At the beginning of last year I was taking some pictures of the open cluster M67, in the constellation of Cancer, when I noticed that I had an additional slow-moving object appearing in the frames that I had recorded. To capture the star cluster I was using a 102mm (4 inch) aperture Celestron NexStar telescope with a digital SLR camera at prime focus (an old Fuji S2 Pro). I was taking short ten second duration frames at ISO 800 and subsequently went on to process these frames in DeepSkyStacker (an astronomical imaging software package). The final image can be found here on the Society’s Imaging Group Flickr page.
Figure 1, above, shows a composite of five of the frames. M67 is the cluster of stars at the bottom left of the picture, and north is roughly up. The unidentified object is seen as the elongated streaks and is heading eastwards towards M67. The trail is not complete, not because the satellite was intrinsically fading in and out, but because the camera was preparing to take the next 10 second frame in the sequence and so didn’t record the motion of the satellite.
So what is this object? I was initially puzzled because I am used to seeing earth satellites moving across the sky at a reasonable lick. Anyone who has seen the International Space Station will tell you that this bright object moves across the sky in a matter of minutes, and if you do happen to get such an object in your telescope it will cross your field of view in a flash! So what was happening here? The picture shown above is in fact a crop of the full frame of the camera which is 2.0 x 1.3 degrees. The angular distance travelled by the object is similar to the distance between the two stars HD75700 and HD75299, marked in Figure 1, and this is 0.62 degrees. The first frame started at 21:57:51 and the last ended at 22:00:27 GMT, a timespan of 2 minutes 36 seconds. So the object is moving at just under a 1/4 degree per minute. This is slow! What’s more, it is quite faint. I have marked the magnitudes of a couple of other stars in Figure 1, one at 9m.4 and one at 10m.5, and you can see that the object’s brightness is somewhere in between these two values.
I initially had a brief discussion about this with our expert on satellites at the ASE, Alan Pickup, and he suggested that this might be an earth satellite in a geosynchronous orbit. These orbits are used for communications satellites, and these objects orbit the earth once every sidereal day (which is 3 minutes 56 seconds shorter than a ‘normal’ day of 24 hours). So these satellites, approximately, move through 360 degrees in 24 x 60 = 1440 minutes, or 360/1440 = 1/4 degree per minute, in rough agreement with what I observed.
Later I showed these pictures to the Imaging Group and Alan said that he might be able to identify exactly which satellite this was – and he was successful! It turns out that this object is the remains of a Chinese Long March 3C rocket that was used to place a Beidou (or Compass G3) communications satellite into geosynchronous orbit on the 2nd June 2010. Its initial orbit measured 35582 x 202 km and had an orbital period of 627 minutes. By the 6th February 2013 (when I detected it) its orbit was 27080 x 168 km and its period was 469 minutes. Because of the highly elliptical nature of this orbit, this rocket will be moving slowly across the sky when it is further from the earth, and faster when it is close by. I think it is fortuitous that I happened to observe it move at 1/4 of a degree per minute (which happens to match the average speed of a geosynchronous satellite).

Figure 2 – Predicted and observed positions of the Long March rocket (green dots and red dashes respectively).
Figure 2, above, shows a map of the sky near M67 where the rocket was observed (this was produced by Alan using SkyMap Pro). Using the date and timings of my five frames he produced a prediction of the position of the rocket using the Heavensat software (green dots). My observations of the rocket have been roughly drawn on this map and are shown as red dashes. You can see that there is a pretty good agreement between the two! Alan reckons that there is an along-track residual error of about 10 seconds between the observed and predicted positions. I am not really surprised by this as I don’t think the clock on my camera would have been set that accurately. He notes also that the rocket would have entered the earth’s shadow at a point near the end of the final trail at the left of the figure, and sure enough the object does not appear in the frames that I took of M67 seconds after this.
So it just goes to show how things can turn up unexpectedly in astronomy, and how a bit of research can lead to an interesting result. Who would have thought I would have seen a Chinese rocket through my telescope!
Many thanks to Alan Pickup for suggesting that I look into these observations in more detail, and for providing predictions of the rocket’s positions, as well as his analysis.
Duncan Hale-Sutton