jump to navigation

Grouping the planets December 4, 2009

Posted by Jorge Candeias in Definition of planet, Dwarf planets, Giant planets, Sedna, Terminology.
Tags: , , , , , , , ,

Thoughts are like pistachios: you put one in your mouth (or in your head… doesn’t matter) and you’re on for a long ride. So, when I ranted about the terminology astronomers come up with, that sent my head spinning in new directions. However, as often happens, I’ll have to take a step back in order to explain it all properly.

As most people who deeply dislike the definition of planet the IAU came up with, particularly those who aren’t obsessed with Pluto (yeah, I know, there should be more of us), I think that a planet, like a human, a tree or a cloud, should be defined by what it is, i.e. by its own characteristics, and not by where it is. You don’t say that a human in space or under water is no longer a member of the human race, trees are trees no matter if they belong to a forest, are planted in urban streets or grow isolated in some field somewhere, and if something is composed by countless liquid or solid particles suspended in a gaseous medium, it’s a cloud, be it on Earth, on Venus or on 47 Ursae Majoris b. By the same kind of reasoning, to define what a planet is, where it is should matter not at all.

And the single most obvious thing that sets planets apart from other substellar objects is shape. Despite all their differences, they all show the same overall shape, a shape we know is due to a fundamental physical process that rounds them up if their mass is high enough to crunch them into a relatively low-energy state. Hence my definition for planet.

This means that all planets have differentiated and at least partially layered interiors, which implies the presence of geological processes going on at some point in their history (although you may have a tough time if you try to study the geology of gas giants. Still, they are differentiated like the others).

And this is where we come back to my little rant below.

So. Let’s suppose astronomers have the sense to start calling belt planets to what they currently call dwarf planets, using a location qualificative to set a subcategory that is based on location, and saving a size qualificative for another subcategory based on size. If they do, all of the planets that are currently known as dwarf planets would be both belt planets and dwarf planets, but you can use the term “dwarf planet” with other small planets that, as far as is known, do not reside in belts. Sedna, for instance, which is almost certainly a dwarf planet although it hasn’t yet been declared as such, was in that situation for a while. Its discovery was somewhat surprising, because it was too far to be a Kuiper Belt object but too close to be a denizen of the Oort Cloud… and for a while it was alone in its area. Actually, it still is very much alone out there. Sedna is dinamically classified as a detached object, together with only a dozen or so other known objects. If you consider that the outer edge of the Kuiper Belt lies at about 55 AU from the Sun and the theoretical inner limit of the Oort Cloud (also pretty much theoretical at this point) lies at about 2,000 AU, you can get a pretty good idea of how isolated Sedna really is out there. Even if you throw in the scattered disc objects to the mix, a relatively small population of objects with very elliptical orbits that make them travel from within the Kuiper Belt to large distances, sometimes well beyond 100 AU. Eris among them.

So, as far as we know for a fact, Sedna is not a belt planet because there’s no belt out there. And none is thought to exist. Astronomers think that is a very scarcely populated area, although they also say that discoveries out there are mostly a thing of the future. And yet, Sedna is undoubtedly a dwarf planet: with a diameter estimated at more than 1000 km, it’s definitely massive enough to have been rounded by its own gravity… and with a diameter of no more than 1600 km (yeah, the uncertainties are large), it’s definitely a small planet. Therefore a dwarf planet.

And then, of course, there’s PSR B1257+12 D. Also a dwarf planet which is not a belt planet, as far as we know.

But hang on: how can we draw a border between what’s an average sized planet as our own and a dwarf planet?

Well, ideally, we’d look at other planetary characteristics and find a suitable one. For instance, the presence of an atmosphere capable of creating all sorts of processes that transform the planet’s surface and of protecting it against at least some of the impactors. In other words, yet another layer of geology that sets living planets such as the Earth, Mars or Titan apart from pretty dead worlds like the Moon, Mercury or Mimas.

