In the final analysis, when we send a probe to study a planet, we are not studying its orbit. We care about its orbit mainly in the sense that it allows us to navigate to the planet through space When we get there, we are studying the planets intrinsic physical characteristics, its geology, its atmosphere, its surface features, its interior, its magnetic field, its composition, its morphology, its evolution. Those are the things we care about.
Those are the things that make it a planet, and make it interesting.
Quote from: llanitedave on 03/26/2017 03:48 amThat the IAU insisted on making it about Pluto is what led them into this error in the first place.I doubt many people think it was the IAU that has made it "Pluto specific"...
That the IAU insisted on making it about Pluto is what led them into this error in the first place.
I was clearly referring to the IAU definition of a "dwarf planet". Vesta isn't a "dwarf planet" because it is no longer capable of being in hydrostatic equilibrium. (Ditto Pallas). However, Vesta is differentiated. Therefore the only justification you came up with for a geophys definition fails. Lack of "roundness" doesn't tell you that a body hasn't differentiated.
Actually Eris is an SDO, not a KBO.
As with any continuous sequence (mass in this case) as you approach the boundary (mass necessary to be spherical) you run into borderline examples.
Quote from: DRussell on 03/27/2017 01:59 pmAs with any continuous sequence (mass in this case) as you approach the boundary (mass necessary to be spherical) you run into borderline examples.My point was that there aren't boundaries in a continuous sequence.
If "moon" is just a subclass of all substellar objects in the solar system, then so are "planets", by any definition. And no more or less valid than any other subclass, such as SDO or KBO or main-belt object. llanitedave was trying to suggest that we never classify things after their location or extrinsic properties, when it's actually very common in the solar system. And "moon" is the perfect example. A whole classification based solely on a single extrinsic property, a substellar object in the solar system being in orbit around some other substellar object, regardless of whether that "some other" is Jupiter or a small asteroid.
It wasn't about Pluto at all; it was about Eris - was Eris a planet or not? The IAU would probably have preferred to stay out of it, and let the argument rage on without conclusion. Unfortunately, because of decisions made decades earlier, the IAU has two different systems for naming planets and other bodies. It had to choose one or the other.
IMO -- and I'm not the only person to express this opinion -- we have a perfectly fine label for any round sub-stellar body: world. I'd be for world to be the label for any sub-stellar body that's round. There would then be three classes of worlds: planets (orbit stars), moons (orbit planets), and free-floating (orbit neither stars nor planets [but likely orbit much larger constructs, like star clusters or a galaxy as a whole]). Less massive objects, never round, would still be called asteroids.
Quote from: Paul451 on 03/27/2017 02:34 pmQuote from: DRussell on 03/27/2017 01:59 pmAs with any continuous sequence (mass in this case) as you approach the boundary (mass necessary to be spherical) you run into borderline examples.My point was that there aren't boundaries in a continuous sequence.I'm talking about the lower mass boundary for the geophysical planet definition.
The lowest mass "round" body in the Solar System is Mimas and its mass serves well enough as the lower mass limit for "planet"
Instead of a distracting focus on why Pluto is no longer a planet, the focus for kids could have been an awe factor as to how many new planets there are.
The focus of instruction becomes understanding the structure and types of bodies within the Solar System: 1. What is significant about planets being round?2. What types of orbits do planets have? Dynamically dominant (Principal planets), dynamically not dominant (Belt planets), planets that orbit a bigger planet (moons), planets that don't orbit any star (rogue).[...]And the concerns of the dynamical perspective are addressed - not ignored. In fact, calling the IAU planets "Principal planets" would make pretty big impression on students in school - "You know how the Principal is the person in charge of the school? Well, the Principal planets - like the Earth and Jupiter - are the big planets that dominate the orbits in the Solar System."
Quote If "moon" is just a subclass of all substellar objects in the solar system, then so are "planets", by any definition. And no more or less valid than any other subclass, such as SDO or KBO or main-belt object. llanitedave was trying to suggest that we never classify things after their location or extrinsic properties, when it's actually very common in the solar system. And "moon" is the perfect example. A whole classification based solely on a single extrinsic property, a substellar object in the solar system being in orbit around some other substellar object, regardless of whether that "some other" is Jupiter or a small asteroid. And in bold is not what illanitedave is suggesting.
