In my world an alien creature claims to be from an earth-like planet orbiting a 500 light-years away solar analog star. It tells us that the planet has x mass, y diameter, z orbital period, that specific atmospheric composition etc etc.

We're in the second half of 2020s so there are all of the future space telescopes (TESS, CHEOPS, JWST, PLATO and WFIRST). Obviously, we want to verify the creature's claim and we realize that we haven't discovered that planet yet.

Bearing in mind that the creature isn't lying and that we could detect that planet through the transit method (since we need to verify the atmospheric composition - thanks Mike Scott), would it be so unlikely to not have detected it before? If not, why?

(I can adjust the planet characteristics to make it plausible)

  • 1
    $\begingroup$ Instead of looking at something 500ly away which we can barely verify, look at his ship. Either the alien has a crazy life span, which can be verify by checking DNA. Has cyro-chambers which can be examined, or FTL which can also be examined and duplicated given enough time. Also surely the alien has travel logs we can examine. $\endgroup$
    – cybernard
    Commented Dec 5, 2017 at 4:05
  • $\begingroup$ Is there destructive wave interference of light on the cosmic scale that could explain that sometimes the light from that star isn't even visible, much less allowing detection of a shadow moving in front of that light? $\endgroup$
    – N2ition
    Commented Dec 5, 2017 at 4:57
  • 2
    $\begingroup$ @cybernard Verify a crazy lifespan by checking his DNA? We are nowhere near capable of analyzing DNA that well. $\endgroup$
    – Brilliand
    Commented Dec 5, 2017 at 23:10
  • $\begingroup$ @cybernard Perhaps also likely that one with a crazy lifespan that can travel that far has already manipulated DNA or its counterpart in order to achieve that lifespan? Would we be able to recognize the really clever coding modifications even? $\endgroup$
    – N2ition
    Commented Dec 6, 2017 at 0:45
  • $\begingroup$ @Brilliand in the second half of the 2020's (2025.5-2030) we will have that all nailed down. I am sure we will have more genetic breakthroughs, and what we know now will pail in comparison. $\endgroup$
    – cybernard
    Commented Dec 6, 2017 at 4:07

6 Answers 6


A few things.

First, there are such strong biases toward finding planets very close to their stars that an Earth-like planet around a Sun-like star is still a big reach. That is, only one (Kepler-452 b) has been found to date. But Kepler can't do this anymore, and radial velocity can't find a 10 cm/s amplitude signal at 1-year period. So, from that point of view, finding your planet in 10 years is doubtful just because it's hard.

Second, you way overestimate the power of the next generation of planet-finders. Of the ones you listed, only PLATO is capable of finding planets on Earth-like orbits (WFIRST maybe in certain cases but only for a statistical sample not any individual star). But PLATO 1) is, sadly, likely to end up being significantly less powerful than proposed due to budget issues, and 2) isn't scheduled to launch until 2026 at the soonest. Plus, to find a planet at 1 AU would require several years of observations so you're looking at 2030 at the earliest and then only if lots of things go right.

Third, it's actually extremely hard to detect 1-year orbital periods. This is because it's easy to introduce an artificial signal into data with certain characteristic frequencies (especially 1 day and 1 year), so astronomers are very skeptical about planets with exactly those orbital periods. So, your planet could effectively be hidden in plain sight, but have been filtered out of the data. (This is more true of the radial velocity method than the transit method but it could still happen).

Given these issues, your planet could be right next door to the Sun, even orbiting a very bright star, and we would almost certainly not find it before 2030 at the earliest.

To make your story plausible, I would push the timeline back 20+ years.

Also, just to emphasize my point, please note that I'm an astronomer who works in part on searching for extra-solar planets.

  • 1
    $\begingroup$ Also, just to emphasize my point, please note that I'm an astronomer who works in part on searching for extra-solar planets. $\endgroup$ Commented Dec 5, 2017 at 21:00
  • 1
    $\begingroup$ I would consider adding that to your answer to help give it credibility. Comments are second class citizens of course. $\endgroup$
    – Matt
    Commented Dec 6, 2017 at 13:24
  • 1
    $\begingroup$ @SeanRaymond Could be simply due to the position of the planet. If the target solar system is not edge on (or close to it) we probably won't be able to detect it at all. Correct me if I'm wrong, but wouldn't that leave us with direct imaging as the only option? Direct imaging of a Earth sized planet will require an enormous telescope - something that has a mirror that's measured in kilometers. $\endgroup$
    – ventsyv
    Commented Dec 6, 2017 at 17:24
  • $\begingroup$ @ventsyv -- My understanding of the setup was that the planet does indeed transit its star, i.e., the planet's orbital plane is aligned with our viewpoint from Earth. If that is not the case then the chances of finding this planet in the coming decades are pretty slim. The radial velocity signature of Earth is a 10 cm/s amplitude sine wave with a period of 1 year. Given measurement erorrs and systematic uncertainties (stellar fluctuations) that is not detectable on the near term (but who knows, maybe in 20 years). Direct imaging: doable in principle with a 10xELT (~500m diameter 'scope)... $\endgroup$ Commented Dec 6, 2017 at 19:03
  • 1
    $\begingroup$ @SeanRaymond Just read that the ESPRESSO instrument that's being installed on the VLT in Chile should have ~ 2cm/s precision. That's on 16m scope, but by late 2020s EELT with it's 39m mirror should be online so in theory that should be able to detect the planet no? $\endgroup$
    – ventsyv
    Commented Dec 6, 2017 at 19:32