This, however, won’t work, because there’s a whole range of gases that remain gaseous at the various temperatures the distance from the Sun creates and that don’t get blown away to space, especially at large distances. As a consequence, Pluto has an atmosphere, at least during part of its orbit (temporary atmospheres are another reason why this is not a good criterion for the same reason the barycenter criterion is bad to define double planets. See below), Triton, also smallish, too, and Mercury, much larger, does not. And all the other criteria that were thrown back and forth during the early times of the planet redefinition debate (presence of satellites, presence of volcanism, etc.) are so flawed that I’m afraid we’d only have one alternative: go arbitrary on this. As I wrote several times, I really hate arbitrary groupings, but I have to admit that sometimes we just don’t have any good choice. This is one of them.

So the problem becomes finding a number that suits us well. Let’s see… I’m sure most people would want to keep Mercury as a medium planet, for all sorts of reasons, which gives us a maximum diameter for the limit of 4879 km. Most people would also want all the belt planets to fall in the dwarf planet category, which means that the limit has to be superior to the diameter of Eris: 2600 km. It would be nice to be a neat, round number, which leaves us with 4000 or 3000 km. Just pick one.

Personally, I prefer 4000. If you do it my way and add satellites to the mix as secondary planets (in italics), you end up with these three size-based subcategories of planet in the Solar system:

  1. Giant planets: Jupiter, Saturn, Uranus, Neptune. 4 in total.
  2. Medium planets: Earth, Venus, Mars, Ganymede, Titan, Mercury, Callisto. 7 in total, 3 of which secondary.
  3. Dwarf planets: Io, Moon, Europa, Triton, Eris, Pluto, Titania, Rhea, Oberon, Makemake, Iapetus, Charon, Umbriel, Ariel, Haumea, Dione, Tethys, Ceres, Enceladus, Miranda, Mimas plus a large number of other objects that are still in the lists of dwarf planet candidates. 21 for the time being, 16 of which secondary, a few dozens more already discovered (Sedna, Quaoar, etc.) and maybe many hundreds to be discovered.

(If you prefer setting the limit at 3000 km, Io, Moon and Europa go up to the medium planet zone, increading their numbers to 10; Dwarfs remain in the hundreds.)

And, according to location (in italics the belt planets except Charon, the only secondary, in bold the main planets):

  1. Inner planets: Mercury, Venus, Earth, Moon, Mars. 5 in total, one secondary.
  2. Asteroid belt planet: Ceres. 1 in total.
  3. Outer planets: Jupiter, Io, Europa, Ganymede, Callisto, Saturn, Mimas, Enceladus, Tethys, Dione, Rhea, Titan, Iapetus, Uranus, Miranda, Ariel, Umbriel, Titania, Oberon, Neptune, Triton. 21 in total, 17 of which secondary.
  4. Kuiper belt planets: Pluto, Charon, Haumea, Makemake. 4 for the time being, 1 secondary, more already discovered and waiting for classification, probably more yet to discover.
  5. Scattered disc planet: Eris. 1 for the time being, a couple more already discovered, pretty certainly more to discover.
  6. Detached planets: none as yet, but at least Sedna will most certainly make the list, sooner or later. And more discoveries are likely.

Workable? I think so. And much better than what we have today because not only this planet subdivision keeps the actual structure of the Solar System visible (small number of large objects, increasingly larger numbers of increasingly smaller objects; each zone has its own planets in the list), instead of simplifying it to the extreme as the 8-planet approach does, but it can also be neatly used with extrasolar planets, demanding very little information to start with. Which is good.



1. rikchik - December 4, 2009

Not that I’m proposing this, but what would you think about using tidal locking to differentiate primary planets from secondary planets? That would make Charon and Pluto co-secondaries of each other. (If Mercury had turned out to be locked to the sun like we thought 20 years ago, it would be a secondary of the sun.) Should be more reliable than barycenter position.

2. Jorge Candeias - December 4, 2009

That’s actually a pretty interesting idea. My only objection is that tidal locking is a process that takes a very long time to set in and can be disrupted relatively easily by strong impacts or close calls (which aren’t all that common after the initial stages of planetary formation, but can still happen thoughout the whole history of a planetary system), giving rise to a certain instability in categorization. But it certainly is better than the barycenter idea, in my humble opinion.

3. rikchik - December 4, 2009

Isn’t gravitational smoothing similarly slow and disruptable by impacts? I don’t know as much about the science as I should but my impression is that both processes are similarly predictable based on the mass and composition of the bodies involved.

4. Jorge Candeias - December 4, 2009

Yes, both processes are predictable based on the mass and composition of the bodies, but gravitational smoothing takes place at a much, much faster pace. Most planets are still in the early phases of planetary formation but are already round because they are molten by the heat involved in that formation. And that roundness is quite hard to disrupt later on. Evidence of that is the “death star moon”, Mimas, whose gigantic Herschel crater is almost one third of the satellite’s diameter. It is thought that had the impactor been a little more massive, the slamming would have shattered the sattelite. Even so, it remained spherical.

In the whole of the (known) Solar System, there’s only one object which has probably a story of roundness disrupted by a massive impact: Vesta. And it was pretty small to begin with.

5. rikchik - December 4, 2009

Thanks – that makes a lot of sense.

6. Bob Shepard - December 5, 2009

As you’ve noted, any kind of size cutoff to denote “dwarf” versus “medium” versus “giant” is going to be arbitrary, but I can definitely live with your system. I’m prepared to completely rethink how we classify these big, “round” objects.

Another possibility might mirror the spectral classes of stars (O, B, A, F, G, K and M). We could (using Earth=1) define a range of masses on a logarithmic scale.

A – 0.00001 (Mimas, Miranda)
B – 0.0001 (Ceres)
C – 0.001 (Pluto)
D – 0.01 (Mercury)
E – 0.1 (Venus, Mars)
F – 1 (Earth)
G – 10 (Saturn, Uranus, Neptune)
H – 100 (Jupiter)
I – 1000 (Super-Jupiters)

We could then define classes A, B and C as “dwarf”, D, E and F as “medium” and the rest as “giant”. Or we could simply speak of “F-class” planets, etc., if we want to be more precise.

Or we could use mnemonics, like this:

I – 0.00001 (mImas, mIranda – M is already taken, so use the second letter instead)
C – 0.0001 (Ceres)
P – 0.001 (Pluto)
M – 0.01 (Mercury)
V – 0.1 (Venus, Mars)
E – 1 (Earth)
N – 10 (Neptune, Uranus, Saturn)
J – 100 (Jupiter)
S – 1000 (Super-Jupiters)

This one might make sense, since astronomers are already talking about “super-earths” and “super-jupiters”. Why not a “Ceres-class” planet?

(For a more international flavor, we could use T for “Terra” instead of E for “Earth”.)

Yet another possibility, rather than using Earth as the measure, might be to take the logarithm of the mass, in kilograms:

10^19 – I – (mImas, mIranda)
10^20 – C – (Ceres)
10^21 – O – (Oberon, Titania)
10^22 – P – (Pluto, Triton, Earth’s Moon)
10^23 – M – (Mercury, Mars, Ganymede, Titan)
10^24 – E – (Earth, Venus)
10^25 – U – (Uranus)
10^26 – N – (Neptune, Saturn)
10^27 – J – (Jupiter)
10^28 – S – (Super-Jupiters)

I, C, O and P could be the “dwarf” planets, M and E the “medium” planets, and the rest “giant”.

Note that this list groups Earth and Venus together in one category, also Mercury and Mars in another. Some may find this more appealing than grouping Venus and Mars together, as the other lists do.

P.S. I’m using the following site as reference:


7. Jorge Candeias - December 5, 2009

Hi, Bob.

I don’t have any fundamental disagreement with any of your classification ideas, but I have to say I don’t see the need for such a detailed hierarchy like the ones you propose. Maybe such need will be apparent in the future, when the number of known planets rises to the millions, but for now I think something simpler is probably quite enough.

Anyway, this is giving me ideas for a new post. Stay tuned.

8. How about hot jupiters and super-earths? « Thousands of Planets - December 5, 2009

[…] and some more ideas coming out of both the posts themselves and the comment boxes. Particularly this comment by Bob Shepard, where he proposes a very detailed classification scheme for the planets, […]

9. uk superior papers - May 20, 2012

This is my research paper topic. How can there be liquid water if the temperature is over 3,000 degrees? Only liquid water on the side not facing the sun?

10. 어린이공방 - June 4, 2017

This info is priceless. How can I finbd out more?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: