So long, and thanks for all the Fish Planets

The captions in the NASA Ames Research Center video What Will Happen to NASA’s Kepler Spacecraft? read as follows:

  • NASA’s Kepler space telescope found thousands of planets outside our solar system. With its mission completed, the telescope will remain 94 million miles away in an orbit trailing Earth.

  • Kepler is traveling in a somewhat larger, slower orbit than Earth. Over time it will trail farther and farther behind.

  • By 2060, Kepler will fall so far behind that Earth will almost catch up to it.

  • As Earth approaches, its gravity will tug at the telescope and send Kepler into a closer, faster orbit around the Sun.

  • In its closer, faster orbit, Kepler will slowly speed ahead of our planet.

  • In 2117, the telescope will almost catch up to Earth from behind. Earth’s gravity will tug on the telescope as before, but this time pull it back into the wider, slower orbit.

  • For the foreseeable future the pattern repeats as Kepler is tugged inside and outside Earth’s orbit. The telescope never comes closer than a million miles to Earth — more than four times the distance from Earth to the Moon.

This suggest that roughly every 108 years Kepler's orbit will complete one full cycle, spending the first half, or one synodic period (also here) in a heliocentric orbit higher and slower than Earth's, then another synodic period in a faster, lower one.

Question: RIP Kepler, how shall we call your orbit? Does this cyclic flip-flop process have a name?

animated GIF:

Kepler Space Telescope, orbital flip-flop

  • 1
    $\begingroup$ Is this the same thing than between Janus and Epimetheus? en.wikipedia.org/wiki/Janus_(moon)#Orbit $\endgroup$ – Eth Oct 31 '18 at 10:24
  • $\begingroup$ @Eth that's a very, very interesting point... hmm... The video looks weird and a bit cartoonish/schematic-ish, and it's also in a fixed frame and those co-orbital moon diagrams in your article are in a rotating frame, so this is really interesting! That might be the answer. $\endgroup$ – uhoh Oct 31 '18 at 10:30

Kepler's is a horseshoe orbit:

A horseshoe orbit is a type of co-orbital motion of a small orbiting body relative to a larger orbiting body (such as Earth). The orbital period of the smaller body is very nearly the same as for the larger body, and its path appears to have a horseshoe shape as viewed from the larger object in a rotating reference frame.

The loop is not closed but will drift forward or backward slightly each time, so that the point it circles will appear to move smoothly along the larger body's orbit over a long period of time. When the object approaches the larger body closely at either end of its trajectory, its apparent direction changes. Over an entire cycle the center traces the outline of a horseshoe, with the larger body between the 'horns'.

  • 1
    $\begingroup$ My GIF from the animation does not "look" like a horseshoe orbit, even if it "sounds" like one. Like they say, "In space, no-one can hear you gallup". $\endgroup$ – uhoh Oct 31 '18 at 13:23
  • 1
    $\begingroup$ The video talks about the spacecraft passing the Earth. In a horseshoe orbit, this never happens. This is why I took the time to include the behavior as a GIF as well as quote the entire video. $\endgroup$ – uhoh Oct 31 '18 at 13:25
  • 5
    $\begingroup$ "its path appears to have a horseshoe shape as viewed from the larger object in a rotating reference frame" : Spin your head counter-clockwise while watching the video. $\endgroup$ – Russell Borogove Oct 31 '18 at 13:27
  • $\begingroup$ Wait, the animation shows the two side-by-side as they appear at the bottom of the screen. In space lingo that means they have the same true anomaly and that does not happen with horseshoe orbits. No amount of heat turning can make the animation fit. That doesn't necessarily mean that in reality the orbit isn't a horseshoe, just that the animation isn't a horseshoe. I am beginning to think the animation is just not right. About the horseshoe, it remains to be seen. $\endgroup$ – uhoh Oct 31 '18 at 13:36
  • 1
    $\begingroup$ @uhoh: I tend to agree, the animation looks wrong for a horseshoe orbit. In fact, it looks more like just an ordinary orbit around the Earth, viewed from a solar reference frame. (But it might not actually be that, either; it might just be tracing some arbitrary unphysical path "on rails".) The spacecraft is moving so slowly when it leaves the frame that, if the animation would continue further, the Earth would probably catch up to it. In any case, the Moon's orbit in the animation is definitely not to scale, either. $\endgroup$ – Ilmari Karonen Oct 31 '18 at 17:01

As Russell Borogove already noted, the NASA video in your question describes a classical horseshoe orbit. Whether that's a correct description of Kepler's actual orbit is another matter, which I will address below.

However, in any case, the animation shown starting at about 0:27 in the video does not illustrate a typical "horseshoe bounce" interaction. As noted in the comments, the animation shows the Earth and the spacecraft entering the frame side by side, with the same true anomaly, which is impossible for a horseshoe orbit.

In fact, it's clear that the animation is physically inaccurate in other ways too, like in its depiction of the Moon's orbit. It's hard to say whether the animation actually shows any physically realistic orbital interaction at all, or whether it's just an animated "artist's sketch" with the Earth and the Kepler spacecraft moving "on rails" along arbitrary non-physical paths.

A (probably) more accurate animation (created by Tony Dunn / Orbitsimulator.com) of Kepler's orbit can be found as an illustration to this seti.org article:

GIF animation of Kepler space telescope orbit by Tony Dunn / Orbitsimulator.com

As this animation shows, the orbit indeed features some horseshoe-like behavior, but it's also clear that it's not a "pure" horseshoe orbit. In particular, the eccentricity of Kepler's orbit causes it to trace a looping corkscrew path when viewed from a reference frame co-rotating with the Earth, instead of a clean horseshoe.

Also, while the animation shown above does feature a nice textbook "bounce" during the ~2060 encounter, the next encounter around the year 2110 instead shows the spacecraft passing the Earth and interacting with it, but continuing in a < 1y period orbit. Presumably, this is also due to the eccentricity (and possibly the inclination) or Kepler's orbit complicating the situation and causing the dynamics of the encounter to depend on the relative phases of Earth and Kepler in their orbits.

In any case, this is contrary to the NASA video, which claims that there should be another "bounce" around 2117, and that this process should continue "for the foreseeable future". At this point, I cannot definitely say which of these contradictory predictions is correct, although personally, on the balance of the evidence, I'd be more inclined to trust the Orbitsimulation.com animation, if only because it seems to make fewer simplifications and "artistic liberties". Of course, I cannot even completely rule out the possibility that both are wrong.

  • 1
    $\begingroup$ Excellent find by the way, complements to Tony Dunn / Orbitsimulator.com $\endgroup$ – uhoh Oct 31 '18 at 18:50
  • 2
    $\begingroup$ That's really fascinating! $\endgroup$ – Organic Marble Oct 31 '18 at 19:16
  • 2
    $\begingroup$ Thanks for the edit! The GIFFed section of the NASA video does not show a classical horseshoe either (see this explanation) but that's a secondary concern. Your GIF is making me very sleepy... my eyelids are feeling heavy... $\endgroup$ – uhoh Oct 31 '18 at 19:24

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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