I remember a few years ago hearing that Jupiter was an anomaly in the landscape of exoplanets. Back then, most Jupiter-mass planets discovered were Hot Jupiters, orbiting very close to their host star.

In the last few years since I heard this, many new exoplanets have been discovered. Do we know now if there are many more Hot Jupiters and our own Jupiter is a rather rare occurrence, or if this was an observational bias and there are many more "Cold Jupiters"?

  • 6
    $\begingroup$ Jupiter is a possible solution to the Fermi Paradox, the rare Earth hypothesis. en.wikipedia.org/wiki/… It's hypothesized that Jupiter protects Earth from comets by sweeping up stray debris with its large gravitational field. Along with a big moon and a few other attributes. $\endgroup$ – Chloe Dec 5 '19 at 22:31
  • 15
    $\begingroup$ Note that all else equal, Hot Jupiters will be much easier to find than Cold Jupiters, using the techniques we use to observe planets in other solar systems. This is a common problem in astronomy :) $\endgroup$ – Luaan Dec 6 '19 at 10:08

It depends on how you define Jupiter analogues. There are several possible factors, including mass, eccentricity and orbital period cutoffs. Given there's no consistent definition, comparison of results between the various papers is difficult.

For example, the recent paper by Wittenmyer et al. considers "cool Jupiters" to be planets with masses greater than 0.3 Jupiters with orbital periods longer than 100 days. These planets do seem to be much more common than the hot Jupiters but this category is a lot broader than just "Jupiter analogues". It includes objects like HD 208487 b, a planet which would be located between Mercury and Venus in our Solar System and has a far more eccentric orbit (e=0.3) than any of our major planets: hardly a Jupiter analogue.

Many of the long period planets have high eccentricities. Imposing an eccentricity cutoff would tend to change things a bit. Other considerations might involve imposing upper limits on the mass, or a different lower limit. The paper notes that their conclusions about the rate of occurrence of Jupiter analogues is consistent with previous studies once the different criteria are imposed.


A recent study indicates that Cold Jupiters similar to Saturn and Jupiter greatly outnumber Hot Jupiters. The authors studied 18 years worth of data to find long-period exoplanets, that is planets far from their host star.

Cold Jupiters, being farther from their host star, have longer periods than Hot Jupiters. Therefore, they need to be observed over a longer time frame to see multiple transits. ETs would need to observe the Sun for 12 years before seeing two transits of Jupiter, and that still wouldn't be enough to confirm its presence.

Planets that are further out also induce smaller variations in the velocity of their host star, so the spectrum of the star is shifted by a smaller amount. Therefore higher resolution spectrographs are needed to detect Cold Jupiters using the radial velocity method.

Finding more Warm and Hot Jupiters at the beginning of the era of exoplanetary discovery was an observational bias due to the limits of the instrumentation available at the time, and the amount of time needed to find long-period exoplanets.

  • 5
    $\begingroup$ Is this because the closer a planet is to the star, the more regular the dips and thus easier to be detected? Something like Jupiter only dips in front of the Sun once every 12 years, so it only would have been seen once or twice since we found our first ever exoplanet in 1995 - very difficult to draw a conclusion from that!! $\endgroup$ – corsiKa Dec 5 '19 at 21:13
  • 12
    $\begingroup$ @corsiKa, it's because most exoplanet-detection methods require you to watch the parent star for one full period of the planet's orbit. To detect Jupiter, you need to watch the Sun for 12 years. $\endgroup$ – Mark Dec 6 '19 at 0:49
  • $\begingroup$ If you are using transits, you need to see the planet transit at least twice so you need to watch for a long time. If you are using radial velocity of the host star, a planet closer in will "tug" more at the star than a planet of equal mass further out, I think ? So you need better resolution spectroscopy to find them. $\endgroup$ – usernumber Dec 6 '19 at 10:36
  • 5
    $\begingroup$ @usernumber (and Mark) Actually, for transit you need at least three of them. When you see 2 dips, there's still no way to know whether those 2 dips were different planets, maybe 2 transits of something else entirely (possibly a solar system object), etc. Only with 3 transits you get 2 intervals, meaning you can confidently tell there is some rythm. $\endgroup$ – Emil Bode Dec 6 '19 at 18:04
  • 3
    $\begingroup$ Also, even for spectroscopy you need to be able to see some change. Just seeing a star move doesn't tell us anything, stars have all kinds of motions. Only when you can see there is some pattern to the movement, can you conclude there is a planet. $\endgroup$ – Emil Bode Dec 6 '19 at 18:09

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.