From Earth, we can detect extrasolar planets by a number of methods; primary are detecting the wobble in a star's motion caused by a large orbiting planet, and the dimming of the star's light as the planet passes between it and Earth. Neither of these seems very amenable to quickly finding planets, nor to finding planets while at the star.

So: You're looking for habitable planets and have used your hyperspeed device to reach a distant G-class star. Now what can you do to find its solar system? I'm trying to see how to do this relatively quickly (on the order of weeks), with reasonable extrapolations of current technology (except that we can move really really fast). Scaling up the methods we currently use to find e.g. Kuiper belt objects don't really seem amenable to this time frame; could we even do a complete astronomical survey in a few weeks? Would it give us enough information to find the planets?

We're looking for habitable planets, so I'd be OK with just looking in the Goldilocks zone, but finding all of the planets would also be interesting.

EDIT: It looks like the simplest method would be to scan the sky from a bunch of points in (or near) the system, then compare the pictures to 3D-locate nearby points. I'm still not certain of the optical constraints here, though. Could someone who knows something about astrophotography comment on how big / sensitive the camera would have to be to pull this off?

  • 15
    $\begingroup$ If you can make a hyperdrive, I'm pretty sure you can have your computer do some image processing with your telescope... $\endgroup$ – Telastyn Apr 10 '15 at 16:00
  • $\begingroup$ Of primary concern is the shielding you're using to protect yourself against being consumed by stellar fusion, and the effect that said shielding has on your instrumentation. Oh, you mean you made it to the (not particularly close) vicinity of another star.... that may affect the answers. $\endgroup$ – Ben Voigt Apr 12 '15 at 21:01

It's pretty easy to spot objects in space that are moving differently from most of the stars. How quickly you can identify them depends almost entirely on the accuracy of your measurements—I would imagine that even modern instruments could do it within a few days, maybe even less; future technology could make this even faster. So, you find these things that are not stars, and they are probably planets.

Simply recognizing that some stellar bodies had different movement than the rest is how ancient civilizations recognized Mercury, Venus, Mars, Jupiter, and Saturn. That is, five other planets in our solar system were identified by people with no precision equipment, or even knowledge of orbital mechanics. Future scientists, who know what they're looking for, shouldn't have much trouble.


Hyperspeed your way to one side of the goldilocks zone, and have your computer use an optical telescope to record the position of all the brighter stars in a 360 x 360 sphere around the ship.

Then hyperspeed your way to the other side of the goldilocks zone and repeat the process. If a bright dot has appeared, disappeared, or has moved in the sky, then it's a planet and not a star (the parallax should be insignificant for stars but big for planets in the goldilocks zone). With trigonometry you should be able to identify the distance from the planet to the star to narrow down what planets you want to make a closer inspection of.

  • 1
    $\begingroup$ Maybe I'm overestimating the problem. I know the Hubble couldn't do a full sky survey in a few weeks, but how big a telescope would you actually need to find just planets? $\endgroup$ – gilgamec Apr 10 '15 at 17:08
  • 1
    $\begingroup$ @gilgamec From what I could find, a telescope of 6" should be able to show a planet within several AUs distance as a disc. I'd think a reasonable extrapolation of computer control and image processing would be able to cover the sky in less than a few weeks, but that's just my intuitive guess. $\endgroup$ – Darth Wedgius Apr 10 '15 at 17:41
  • 1
    $\begingroup$ @gilgamec: Assuming you're parked somewhere in the Goldilocks zone, you could do it with the naked eye if you're willing to wait around a bit. After all, it's perfectly possible to see (and I have seen) five other planets from Earth surface. Venus and Jupiter are almost impossible to miss, Mars and Saturn a bit more difficult. Seeing Mercury requires a clear horizon at sunrise or sunset, so it might be years between sightings. (But if you're a space colonist, do you reallt care about Mercury-like planets?) $\endgroup$ – jamesqf Apr 10 '15 at 18:30
  • 2
    $\begingroup$ We expect the plane of the ecliptic to be the same as the star's own rotation. Small planets may end up in odd orbits though while still being a small amount of angular momentum "off" compared to the norm, if the giant planet contains the bulk of it. $\endgroup$ – JDługosz Apr 10 '15 at 21:52
  • 6
    $\begingroup$ @gilgamec Hubble was designed for peering deep into space with an incredibly small field of view. If you're cruising over to a star system, you'd use a much wider field of view camera because the objects you are trying to spot are much brighter than the things Hubble is looking at, so you can build your detector and telescope differently. $\endgroup$ – Cort Ammon Apr 11 '15 at 6:00