I can think of no other scientific definition of any class of objects where intrinsic characteristics of the objects themselves are completely absent.
We are suggesting that [...] at the top of the sub-stellar taxonomy "planet" should be the set of sub-stellar bodies that are round no matter their dynamical circumstances ... or composition.
There is no "mass boundary". Any intrinsic property depends not just on mass, but on the composition, age, closeness to the sun of the object, and varies wildly between objects of the same mass.
You can draw arbitrary lines on a continuum, just for the sake of nomenclature, if no other arrangement presents itself. But if the universe presents you with an existing wide separation of classes of objects, why reject it? Why choose a greatly inferior classification system?
Serves what? What scientific need is served by it?
The people who fussed over the "demotion" of Pluto are not the sort of people who would have explained dynamical dominance, or hydrostatic equilibrium, or anything else. Those who didn't understand the concepts made no attempt to understand the IAU definition, and those who did understand it, not only made no attempt to explain the science behind it, but went out of their way to encourage public misunderstanding.Your "won't somebody think of the children" plea assumes the people who opposed the IAU definition are acting honestly and with the broader interests of science at heart. But I see nothing to suggest that.
How does "moon" not meet llanitedave's requirement of a classification based purely on extrinsic/location/environmental properties?"Moons" are implicitly a sub-category of the broader class of all substellar objects, but explicitly so are IAU-Planets. Both are dynamical subcategories of the broader intrinsic class (substellar masses).Just as KBOs are a dynamical sub-category of the broader class of substellar non-planetary objects.
Why?IMO, extrinsic classification of even sub-planetary objects is more scientifically useful. KBO, SDO, main-belt asteroid, Trojan, etc.
What is the point in placing an arbitrary intrinsic property first? You separate objects of extrinsic similarity. For what reason? What is the benefit of grouping Ceres with Pluto, Pluto with Eris, instead of grouping them first with their neighbours and siblings. What scientific usefulness is served by such a grouping? What science has been harmed by the previous lack of such "roundness" classification?
QuoteIMO, extrinsic classification of even sub-planetary objects is more scientifically useful. KBO, SDO, main-belt asteroid, Trojan, etc. Why?
IMO, extrinsic classification of even sub-planetary objects is more scientifically useful. KBO, SDO, main-belt asteroid, Trojan, etc.
Quote What is the point in placing an arbitrary intrinsic property first? You separate objects of extrinsic similarity. For what reason? What is the benefit of grouping Ceres with Pluto, Pluto with Eris, instead of grouping them first with their neighbours and siblings. What scientific usefulness is served by such a grouping? What science has been harmed by the previous lack of such "roundness" classification? These questions can all be turned right back at you about the system that you prefer. Why exclude all spherical bodies from the class "planet"?
The problem with classifying due to orbital neighbors is that it is open to too much opinion and leeway and isn't in any way a scientific way of looking at the matter. Someone jokingly pointed out that Neptune isn't a planet according to the definition in use at the moment as it hasn't cleared Pluto out of it's orbit. This was probably suggested in jest, but it does show how poor the definition is. Jupiter could be demoted too as it hasn't cleared it's orbit of it's forward and aft trailing trojan debris. How much 'debris' is acceptable? How do you choose and where do you draw the line?What happens if we find a 'Plutoid' out in the Oort cloud that is larger than Mercury and is perfectly round? What do we do if we find out that some of the planets that we are seeing around other stars are sat inside a debris belt? Using a different field of study, we can definitely tell if a particular creature is, for example, a dog. The dog has certain points that differentiate it from other creatures. These points are easily determined and allow those who have never seen a particular specimen to determine if the creature they are looking at is, or isn't, a dog. The dog is still a dog, no matter where it is. If it is in isolation, or if it is surrounded by sheep or cats, it is still a dog. It doesn't change it's taxonomy depending on it's location or surroundings.We need a similar taxonomy system for planets, based on their physical attributes.
The problem with classifying due to orbital neighbors is that it is open to too much opinion and leeway and isn't in any way a scientific way of looking at the matter.
Someone jokingly pointed out that Neptune isn't a planet according to the definition in use at the moment as it hasn't cleared Pluto out of it's orbit. This was probably suggested in jest, but it does show how poor the definition is.
