordoewer border post
and another 180km from Keetmanshoop could hear loud explosion/thunder sounds
and the former also allege to have felt the explosion. 

We concluded it to be a satellite entry and looking at your recent decay
list, probably Norad#25316 (98- 25 B) Cosmos 2350 Proton r  from your list
below posted to SeeSat-L.

>    #    Designation    Name               ----- Decay (1998) -----
>                                             Predicted   Actual?
>  25298   98- 19 G   Iridium 55 deb                    April 30.3
>  14034   83- 38 A   Cosmos 1456                       April 30.7
>  25316   98- 25 B   Cosmos 2350 Proton r                May  2.10
>  24083   94- 29 EE  STEP-2 Pegasus deb                  May  3
>  23910   86- 17 KY  Mir deb                    May  4.2

Does this sound possible and would it's last orbit fit the description?  
Could such a rocket body have produced such a sighting?
How do you determine the "actual" decay date?


--------To which Alan replied---------

Hello Willie,

I think that the decay of the Cosmos 2350 Proton rocket could well explain
your reports from the morning of May 2. Your description fits well with
those of earlier decays and the object was certainly big enough to give
quite a display. Did you see my decay list on SeeSat-L or on the newsgroup?

I reported my conclusion about this decay to SeeSat-L later on May 2:
>The final published elset ...
>C 2350 Proton r                                  146 x 136 km
>1 25316U 98025B   98121.87208462  .10826023  12723-4  13065-3 0   214
>2 25316  51.6148 280.5694 0007176 301.2392  59.3044 16.49391841   448
>... suggests that it was running 4.7 seconds early against my final
>prediction. The two previous elsets had only 0.07 sec early and 0.16 sec
>late. I suspect that it decayed at about May 2.10 (02:30 UTC), though it
>may have come down as much as 90 minutes earlier.

At the time I was a little skeptical about the final elset so I gave
greater weight to the previous two elsets which fitted so well with the
prediction for decay at May 2.10 (which I'd posted to SeeSat-L at around
19:00 on the 1st). However, when I forced the evolution through the final
elset, I got a decay time of May 2.0485 which is May 2 at 01:10 UTC - this
is why I made the comment about it coming down as much as 90 minutes
earlier than 02:30 UTC. At 01:10 UTC it would have been near latitude 22
deg S, 24 deg E having passed near Keetmanshoop two minutes before!

In fact, there must be uncertainty about the decay location along the orbit
since the orbit is not far from circular. However, its perigee was located
in the S hemisphere on its approach to the equator, and it would have been
running into the atmospheric bulge over the equator as it tracked over
southern Africa - these factors make that sector of its orbit the most
likely for decay.

My final SatEvo-predicted elset for the previous northbound equator
crossing is:

C 2350 Proton r                                  128 x 122 km
1 25316U 98025B   98121.99295038  .38501339  20253+1  18635-3 0 90212
2 25316  51.6132 279.8740 0004026 301.7589  58.2019 16.55442427   464

and the the ground track towards the end of that rev would have been:

   Time (UTC)  Latitude  Longitude

    01:00       47.4 S    14.1 W
    01:01       45.6 S     9.0 W
    01:02       43.6 S     4.2 W
    01:03       41.3 S     0.3 E
    01:04       38.9 S     4.4 E
    01:05       36.3 S     8.2 E
    01.06       33.6 S    11.8 E
    01:07       30.8 S    15.1 E
    01:08       27.8 S    18.3 E
    01:09       24.8 S    21.2 E
    01:10       21.8 S    24.0 E
    01:11       18.6 S    26.7 E
    01:12       15.5 S    29.2 E
    01:13       12.3 S    31.7 E
    01:14        9.0 S    34.1 E
    01:15        5.8 S    36.5 E
    01:16        2.5 S    38.8 E
    01:17        0.7 N    41.1 E
    01:18        4.0 N    43.4 E

I think that these times are unlikely to be out by more than 10 seconds and
the ground track should be very good. I don't give altitudes in this table,
but if it was producing a re-entry display it was probably below 80 km. The
fact that sonic effects were heard too suggests that it might have been
lower still (<60km?) but we are getting outside my experience/knowledge.

You ask how I produce the "Actual?" decay times. I collect all the
available elsets on all decaying objects and attempt to satisfy them by
fitting an orbital evolution through them using my SatEvo program (see my
WWW page). This gives me the decay time which I report in my SatEvo decay
lists. For the majority of objects, I have now way of checking whether my
calculations match reality - your report is one of the few exceptions and
I'm very grateful for it.

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


Thanks again Alan.

                        Willie Koorts   

       Cape Town,  Observatory   33d 56' 03"S   18d 28' 36"E   GMT + 2h
       Wellington, South Africa  33d 38' 56"S   19d 00' 52"E   GMT + 2h

       For - Amateur Astronomy - Telescope Making - Satellite Tracking -
                   Visit  ....