Your simple answer is that no one has got round to checking that star yet. There are perhaps 2 million stars within 500 light years, of which about 150,000 are G-class like our sun. Corot and Kepler looked for transits on a lot of stars at once, but had fixed fields of view and couldn't scan the whole sky (and Corot couldn't have detected a planet that small). TESS will only be looking at a small number of bright stars in detail. PLATO would probably spot your planet, but isn't even scheduled to launch until 2026, and will then take several years to establish the periodicity of the transits (which will be about once a year, since it's an Earth-like planet around a G-class star).

  • 6
    $\begingroup$ "and will then take several years to establish the periodicity of the transits" > Yep. But would it be a good proof if the alien creature would tell us "hey, according to my calculations my planet will transit in front of its star in 142 hours"? $\endgroup$
    – Lupetto
    Commented Dec 4, 2017 at 13:20
  • 23
    $\begingroup$ @Mołot Even if they are just a prepared liar, anyone who is able to arrange something that looks convincingly like a stellar transit of a star 500ly away just to back up a story is still probably someone you should take seriously. $\endgroup$ Commented Dec 4, 2017 at 13:37
  • 13
    $\begingroup$ @Lupetto All it would really prove is that he has a better telescope than we do, not that he's from that particular planet. There's nothing we can see from 500 light years away that he couldn't have seen first if he has the technology, so he can't prove he's from that planet. $\endgroup$
    – Mike Scott
    Commented Dec 4, 2017 at 13:38
  • 4
    $\begingroup$ @Mołot Ah, you mean pointing at a convenient but unrelated stellar transit. That's more plausible, I guess. You're right, though; any creature that appears to be alive, let alone sentient, while exhibiting biology totally unrelated to the genetic makeup of any Earth creature... that'd be pretty strong evidence on its own. $\endgroup$ Commented Dec 4, 2017 at 13:42
  • 5
    $\begingroup$ @Lupetto Yes, if someone can predict an exoplanet transit that our best known instruments haven't yet detected, we know that something pretty odd has happened. $\endgroup$
    – Mike Scott
    Commented Dec 4, 2017 at 14:08

Another couple of possibilities:

Obscured transit

If another object passes between us (our telescopes) and the target star system at the same time as the remote transit, that transit will be obscured. For example retrograde movement of a planet in our solar system. The sun is also a good candidate - especially if the orbital period of the target planet is similar to our year, in which case each transit will be obscured. Notably, the telescopes named in the questions are based around Earth, the Earth's L2 point and lunar orbits, and so will all suffer from this blind spot.

Inclined orbit

Transits are only visible if we are in the same plane as the orbital plan of the exoplanet. The majority of exoplanets cannot be detected by their transits because of this. This can be leveraged: a transit that passes near the 'top' or 'bottom' of that star as seen from Earth would give an occlusion reduced in both duration and magnitude.

Solar activity of the target star

If the target star happens to flare at the same time as the transit, the darkening effect of the transiting planet could be reduced or countered by the flare. The duration of a transit is similar to that of a long flare, but bear in mind that this has a markedly more recognisable pattern (see this excellent explanation).

A combination of these or other factors would be enough to fool an automatic detection system into failing to signal a transit, while still being able to provide collaboration of the alien's claims under scrutiny of the data.

  • 10
    $\begingroup$ An "inclined" orbit is by far the most likely. Assuming that stellar orbital planes are uniformly randomly distributed (unknown), there's only a very, very small percentage that will happen to be lined with our solar system. This percentage is smaller for stars which are father away and for planets which orbit their star farther out. Thus, it's highest for stars which are close to us with large diameter/mass planets which orbit near their star. I put "inclined" in quotes because it's only inclined from our POV. It isn't, necessarily inclined from the POV of that stellar system. $\endgroup$
    – Makyen
    Commented Dec 4, 2017 at 18:47
  • $\begingroup$ If the orbit is inclined close to perpendicular to our line of sight, we wouldn't be able to use the radial velocity method, either, would we? So we'd be limited to detecting very small shifts in the star's position, which I'd guess would be well beyond current tech. $\endgroup$
    – jamesqf
    Commented Dec 4, 2017 at 18:52
  • $\begingroup$ @Makyen you're absolutely right. I interpret OP's question as including the requirement of the exoplanet being detectable [by transit] and used that as basis of my answer; please let me know if that's not clear. $\endgroup$
    – pbeentje
    Commented Dec 4, 2017 at 22:17
  • 2
    $\begingroup$ @pbeentje, I interpreted the OP's Q to be far more naive. More that it was coming from someone who didn't expect that the vast majority of exoplanets are not detectable with the transit method (or current methods, generally). However, I now realize my interpretation is biased. The question clearly sates "that we could detect that planet through the transit method". I interpreted that as someone stating a naive scientific opinion, not as it should be interpreted: as part of the world setup. However, I'd still argue it's sufficiently unlikely that it could strain suspension of disbelief. $\endgroup$
    – Makyen
    Commented Dec 4, 2017 at 22:32
  • $\begingroup$ @jamesqf, if the extrasolar system were inclined so that we were looking at it from a high angle compared to its ecliptic, you're right: it would be essentially impossible with current tech to detect the presence of a solar system. In our solar system, if all the planets were in perfect alignment, the barycenter would be about 500,000 km above the sun's surface, or less than 1 solar radius. At a distance over several light years, it would be almost impossible to visually detect that amount of wobble, and that's the theoretical maximum for the sun, when it would typically be a lot less. $\endgroup$ Commented Dec 4, 2017 at 22:55