You could quite easily detect the presence of any planets by taking an image of the sky in a specific direction, move some small distance, and take another image of the sky in the same direction. There is no need to move to the opposite end of the system for this (as suggested by Darth Wedgius); moving a fraction of the distance of the orbit around the star at the distance you are should be plenty enough for our purposes.

Stars are so far away that they won't have moved, or if you shifted the perspective enough they will all appear to have moved by the same amount in the same direction, which is trivial to detect by simply overlaying the images. However, planets, especially in the habitable zone, are going to have moved enough that the difference is relatively easy to spot. This is the parallax effect.

It is also very similar to how Pluto was discovered after its approximate position was known.

In the situation you're describing, you wouldn't know the position of any orbiting objects (the planets) but you would know that anything that has moved when comparing two pictures taken some distance apart but near each other in time must be moving relative to the background stars, and thus is a very good candidate for being a planet that you can aim your high-resolution equipment toward and take a closer look.

Once you know in what direction to look, move some distance orthogonally to your original direction of movement and look again. The two angles will form a triangle with a focal point at the planet, which with some trivial trigonometry will give you the distance from your current location to the planet. Since we can likely assume that you know your distance and direction to the central star, this tells you both how to get to the planet and how far from the star it currently is.

Given a high enough speed of movement between taking the images that the planet won't have moved significantly in its orbit in the interim, with a maximum of (I think) four such maneuvers, you will have located the planet in 3D space.


Planets will normally orbit a star in the same plane as its own rotation. This cuts down the area you need to search. If you are parked in the habitable zone, a planet will be in the top 20 or so brightest objects visible in the entire sky, so finding one or more in the ecliptic is a giveaway. Once you spot them, watching them over a few days will show their motion, and thus their orbit.


While at hyperspeed, you simply look out the window.

At speeds great enough to make traveling to another star feasible, the difference in distance between stars and planets would make any planets you could see seem to fly across the star-field.

This would make looking for planets much like looking for satellites while stargazing on earth.

  • 2
    $\begingroup$ Since it's not specified how this "hyperdrive" would work it's unknown how the outside would be percieved and whether it's possible to "see" stars at all. $\endgroup$ – Ghanima Apr 10 '15 at 20:06
  • $\begingroup$ +1 for knowing before you get there (driving without a detestation is dangerous). You don't stand inside something to find it; you look at it first. -Stop 4 light years out and take two pictures. $\endgroup$ – Mazura Apr 11 '15 at 1:50
  • $\begingroup$ @Mazura I think you mean a "destination". I rarely travel to places that I detest. Except when I had to go to my ex-wife's house to pick up the kids. $\endgroup$ – Jay Dec 17 '15 at 20:45

Light takes time to travel. Fortunately, for us, we have faster than light travel, meaning that on the way there, it will be somewhat trivial for us to see back in time (relative to the star system in question).

There are a couple of things that could happen depending on exactly how hyperdrive works. Assuming that normal space is visible at FTL, we could tote along an array of hardware not all that different from the Hubble telescope. Focusing it along the necessary bearing while moving is going to be hard but, hey, hyperdrive is a thing. This way you get to fast forward through taking a time lapse of many months worth of system history.

Assuming it is not visible, approach the star system to some arbitrary distance, say, 3 light years out. Take readings of the system at this point, and then cruise towards the system, say, half a year at a time. You'll still get a time lapse, just a little more fragmented. Once you get into the system, if you don't feel you've identified everything quite yet - which, you probably did - cruise back out a year or so from the system along any suspect bearing. Any planetary bodies should become pretty obvious.

These scenarios notably only really work in the event that hyperdrive doesn't take weeks to charge, and that the resources required for it don't skyrocket if you frequently stop and start.


You can do it most effectively by looking before you get there.

