The following image shows trails of SpaceX’s Starlink satellites as seen in the sky above the Cerro Tololo Inter-American Observatory (CTIO) in Chile:

enter image description here

Image Source: SpaceX’s Newly Launched Starlink Satellites Block Galaxy Observation

In the top left corner of the image, it can be seen that two satellites travel in nearly parallel lines and not in coincident lines. Or in other words, it seems that the two satellites travel in two different orbits. What is the reason for this? Why is that one particular satellite not following the queue, even though all satellites have the same orbit parameters (radius, inclination, etc.)?

Here's a zoom in on the upper-left corner:

enter image description here

Here's another image:

enter image description here

Image Source: Elon Musk made a ‘satellite train’ in the night sky – Here’s how to see it

Here, everyone is going in a straight line!

I think Earth's rotation is not the culprit here, because if that was the case, then there should have been a shift for the rest of the satellites in the first image. So, why is there a deviation in one image and not in the other?

Thank you in advance.


3 Answers 3


Basically they are not quite in the same plane. As a satellite raises or lowers, not only does it change the relative position within an orbital plane, it also will slowly shift the longitude of the ascending node with respect to the other satellites, called the "Nodal Precession". In fact, this happens every day. There is a lot in this, but the bottom line is a satellite such as the International Space Station will rotate its longitude of the ascending node completely around the globe in about 2 months, hence why there is a few minutes difference in a launch window to the ISS from day to day. This effect is caused because the Earth isn't a perfect sphere. If two satellite are at the same inclination, but different altitudes, the rate of change of the nodal precession will differ slightly, causing the two satellites to not quite be in the same line.

I would need to identify exactly which satellites these were and see, but that pretty much has to be the answer as to why the two are different. Any difference of inclination will be microscopic, no more than a small fraction of a degree.

  • $\begingroup$ As a bonus factoid: thanks to nodal precession we can have Sun-Synchronous satellites :) $\endgroup$
    – gerrit
    Commented Nov 23, 2019 at 12:29

The streaks are due to extended exposures.

From a non-rotating earth, the satellites (this close to release) would follow each other in the same arc across the sky. That arc is the projection of the orbit and never changes.

From a rotating earth, that arc is not always in the same place: the earth rotates under it. So a satellite is moving along an arc that is itself moving with respect to the camera.

If you take an instantaneous snapshot, you see dots on the arc where it is at that moment.

If you take an extended image, the satellites will have moved a bit sideways by the time they reach the end of the page. And that causes the overlap.

  • $\begingroup$ at 300 km and 30 degrees from the horizon they should be moving about 0.5 degrees per second, and the Earth rotates about 0.004 degrees per second, I'm not sure that this explanation works quantitatively, but I'm also not sure I'm ambitious enough to do the calculation either i.sstatic.net/nmnHu.png $\endgroup$
    – uhoh
    Commented Nov 23, 2019 at 3:08
  • $\begingroup$ The motion along a 90 minute orbit and the motion across it (if polar) would be in a 24/1.5 = 16 ratio. This can change by a factor 2 or so due to inclination, but the ratio won’t be the roughly 125 from 0.5/0.004. Doesn’t an object 300 km away appear to move faster than just the rotation rate of the earth? $\endgroup$ Commented Nov 23, 2019 at 3:17
  • $\begingroup$ I need more coffee... will get back to you ;-) $\endgroup$
    – uhoh
    Commented Nov 23, 2019 at 3:18

Stalking Starlink's DartSat (Dark Energy Camera)

Starlink satellites trail through the field of view of the Dark Energy Camera. Credit: DELVE Survey/CTIO/AURA/NSF

From Stalking Starlink's DartSat, Field of view diameter is about 2.2 degrees

(even better image: https://i.sstatic.net/EBpeP.jpg from NYTimes)

NGC 5353/4 galaxy group + Starlink satellites, Lowell Observatory, ann19035a

An image of the NGC 5353/4 galaxy group made with a telescope at Lowell Observatory in Arizona, USA on the night of Saturday 25 May 2019. The diagonal lines running across the image are trails of reflected light left by more than 25 of the 60 recently launched Starlink satellites as they passed through the telescope’s field of view.

Although this image serves as an illustration of the impact of reflections from satellite constellations, please note that the density of these satellites is significantly higher in the days after launch (as seen here) and also that the satellites will diminish in brightness as they reach their final orbital altitude.

Credit: Victoria Girgis/Lowell Observatory

The first batch of ~60 Starlink satellites was a scrappy lot. Below is a plot from What are these four “debris” objects along with the Starlink satellites? about two weeks and then six weeks after launch, showing that they took substantially different paths to reach their ultimate target orbit.

Ignoring time zones, the photo was taken on 18-Nov-2019, a week after the 11-Nov-2019 launch. So in addition to whatever differences in their orbit accumulated due to initial order 0.1 to 1 m/s deployment differential, drag differences due to initial random attitudes and Earth's lumpy gravity field, each took it's own course once it stabilized in attitude and started climbing. So in any given snapshot during their individual journeys there should be no expectation for them to be in he same orbit.

I don't know if their destinations are all in a single orbital plane, but if so they will need to distribute themselves evenly by "phasing" which means some need to be a little higher and some a little lower in order to spread out along the track. Viewed from the side and not from underneath, different altitudes will appear as offsets in their paths, so this can be contributing as well.

Starlink satellites gaining altitude

  • $\begingroup$ If the difference was due to visible offsets in the orbits, wouldn’t that also show up in the “dots in a line” photo? $\endgroup$ Commented Nov 25, 2019 at 2:32
  • $\begingroup$ @BobJacobsen different photos, different times after launch, different altitudes, etc. The "dots in a line" photo was probably much sooner after deployment considering how close they are. The point of my answer is that things evolve over time. update: It was taken " by satellite tracker Marco Langbroek in Leiden, the Netherlands on May 24, 2019, just one day after SpaceX launched..." fyi I'm still working on that coffee... $\endgroup$
    – uhoh
    Commented Nov 25, 2019 at 9:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.