Like most Low Earth Orbits (LEOs), the Starlink satellites will eventually have their orbits decay and burn up in the Earth's atmosphere. How long, on average, will this take? I am assuming that there are zero factors other than gravity and aerodrag. For these purposes, the satelittes are at 550 km. So, how long can the SpaceX Starlink satellites survive before they deorbit?

  • 1
    $\begingroup$ 550km as most Starlink satellites are launched to. $\endgroup$ Commented Jun 26, 2022 at 20:39
  • $\begingroup$ I had always assumed that mass and cross sectional area would also be factors in the rate of orbital decay, in addition to altitude and solar activity. Is this not true? $\endgroup$
    – Mike H
    Commented Feb 3, 2023 at 19:56
  • $\begingroup$ Mass would not for sure. Cross sectional area matters a small amount. @MikeH $\endgroup$ Commented Mar 19, 2023 at 11:21
  • $\begingroup$ @MikeH mass and cross-sectional area both matter (and are frequently considered together as "ballistic coefficient." It's trivially provable that for a given drag force (determined by the shape and area of the body), drag acceleration (i.e. decay) scales with the inverse of mass. See e.g. the simplified model at the Wikipedia page for Orbital decay $\endgroup$
    – Erin Anne
    Commented Mar 19, 2023 at 11:29

1 Answer 1


Around 5 Years

Starlink satellites are launched into orbits between 335 and 354 miles above the Earth. SpaceX had originally planned to fly some of their constellation at 800 miles, but petitioned the FCC to change the architecture such that all their satellites fly below 380 miles. This was requested specifically to reduce the time it would take for a dead satellite to de-orbit.

SpaceFlight Now - FCC clears SpaceX to fly satellites at lower altitudes

SpaceX publishes a graph showing expected decay times as a function of altitude:

SpaceX graph of decay time vs altitude

The reason I said 'around' 5 years is that decay times can be affected by solar storms and other factors, and the satellites will be at slightly different orbits. But that wouldn't affect decay times at that altitude very much.

  • 12
    $\begingroup$ Space Junk is a problem. The satellites are going to have a fixed lifespan, and some are going to fail early. You don't want thousands of dead satellites floating around in orbit. The satellites won't deorbit so long as they are functional, as they have ion thrusters for maintaining their orbit. But once they die, you want them out of orbit as quickly as possible. $\endgroup$
    – Dan Hanson
    Commented Jun 26, 2022 at 20:25
  • 10
    $\begingroup$ Satellites are zipping around the Earth at very high speed. Satellites do collide. For example, in 2009 an Iridium satellite collided with a defunct Russian satellite. Two collisions happened in 2013, and another in 2019. The big problem here is that the collision between two satellites creates thousands of little tiny satellites, increasing collision risks exponentially. $\endgroup$
    – Dan Hanson
    Commented Jun 26, 2022 at 20:46
  • 8
    $\begingroup$ While space is big, LEO is not so big and is already crowded. $\endgroup$
    – ikrase
    Commented Jun 26, 2022 at 21:20
  • 5
    $\begingroup$ @BruceWayne they'll de-orbit 5 years after failing or running out of station keeping prop. Although since not being controllable makes them a hazard, SpaceX will probably deliberately deorbit them before they run dry. $\endgroup$ Commented Jun 27, 2022 at 13:51
  • 4
    $\begingroup$ @Tim link, and the thrusters are to keep the constellation structured, with even coverage, instead of a random chaotic swarm - keeping the satellites spread evenly in their orbits - plus make sure the lifetime is 5 years, regardless of solar weather bloating our atmosphere and accelerating orbital decay. $\endgroup$
    – SF.
    Commented Jun 27, 2022 at 19:10

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.