  1. Point your big telescope at the sky in the direction of your destination star. Record the image it captures.

  2. Repeat step 1 every month for a year or so.

  3. Combine the resulting images into one. Mark anything that appears round the star that is not another star, and using the varying positions, draw on a rough orbit.

  4. Use the orbits and sizes of the objects around the star to determine what they are likely to be:

    • Comets have trails and usually very eccentrically elliptical orbits. They are small and their trails are long.
    • Asteroids are small and usually have fairly circular orbits.
    • Planets' orbits can vary, but they are larger than comets and asteroids and appear brighter.
  5. Use the size of the star to work out its Goldilocks zone. Note any planets in this zone. These are your target planets.

  6. Compute, as a function of time to get to the star and orbital period, the position of the target planets when you get there.

  7. Go there. Investigate the planets and decide on one or more to colonize.


  • In Step 2, I say a year or so because planets' orbital periods vary but a year should give you enough data to be able to extrapolate their orbits.
  • Step 4 may not be completely accurate, so you'll want to check your conclusions as much as possible on the way, and of course when you get there.
  • $\begingroup$ This sounds great, but we're currently having problems doing this to Alpha Centauri, only four light years away. How big a telescope would we need to scan a star system hundreds of light years away? $\endgroup$ – gilgamec Apr 16 '15 at 17:58
  • $\begingroup$ @gilgamec Very. However, any society with hyperspeed drives and interstellar travel almost certainly has bigger telescopes than us. $\endgroup$ – ArtOfCode Apr 16 '15 at 18:04

With off the shelf technology today, it would take at most a matter of hours to find the habitable planets in a Goldilocks zone, as well as planets of comparable sizes / orbits to Jupiter and Saturn.

Simply take a 360 panorama of the sky using a standard camera. Take another one a little while (even fifteen minutes should be sufficient). Maybe take a second set of photos to cover the full celestial sphere.

Subtract one image from the other. Since the stars are essentially stationary, they will disappear from the combined image, leaving only objects that have moved a measurable distance in that fifteen minutes or so.

From there, it's quite simple to determine the orbital mechanics of the planets you've found and do a more in-depth survey.

You may - depending on the mechanics of your warp drive - want to reposition and do a second survey to make sure you didn't miss any planets obscured by the star. Similarly, instead of waiting to take your second photo you could warp to another location.

This is basically the same method that was used to find Pluto, although Pluto wasn't found using a digital camera / images.


Just look out of the ship's window. Planets are brighter than most stars, and the position of any star bright enough to be mistaken for a planet (e.g. Sirius viewed from sun) may be calculated in advance.

  • $\begingroup$ Planets are brighter than most stars - not in many cases - anything beyond Saturn was tough for astronomers to find. $\endgroup$ – HDE 226868 Apr 10 '15 at 19:37
  • $\begingroup$ @HDE226868 Uranus was discovered in 1781 and Neptune in 1846 so it was hardly rocket science. $\endgroup$ – David Richerby Apr 10 '15 at 21:03
  • $\begingroup$ @DavidRicherby After a couple centuries of studying the other planets with telescopes. Uranus was observed a couple times prior to Herschel, but it was mistaken as a star. Neptune was only found by studying Uranus. All of the other planets were noted by the ancients, but it took millennia for Uranus and Neptune to be discovered. $\endgroup$ – HDE 226868 Apr 10 '15 at 21:14
  • $\begingroup$ @HDE226868 But they were discovered long before modern technology existed. This suggests that, while it's considerably harder than just looking up at the sky, it's also not particularly difficult. For example, both were discovered before modern steelmaking, whereas this question is asking about how to find planets after we've invented interplanetary travel. $\endgroup$ – David Richerby Apr 10 '15 at 21:21
  • $\begingroup$ The answer implies that detecting a planet in an unknown location would involve simply "looking out of the ship's window". I actually agree with @HDE226868 in this case, it would be nowhere near that easy. For another counterexample, look up photos taken by the astronauts on the surface of the moon; there are no stars visible in them. While not a perfect comparison, the contrast you'd be dealing with between the spacecraft interior and the ambient starlight probably wouldn't be significantly different from this, so this gives you some idea of how hard it would be to visually locate a planet. $\endgroup$ – a CVn Apr 10 '15 at 21:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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