I was reading articles about Starlink and some concerned astronomers that SpaceX Starlink will lead to more space junk and eventually the death of ground based astronomy. I am just wondering how realistic these concerns are?

  • 1
    $\begingroup$ Welcome to stackexchange! Great question. It might be ruled as a duplicate of Has SpaceX made any commentary on StarLink posing a threat to astronomy? and you might like this question and this question. $\endgroup$
    – Christoph
    Commented Nov 19, 2019 at 15:23
  • $\begingroup$ Thanks Christoph for the links. Looks like the same question have been asked before. No proper answers though for the first question. $\endgroup$
    – Taha
    Commented Nov 19, 2019 at 15:31
  • $\begingroup$ It's going to be a big issue for ongoing and future sky surveys (such as LSST). Here is an image taken with a smaller survey camera (DECam on the 4-m Blanco at CTIO) during a pass of the recently launched batch of Starlink satellites $\endgroup$ Commented Nov 19, 2019 at 17:06
  • $\begingroup$ reddit.com/r/space/comments/eaxysu/… SpaceX made a statement. $\endgroup$ Commented Dec 15, 2019 at 13:30

1 Answer 1


I posted this a while back on Twitter (@rocketscient1st) when the first Starlink was launched: Conducted a simple analysis on a complete Starlink Phase 1 constellation. 1584 satellites: 24 planes with 66 satellites each. 53° inclination; 550 km altitude. I assumed the relative phasing as I have no direct information. I do not know the absolute magnitude of the Starlink satellites. I’m simply looking at lighting conditions. Some existing satellites are naked eye visible. Others are much dimmer and require binoculars. Others like old Iridiums flare.

I ran an analysis to determine number of satellites visible under the following conditions: Satellite in full sunlight. Observer with sun -6° below horizon (end of civil twilight). I also limited to satellites visible at least 5° above horizon. I did not try to account for reflection angle (sun-satellite-observer angle). This can significantly impact the brightness of the satellite, especially any flaring. Satellite attitude and surfaces will also affect the brightness. Analysis was for end of May at 38° latitude. This roughly equates to latitude of San Francisco, Washington, DC, Madrid, Istanbul, Seoul. I ran the analysis for 3 days, which should provide a decent average for a constellation of this size.

For satellites in a quasi-terminator orbit (much like just launched satellites) observer may potentially see a given satellite up to 5 times in one night. Satellites may stay in sunlight for up to 5 continuous days before orbit precesses enough to restart eclipses. Satellites in other planes (besides the quasi-terminator) may only be visible once during the night or not at all for any given night.

As others have mentioned, there is already a lot of satellites and junk up there. The impact of Starlink satellites will depend on just how bright they are. Initial reports by @Marco_Langbroek was that they could be seen naked eye, but latest passes required binoculars.

At the start and end of the night about 50 satellites are simultaneously visible that meet the criteria listed above. The satellites are distributed primarily in the direction that the sun set (in the evening) or will rise (in the morning). At midnight 18 satellites are still visible. The visible satellites are distributed north of the observer.


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.