This evening I was watching the Big Dipper when I noticed a bright-enough object (it looked either like a satellite or a really slow shooting star) a little above the line that connects Dubhe and Megrez. It was visible for maybe 1.5 - 2 seconds, proceeding on a straight line out of the Dipper, and then disappeared.

I thought it was a satellite, but why did it disappear? There was not a cloud in the sky, so it couldn't have been hidden. Anyway, oddly enough, half a minute or less later there it was again. It started at the same place and goes AWOL in the same place. Now that was puzzling. But even more, I waited another half a minute and it appeared again for a third time. I was astonished - what could do that?

I'm not an expert, but I would rule out a satellite for the repetition of the trajectory in such a short time frame, as well as the shooting start as slow and impossible to have three in the same place at such a close timing. But what else?? Any clue? This is to me as close to UFO as I could ever realistically think about.

  • 2
    $\begingroup$ It could be a sequence of satellites, e.g. some newly-started starlink satellites which still follow somewhat a very similar orbit, thus reflect the sunlight to your observer position in the same place. Generally commenting on terrestrial or artificial phenomena, it needs a detailed time and place to make anything than general guesses $\endgroup$ Sep 8, 2023 at 19:58
  • 3
    $\begingroup$ You saw a set of Starlink satellites. I have seen numerous times (at least 3 different occasions) where satellite after satellite will flare low to the northern horizon for a second or two, then fade out. (They flare when the reflection of the Sun off of a surface is just right.) You might be able to confirm which set of Starlink satellites caused the flares by using Heavens-above.com. But with over 4000 Starlinks in orbit, it can be challenging to narrow it down. $\endgroup$
    – JohnHoltz
    Sep 8, 2023 at 20:04
  • 1
    $\begingroup$ At what (local) time? At what (approximate) latitude? E.g., to establish if a (LEO) satellite would be in the Earth shadow or not. $\endgroup$ Sep 9, 2023 at 11:47
  • $\begingroup$ Yes, as I mention in my answer it may be pretty straightforward to nail down exactly which satellites you saw if you can give an approximate lat/lon and time/date (note the time zone as well). The position in the sky is so specific I think there's a great chance of success here! $\endgroup$
    – uhoh
    Sep 10, 2023 at 1:59

2 Answers 2


An orbiting satellite becomes far more visible when there is a perfectly aligned angle between the sun, your eye, and a reflective surface on the satellite (usually the solar panels). Imagine looking in a very small mirror at arms' length to see a lamp across the room from you, if you want a more human-sized example. The light vanishes without going behind anything because the alignment changes as the satellite moves away from that perfect point. Imagine moving your mirror by an inch or so (without rotating it) and the lamp is no longer visible in the mirror. This is called a "satellite flare". Some larger satellites (or, more often, space stations) are visible even without the perfect reflective angle, but modern communication satellites are usually too small to pick out with the naked eye without a flare.

Starlink satellites are launched in clusters that very slowly spread out towards their target orbits. The cluster appears as a line of satellites (called a "train"). A few days after launch, if you're in the right place at the right time, you'll see a string of a dozen or more lights in a perfect row. A few days later and they've spread out so that they won't be visible at the same time, but if you're in the right place, you can watch the train go by, each one hitting that perfect reflective angle and moving past before the next one arrives. That's what you saw -- a series of reflective satellites coming by one after another.

  • 2
    $\begingroup$ This is it. With starlink's launch cadence it is pretty common. See satellitemap.space and spin the globe to see what trains there are. Currently I count around 7 of them. There are sites that allow you to try to observe them intentionally like findstarlink.com $\endgroup$ Sep 9, 2023 at 13:29
  • 3
    $\begingroup$ Given that many satellites appear as points of constant brightness over a long stretch of sky, I think that that's the reflection by their body without special alignment. I think they disappear because they enter Earth's shadow. A special alignment of solar panels or antennas cause a satellite flare. $\endgroup$ Sep 10, 2023 at 8:05
  • $\begingroup$ @Peter-ReinstateMonica Yeah, I should be more clear that some orbital objects can be seen at any angle. What I meant was that most satellites are invisible to the naked eye without a flare. $\endgroup$ Sep 11, 2023 at 14:44
  • $\begingroup$ @DarthPseudonym Is that so? Apparently, even the small Starlink satellites are well visible as such. $\endgroup$ Sep 11, 2023 at 18:06
  • $\begingroup$ @Peter-ReinstateMonica Starlinks are visible to the naked eye when they flare, and are visible most of the time on telescopes, but if they're visible to naked eye in a dark-sky area when not flaring, then it's news to me. $\endgroup$ Sep 11, 2023 at 19:03

You may have seen The A-Train or C-Train!

enter image description here

Source: File:A-Train and C-Train constellations - 2019-09

...proceeding on a straight line out of the Dipper, then disappeared.

These are in polar orbits which means their sightings could be associated with the Big Dipper. If there's more information (a rough lat/lon and time/date) we can look into that further.

Anyways, oddly enough, half a minute or less later there it is again. Starts at same place and goes AWOL same place.

If that's accurate, it narrows it down a bit. The two satellites shown in the image comprising the C-train are ~40 seconds apart, but the closest pair shown in the A-Train are about 2.5 minutes apart.

About it always being Starlinks:

While statistically because there are so many of them Starlink is often the answer to these types of one-satellite-following-another-following-another questions, it is far from the only game in town.

Previous examples of groups launched together then spreading out over time are CYGNSS, and, Planet Labs' "flocks" of Dove cubesats are also launched in clumps that over time slowly spread out in a similar fashion though perhaps taking more advantage of differential drag than the more propulsive Starlinks do.

But the last flock was launched in January 2023 so they will be spread out now.

And don't forget SpaceX's early Orbcomm-OG2 where a group of 6 and then a group of 11 satellites were each deployed together and then spread out in the same one-satellite-following-another-following-another fashion.

From https://space.stackexchange.com/q/13217/12102

enter image description here

  • $\begingroup$ Aren't starlinks unusually reflective for their size, too? I'm not sure Doves would even be visible during a flare, as tiny as they are. $\endgroup$ Sep 11, 2023 at 14:48
  • 1
    $\begingroup$ @DarthPseudonym you are right - Starlinks shoujld be much brighter than Doves for several reasons including that they are deployed at much lower altitude (and use their own propulsion to raise themselves), they're bigger and at least in the past their attitude during the raising phase was more favorable for reflecting light. But we don't have magnitude (brightness) specified here and our eyes are logarithmic so I didn't feel it was right to a priori exclude them. I think "Are Doves visible from the ground after deployment?" would be a great new question! $\endgroup$
    – uhoh
    Sep 11, 2023 at 21:16

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .