From the NASA exoplanet archive system it can be seen that in the range of $0.02-0.06~\text{AU}$ distance an exoplanet is to its star, as distance drops down, the planet density increases linearly:

enter image description here

A similar inverse correlation between exoplanet mass VS distance exists too:

enter image description here

So in the mentioned distance range, as we get closer and closer to the star, exoplanets gets more heavier and more dense. Why it is so ? My intuition says that it should be in reverse. The closer to a star, an planet receives more heat, which should evaporate more mass from the planet, making things in an orbital body more gas/plasma like. But instead of that, planet mass accumulates and density increases too. This was unexpected to me. What are reasons behind that? Maybe closer to the star there are more events or collisions of orbital bodies happening which in the end accumulates mass/density or something completely different?

  • $\begingroup$ Have you made sure that both the mass and radius of the exoplanets in your list are measured? $\endgroup$
    – ProfRob
    Jan 17, 2022 at 16:53
  • $\begingroup$ In this question i'm just interested of reasons behind mass-distance correlation. $\endgroup$ Jan 17, 2022 at 16:57
  • $\begingroup$ Well, more massive giant planets are more dense, so your first graph would seem to follow from the second (modulo, some close-in giant planets may be inflated by insolation). $\endgroup$
    – ProfRob
    Jan 17, 2022 at 17:08
  • $\begingroup$ more massive giant planets are more dense In general it does not hold, Jupiter is a lot more heavier that Earth, but Jupiter is $4.2\times$ less dense than Earth. So it may depend what types of planets are compared (giants maybe?) and still I'm not sure if that also does not depends on exact distance range to star also. $\endgroup$ Jan 17, 2022 at 18:04
  • $\begingroup$ But why closer to the star more massive giants should exist ? It's because of accretion disk formation rules or so ? $\endgroup$ Jan 17, 2022 at 18:08

1 Answer 1


The situation is more complex than a simple correlation. A better plot would show the full mass distribution of discovered exoplanets versus their semi-major axis, but for that you need to plot the axes logarithmically. Here is one taken from data held at exoplanets.org. Note that the mass axis is actually "$M\sin i$" because the mass usually has a $\sin i$ ambiguity when measured by the radial velocity technique (as most are).

Mass distribution of known exoplanets

You can see in this plot that the correlation that you saw is actually a gap between the "hot Jupiters" and the "hot super-Earths" - this is known as the hot Neptune desert. The gap seems to close up at around $a=0.06$ au, but is wider at smaller separations.

The predominant explanation for this gap (it is by no means a settled issue) is that a low-mass hot-Jupiter will be unable to hang on to its envelope in the face of photoevaporative radiation from the star it is orbiting. They will lose that envelope and become smaller "super-Earths". The closer to the star the hot-Jupiter is then the more massive it needs to be to hang onto its envelope and this may account for the sloping lower boundary of the hot-Jupiter distribution in this plot.

The reason that the density behaves in a similar way is that, for giant exoplanets at least, there is a strong correlation between mass and density in the sense that more massive giant exoplanets are more dense.

  • $\begingroup$ Thanks, nice explanation. Never knew about Neptune desert. So it seems a system in that region is unstable,- there's a bifurcation point around $M \sin(i) \approx 0.1$. If cosmic body will reach this mass in Neptune desert,- you'll never know if it's own gravity will win and it will become next hot Jupiter or in reverse - star will blow already accumulated atmosphere away and it will degenerate into hot super-Earth. Concurrent processes and chaotic behavior in planet formation, so to say. Interesting. $\endgroup$ Feb 3, 2022 at 17:33

You must log in to answer this question.

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