The simply answer - we didn't looked that way yet and/or we didn't process the data yet.

GAIA is automatically discovering 5 thousands stars a day. Then it "alert" about certain stars that match set description. Then someone need to process that star data in the more specific way.

Then we can observe that region for a longer time/ with more precision instruments. But that take time. And trust me, astrophysics have a different approach to time. If some star they were observing "change" they knew it was a looooong time ago. And 500 light-years means a long time in travel to that place so I'm not sure we would even look for habitable planets in that distance.

The most famous Ultra-Deep Field photo taken by Hubble telescope had a observation period from September to January. Five months. And it's a tiny tiny fraction of the possibly directions we could look.

  • $\begingroup$ No one envisages travelling to any exoplanet, no matter how close, so the travel time is irrelevant. $\endgroup$
    – Mike Scott
    Commented Dec 4, 2017 at 16:27
  • $\begingroup$ @MikeScott Yes they do. They just do it "theoretically". It would be shame if in 300 we would be able to that and travel to a planet that is inhabitable. So now they do a preliminary of targets. $\endgroup$ Commented Dec 4, 2017 at 16:38
  • 1
    $\begingroup$ Gaia isn't designed to detect Earth-like exoplanets; the vast majority of its possible detections will be of Jupiter-mass or higher exoplanets. $\endgroup$
    – HDE 226868
    Commented Dec 5, 2017 at 3:14
  • $\begingroup$ @HDE226868 That's why I write it discover starts. The start then go through selection the data is then matched with other data about planets and data about spectrum and then they look more closely and so on and so on. That's why we have only 3 candidates to "earth like" planets. $\endgroup$ Commented Dec 5, 2017 at 8:15
  • $\begingroup$ @SZCZERZOKŁY My point is that Gaia isn't relevant here; the OP's planet is Earth-like. $\endgroup$
    – HDE 226868
    Commented Dec 5, 2017 at 13:18

If the star has significant "proper motion" (change in a position in our sky), and corresponding change in position of our Sol in the star system's sky, so that the degree to which our line of sight matches up with the plane of the extrasolar system's ecliptic changes significantly, what doesn't transit as much now, may transit a bit more in a few years.

Take a star 500 ly away (4.73*10^15). Say it's moving a relative 50 km/s perpendicular to our line of sight to it (transverse motion(. In 10 year it will have moved 0.00166782048 ly. So taking the inverse tangent, that's a .000191 degree change in both movement across our sky ("proper motion") and our angle of view of its ecliptic. For comparison Earth's angular diameter from the POV of Sol is about 17.5 arcseconds or 0.00486 degrees (about 109th the angular diameter of the Sun from Earth, .53 degrees). So maybe it was just on the edge of being able to be seen as transiting, a few years before, and now it's disk is crossing its parent by about 4% more of the planet's disk's diameter.

Hopefully I haven't messed up the math.

If it was closer than 500 ly or going faster, the change in angle would be greater, of course. Anything very fast would already be famous for speeding across our sky. For example, Barnard's star, a red dwarf, is zipping across our sky because it's just 6 ly away and has a transverse velocity of 90 km/s, so its proper motion is 10.3 arcseconds (0.00286 degrees) per year.


I am presuming you are writing some form of fiction. If you want confirmation to take less than a week, here's how. It's also all in the sigmas. A 'discovery' to an accuracy of 1 or 2 sigma would be put aside until more data is collected. Just not enough accuracy to count as a discovery. But, once there is a second-signal, i.e. an alien saying 'i come from that one.', a 1 or 2 sigma accurate stream of data could be counted as a discovery. So, it could conceivably be done in a few days, if conditions were JUST right.

  • $\begingroup$ Would you please re-read the question and then edit your answer to clarify how it answers the question. $\endgroup$ Commented Dec 5, 2017 at 23:36
  • $\begingroup$ I agree with Peregrine Rook, this seems to be answering a completely different question to the one OP is asking. $\endgroup$
    – F1Krazy
    Commented Dec 6, 2017 at 10:47
  • $\begingroup$ Welcome to WorldBuilding jerry! If you have a moment please take the tour and visit the help center to learn more about the site. Have fun! $\endgroup$
    – Secespitus
    Commented Dec 6, 2017 at 22:00

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .