The Solar System Is Weirder Than You Think

When I was a kid, the solar system was simple. There were nine planets and they all orbited in more-or-less circles around the sun. This same sun-and-a-handful-of-planets scheme repeated itself again and again throughout our galaxy, and these galaxies make up the universe. It’s a great story that’s easy to wrap your mind around, and of course it’s a great first approximation, except maybe that “nine planets” thing, which was just a fluke that we’ll examine shortly.

What’s happened since, however, is that telescopes have gotten significantly better, and many more bodies of all sorts have been discovered in the solar system which is awesome. But as a casual astronomy observer, I’ve given up hope of holding on to a simple mental model. The solar system is just too weird.

The Ancients and the Asteroids

It’s probably all Plato’s fault. While all of the ancient astronomers, from the Babylonians to the Egyptians, had noticed that some of the stars seemed to wander around relative to the others, it was Plato who posited that the fixed stars were located on one sphere, and the planets on another. That’s about as simple as you can get.

Ptolemy noticed that some of the planets seemed to wiggle around in their wanderings, and broke the planetary sphere by claiming that planets followed epicycles, and that each planet had its own. Others proposed epicycles on the epicycles, fitting the data better, but making for a very complicated system. Copernicus later managed to shrink the epicycles, explaining the motion of the planets by putting the sun at the center of the solar system. But it wouldn’t be until Kepler that we had a truly simple system again: all six planets, now including the earth, all orbited the sun in ellipses. Neat and tidy; the opposite of weird.

When the seventh planet, Uranus, was discovered, it was only a minor complication: just one more of the same thing that we already understood. Ditto the eighth planet, Ceres, in 1801. And then the ninth, Pallas, in 1802. And then the tenth planet, Juno, in 1804 and the eleventh, Vesta in 1807. Wait, what?

From this inspirational talk at 37c3

Today, we’d call these four planets “asteroids”, but for around 50 years, they were legit planets in their own right – except they all shared essentially the same orbit, which didn’t fit our mental schema at all. I don’t know if it was with relief or exasperation that as many, many more objects were discovered in the 1850, they all got collectively demoted from their “planet” status, and never spoken of in grade-school astronomy classes again. Our mental model of the solar system was simple once more.

But the asteroids are awesome. Vesta, for instance, suffered a gigantic collision, and over 15,000 tiny asteroids are thought to be chunks that split off, some of which rain down on the earth as meteorites. In that way, Vesta is the poster child of the weird solar system, even among the poster-children asteroids. And this is why NASA sending recent probes to Psyche and to the Trojan asteroids is particularly exciting.

Pluto and the TNOs

In the same era as the asteroid discoveries, the solar system’s eighth and outermost planet, Neptune, was discovered. The geopolitics of its discovery are a fascinating subject, but as far as our mental model of the solar system goes, it was just another planet.

Then comes Pluto. Discovered in 1930, it was a planet until 2006. It was a planet when I was a kid, and even though its orbit crossed inside of Neptune’s, we never gave it any thought. Its orbit takes it between 30 and 50 times as far from the sun as the earth and takes around 250 years. It’s a strange fluke of history that we discovered it just as telescopes were getting powerful enough, and as it was approaching closely. But for 76 glorious years, we convinced ourselves that it was a planet, because nine didn’t seem like too many.

Until, like with the asteroids, we discovered more of them. Telescopes and their sensors became powerful enough in the late 1990s and early 2000s to start to resolve more of what we now call trans-Neptunian objects (TNOs). And there are a lot of them. There is a tremendous variety of objects out there in the Kuiper Belt – a disc like the asteroid belt, only something like 100 times more massive. The Kuiper Belt is home to things like the close comets that have incredibly elliptical orbits that bring them close enough in to us to be visible, but also dwarf planets that are large enough to be rounded by their own gravity, and some of which even have moons.

Arrokoth is weird

It’s only luck, and the fact that Pluto is very reflective, that we found it first. If human telescope development were delayed by a hundred years, it would have been further out, and maybe another TNO would have been discovered first. For instance, Eris is bigger, heavier, and since New Horizons flew by Pluto is the largest body in the solar system that we haven’t visited. (Arrokoth, another TNO, also got the New Horizons treatment, and is probably the strangest object that we’ve visited.)

The point with the TNOs: just outside the radius of what shows up on my son’s poster of the solar system are thousands of known objects, and potentially tens or hundreds of thousands of unknown objects that are big enough to be seen with today’s telescopes. Some are “planets” even if they’re not planets, some are comets, some need classification. But things are weird in the Kuiper Belt.


Closer to home, between the orbits of Jupiter and Neptune, we find the centaurs, so named because they’re halfway between asteroids and comets, and halfway between the asteroid belt and the Kuiper belt.

Euler-Diagram bodies in the Solar System by Holf Weiher

Like Pluto, the first centaur that was ever discovered was quickly called a planet in the popular press, upon its discovery in 1977. This “tenth planet” was named Chiron. (I take the irony that neither the ninth nor the tenth planet would hold up more than 30 years as strong evidence for the weird-solar-system hypothesis.) From recent telescope images, Chiron even looks like it has a ring system.

Like the TNOs, there are very many known centaurs, with the vast majority discovered since 2010. And like the TNOs, we owe our knowledge of the centaurs to increased telescope performance. The Hawaiian Pan-STARRS project has found about half of them. Some think that Saturn’s moon Phoebe is a captured centaur, or that the previously-eighth-planet Ceres used to be one.

With the discovery of the centaurs, the weird zones of the solar system were no longer limited to the asteroid belt in the middle and the Kuiper belt on the “outside”, but also the centaurs that are whizzing around between the two.

Strange Moons and Near-Earth Objects

Surely nothing is weird about the solar system closer to home than the asteroids, right? For instance, in our orbit it’s just us and the moon, right? Well, that depends on your definition of “moon”.

Kamoʻoalewa appears to be a chunk of our actual moon that got chipped off millions of years ago, but has been trapped by Earth’s gravity for more than a century. Given the dynamics, Kamo’oalewa will be with us for at least another 100 years. China’s Tianwen-2 mission may go visit Kamo’oalewa, and then we’d know for sure if it’s made out of moon.

Kamo’oalewa isn’t our only quasi-satellite: 2023 FW13 appears to be circling the earth stably in an orbit that runs roughly between halfway to Venus and Mars respectively. (Note the Hawaiian names? That’s Pan-STARRS again.) So don’t say “moon”, but maybe “quasi-moon”. These weird objects are new enough that we don’t have a name pinned down yet, but we will almost certainly discover more of them in the near future.

Additionally, it turns out that of the roughly 12 million asteroids that we know the orbits for, some 35,000 are on orbits that put them inside 1.3 AU of the sun. (One astronomical unit is the radius of the earth’s orbit around the sun.) Of these, roughly 1,600 have a non-zero probability of colliding with the earth.

Indeed, the second-closest flyby of the earth took place just two weeks ago. (Spoiler: it missed.) 2024 LH1 was visibly spinning as it passed overhead, which was observed by it “blinking” 3.7 times per second. The closest recorded approach was back in 2020, and it missed us by only 370 km. It probably would have disintegrated in the atmosphere, but as with the TNOs and the centaurs, it’s a testament to our telescopes that this near miss was noticed at all. The fastest flyby? That would be totally-not-a-spaceship ’Oumuamua.

The Weird Universe

As with all simple models, the planets-orbiting-the-sun model of the solar system is incomplete. Over the past two hundred years, more planets have been added and subtracted from the list than we have planets at all. But it’s not just the planets – the reason for kicking Ceres, and later Pluto, off the list was the discovery of entirely new classes of objects that are hundreds to millions of times more numerous than just eight or nine planets. And we’re finding them everywhere we look, not just in the “belts”. As our observational astronomy gets better and better, the solar system gets weirder and weirder, and the last twenty years have seen the weirding accelerate.

And that’s just the solar system. Don’t get us started on the rogue planets that wander around completely untethered from any suns at all. There’s a lot more going on out there than we know, or than our orderly systems like to admit. At this point in astronomy, it’s more likely that we’ll discover an entirely new class of objects than a ninth planet. So, keep looking up and keep your mind open. Help keep the universe weird!

50 thoughts on “The Solar System Is Weirder Than You Think

  1. 9-planet people are just parroting the narrative of yesteryear. i think the bigger problem is that the educational system still thinks making kids learn large tables of useless (sometimes incorrect) information is a viable education strategy. when people were complaining about pluto, i was actually kind of happy that ceres got promoted.

    then again i think i would have added 2 subcategories under planets for both minor and major. so you could still technically call pluto a planet in casual conversation, correct enough in that context. if you wanted to be more specific you could call it a minor/dwarf planet. but iau decided to pull a lazy, subcatecorize the minors and keep the definition of planet unchanged. humans suck at categorizing things. in either case the list of useless facts that children would have to remember would go up.

    if you want to reclassify something, reclassify astroids as potatoids.

    1. Thinking relies on remembering information – to have something to think about.

      The modern method of teaching people to rely on external memory really dumbs you down, because it takes more time and feels more difficult to think about anything – and people are more likely to relegate the thinking to someone or something else. The lack of internalized knowledge prevents any kind of “a-ha” moment where your brain finds connections between facts that previously seemed to be just “useless tables of information”.

      1. The “information” is often little more than trivia used to help teach about connections.
        The CONNECTIONS are what actually matter.

        But it’s really hard to test comprehension and the ability to relate one set of connections to another set, so we mistakenly test the trivia instead.

        That isn’t to say some raw information isn’t useful.
        Knowing a gallon of water was defined as 8 pounds, a liter of water was defined as 1kg, and 1kg is about 2.2 pounds is surprisingly useful “trivia” in the real world.

        We still need to physically operate in the world, and I’m constantly surprised how few people know basic units that are actually important to be able to give directions or guess if they have enough gas/charge/time to get from A to B.

        There is a lighthouse across the bay from me that is just about 3 miles away.
        When I ask people about how far they think it is I get answers from 1/4 mile to 50 miles.
        Most people are in the 10+ mile range.

        I realize we offload all our navigation to our phones, but it isn’t just turn by turn directions that are needed. “I’ll be there in a half hour” is meaningless if they believe they need to go 45 miles.
        Conversely, when I say I’m 50 miles away and I’m correct, suddenly I look like a jerk when I arrive an hour later, because they don’t have any idea how long it takes to go 50 miles.

        1. >he ability to relate one set of connections to another set (…) so we mistakenly test the trivia instead.

          How do you suppose we retain the connections without retaining the “trivia” as you call it?

          A relates to B by connection, but if you don’t retain information about A and B then how do you infer the connection? If you don’t remember the starting or the ending point, then how can you ever go from A to B?

          1. Also mind, we’re currently at the point where new people aren’t even required to remember the basic 10×10 multiplication table. This is information that you need to even understand facts like a gallon is 3.8 liters.

            If you say you’re “50 miles away”, that has no meaning unless you can add and multiply by rote memory – but that is exactly the kind of rote learning that is shunned today. You’re not even supposed to know that much – it’s more important that you can google it.

          2. @Dude, are we? Apparently my daughter has decided she wants this as homework, printed some worksheets and fakes weekly tests all by herself without me noticing.

      2. If you only memorize a list of information, that doesn’t help you make links. Kids often first learn the alphabet as a series of letters, and then if you ask them to say it backwards or say what’s the 15th letter or the letter 7 before V, they have to page forwards through the sequence in order to find it. So they only linked together the sequence and the pronunciation of the symbols. But at least that’s still something you need to know.

        A common example of rote memorization not helping someone actually understand what they’re memorizing is when someone “learns” history by being able to recite on command that at 7:23 pm on November 3rd 1634 in Husegonanamdis town in the country of Maidep, Johannes Freyburg was beheaded while wearing a red shirt. But they have no idea why those facts were significant. Who was he? Was there some context for this beheading? Did the red signify something, or whether it was before or after sunset? Etc. They should have sped through the facts to hammer the concepts, imo.

        1. Remembering lists is not useful it itself, but for anyone who needs such information they are more able to re-structure such lists any way they want.

          I have trouble remembering the numbers of months in various languages, so I do the same thing – recite in order. Having learned to do that once, I’ve made the point to remember a few, so I don’t have to start from January every time. Learning the entire sequence once has made it possible to build up on my own. I don’t remember the number and name of every month, but I can quickly derive each, thanks to having learned the rote sequence once.

    2. Like when we were told in Chemistry that instead of actually using the handy reference chart designed to help us on tests, we would be attempting to memorize said chart and draw it from memory…

    3. Fair enough, classification schemes become more specific and consistent. But a planet being a celestial object which “clears its own orbit” is still pretty arbitrary. None of these planets really cleared their orbits, as we’ve seen. What is the size limit on the remaining objects? How stable do they have to be? What’s the time limit for clearing the orbit?

      I will be interesting if someday we detect gas giant-sized Kuiper belt or Oort cloud objects (shout out to Peter Watts). Seems far-fetched, but it could still happen one day. It would be kinda funny to have something the size and significance of Jupiter but it’s so far out that it would never clear an orbit to attain planetary status. It would probably mean we’d need to refine the rules a bit more, which is of course just fine

      1. The idea comes from Stern & Levison’s paper “Regarding the Criteria for Planethood and Proposed Planetary Classification Schemes,” and basically from long-used techniques for estimating planetary scattering probabilities. Note that Stern & Levison didn’t use the term “clear the neighborhood” or “clear its own orbit” – the plot title was “dynamical dominance criteria example.”

        In other words, it’s not arbitrary.

        “It would be kinda funny to have something the size and significance of Jupiter but it’s so far out that it would never clear an orbit to attain planetary status.”

        No, you’re misunderstanding – the phrasing means that it’s cleared its orbit of objects *of comparable size*.

        Basically, the “dynamical dominance” criteria is essentially just “mass squared/(orbital distance^(3/2))”, with the “nominal” cutoff being something like a *billion times* smaller value than Jupiter’s – you can fudge that around a bit, but it doesn’t matter.

        In other words, it’s not possible for something the size of Jupiter to not be a planet, regardless of where it is. It doesn’t matter that there are buckets of wimpy objects around it – Earth doesn’t become “not a planet” because a comet strewed dust all through the orbit.

    4. “but iau decided to pull a lazy,”

      Sigh. It’s not lazy. Pluto was never going to be a planet under any sane definition – the IAU wasn’t looking for a “make people happy” definition. It was a way to help people categorize things they discover so you can study stuff like planetary formation. Pluto *was not* formed by the same types of processes and dynamics as the other planets, so it was never going to be one.

      For instance, there’s strong evidence that the more planets you have in a system, the lower the eccentricity of those planets is. In order to make that correlation, you have to have a common definition of what “planet” is, and if you look from both simulations and in the Solar System, the “clear your neighborhood”/”dynamical dominance” definition very clearly separates things.

      If you go with “object big enough to be pulled into a spheroid by gravity” you’ll end up with dozens to hundreds of objects in *any* solar system regardless of its formation dynamics. You’d need a subcategory for “planets that dominate the dynamics” and “major/minor” vs “planet/dwarf planet” is a silly hill to die on.

      It might seem weird to point to Mercury and say “the fact that Mercury exists tells us something about the Solar System formation” but it does. In fact, the ‘least planetlike’ object in the Solar System is actually *Mars* and it’s still orders of magnitude above Pluto.

  2. “Chiron might be the second-largest “planet” in the solar system that has a ring system”

    Perhaps I missed something from the context, but this doesn’t make sense to me, as Jupiter, Saturn, Uranus, and Neptune have ring systems, and they are all much larger than Chiron.

      1. “Non-planet planets”… this phrase, while being a technically accurate metaphor under the new classification system is a symptom of why the new classification system is a bad idea or poorly implemented, if for no other reason that the nomenclature really doesn’t advance communication; perhaps even ret ards it; and the criticism to that end does have some merit.

    1. I thought that for such a classification the barycenter needed to be outside of the surface of both celestial bodies. The barycenter of the Earth-Moon system is a couple thousand kilometers underground.

    2. i think the deal breaker for that one is that the barycenter of the earth-moon system is still inside earth. pluto would be a better case for a binary planet, if it was still a planet. it might have been true for the primordial earth where the moon was a lot closer.

      1. What an object “orbits” is actually just left vague by that definition. Using barycenter to define what an object orbits sucks because it doesn’t really make physical sense, since the planet’s orbits aren’t circular.

        It’s pretty much “the Moon orbits the Earth because we say it does.”

    3. Jupiter is not a planet under the new classification system… Its barycenter with the sun is outside the radius of the mean “surface” of the sun…. and it does not clear its orbit; in fact it does just the opposite as evidenced by the Trojans at two Lagrange points. I would concede a classification of “failed binary” if not for Jupiter not being close to even a Brown Dwarf in mass.

      1. “Jupiter is not a planet under the new classification system…”

        Yes, it is. The classification is intentionally vague (neither ‘orbit’ or ‘clear your neighborhood’ are strictly defined) because if you construct a classifier for the Solar System the ‘gaps’ where you could put cuts for those terms is utterly gigantic: there’s like, an order of magnitude variation where you could define ‘clear your neighborhood’ and it wouldn’t matter.

        They left it vague so that additional exoplanet studies could clarify it. There are “exact” calculated implementations

        “in fact it does just the opposite as evidenced by the Trojans at two Lagrange points”

        You’re misinterpreting that phrase, which is a common criticism of it: ‘dynamical dominance’ was also suggested, and in fact “dynamical dominance” is in the figure of the Stern & Levison paper that it comes from. Basically, ‘clear your neighborhood’ means that a random small object at that orbital distance (so not satellites or trapped bodies like Trojans) is likely to be scattered out of that area by said planet within a Hubble time.

  3. Remember the good old days when an object was a star, maybe also a planet, and maybe also a satellite and that was as complicated as our naming got? I don’t, but I miss it anyway.

      1. There used to be a cute, old German term for galaxy, by the way: Welteninsel.
        It literally translates to “worlds Isle” or “isle of worlds”.
        As if it was, say, a colorful and tropical island but up in the skies, in the cosmic ocean.

        Another of such word was “Weltenraum”, the plural of “Weltraum” (world room/world space; space).
        It’s special insofar, because the word was referring to other inhabited worlds in the vast cosmos. Both words had been in use in early-mid 20th century, still.

          1. Glad to hear. Your welcome. ^^

            I just thought it would be worth mentioning.

            In the past, English language had used a lot of German-like words.
            Overmorrow for “Übermorgen”, for example.
            This was before French had taken the lead in inspiring English language, I believe.

            There are other words that would fit nicely into English, still.
            Like “rainshade” (Regenschirm) for an umbrella.
            It would be a matching counterpart word for “sunshade” (aka parasol).

          2. It’s a Germanic language, so that makes sense. It also borrows heavily from the Romantic languages due to the influence of many foreign occupations. English, like my ancestral heritage, is kind of a mutt with parts from all over the place.

  4. naming systems and mythical characters are arbitrary
    constructs of the human mind,so jupiter,the easter bunny, hundred dollar bills or pluto,have as much meaning for
    as long as any group can keep the hype going

  5. I will always be a 9 planet person, out of pure contrariness if nothing else.

    Anyway, there was a short story or a “what if” article from a magazine about building a swarm of simple self replicating robots to collect all the mass in the asteroid belt between Mars and Jupiter onto Ceres to make it more round and balance it out. At the end of the day it would be around 3% of the Moon’s mass.

    No idea how’d they’d dock/merge Vesta, Pallas, and Hygiea with Ceres and each other. It’d be an interesting math problem to say the least.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.