Jean-Luc Margot's paper. And the earlier Stern/Levison paper. Both give you a good understanding of the scale of the gap between planets and non-planets, even if you only skim the maths.Using the handy wikipedia tables, Stern/Levison score and rank the planets thus:NameRankValue - RankValueJupiter11,000,000,000 -140,000Saturn2 50,000,000 -2 6,000Uranus3 400,000 -5 400Neptune4 300,000 -6 300Venus5 160,000 -3 950Earth6 150,000 -4 800Mercury7 2,000 -7 130Mars8 1,000 -8 50IAU Planet - 1 -- 1Pluto9 0.003 -10 0.03Eris10 0.002 -11 0.02Ceres11 0.001 -9 0.04Haumea12 0.0002 -12 0.008Makemake13 0.0002 -13 0.007 Stern/Levison, 2002 Margot, 2015Note that even using different methods still results in the same clear distinction between planets (>1) and non-planets (<1), with nothing near the margin.
I keep hearing that the dynamical definition is so useful for exoplanetary systems.But clearing the orbit takes time, so in the early solar system Earth wasn't a planet.
it makes very much sense for comparative planetology to have a geophysical definition. Titan is comparable to Earth in many aspects, but how much does it have in common with Saturn's satellite Pan?
Quote from: tea monster on 03/30/2017 05:50 pmThe problem with classifying due to orbital neighbors is that it is open to too much opinion and leeway and isn't in any way a scientific way of looking at the matter.No there isn't. Read back through the thread, the various measures that have been proposed for dynamical dominance show a gap between planets and non-planets of orders of magnitude. There's no ambiguity, opinion, or leeway, it's based purely on the reality of the solar system. (See below)Quote from: tea monster on 03/30/2017 05:50 pmSomeone jokingly pointed out that Neptune isn't a planet according to the definition in use at the moment as it hasn't cleared Pluto out of it's orbit. This was probably suggested in jest, but it does show how poor the definition is.If that someone was Alan Stern, then you should be aware that he co-wrote one of the early papers that defined the difference between the eight planets and the non-planets, in a paper called "Criteria for Planethood". He even used the language "clearing the neighbourhood" to describe the difference. So he when he now publicly "misunderstands" the IAU definition, try to image my contempt for him.Quote from: Paul451 on 02/21/2017 07:44 amJean-Luc Margot's paper. And the earlier Stern/Levison paper. Both give you a good understanding of the scale of the gap between planets and non-planets, even if you only skim the maths.Using the handy wikipedia tables, Stern/Levison score and rank the planets thus:NameRankValue - RankValueJupiter11,000,000,000 -140,000Saturn2 50,000,000 -2 6,000Uranus3 400,000 -5 400Neptune4 300,000 -6 300Venus5 160,000 -3 950Earth6 150,000 -4 800Mercury7 2,000 -7 130Mars8 1,000 -8 50IAU Planet - 1 -- 1Pluto9 0.003 -10 0.03Eris10 0.002 -11 0.02Ceres11 0.001 -9 0.04Haumea12 0.0002 -12 0.008Makemake13 0.0002 -13 0.007 Stern/Levison, 2002 Margot, 2015Note that even using different methods still results in the same clear distinction between planets (>1) and non-planets (<1), with nothing near the margin.
Stern is clearly trolling, and IMO is doing so since he realized he can "energize" support for his programs. There is no other reasonable explanation to him misunderstanding this.
Quote from: DRussell on 03/30/2017 10:22 amQuoteIMO, extrinsic classification of even sub-planetary objects is more scientifically useful. KBO, SDO, main-belt asteroid, Trojan, etc. Why?Because it's how they are actually grouped by researchers in order to do comparative science.Even some of the intrinsic properties that are used in informal grouping, such as composition (icy/rocky), are often considered more interesting because of what it says about the extrinsic nature of such objects, such as their location of formation and what that says about dynamic systems. And vice-versa, there's a strong overlap between intrinsic and extrinsic properties. Being a KBO or a main belt asteroid is as much intrinsic as not.Extrinsic/location/environmental classification as a top-level-sort doesn't ignore intrinsic properties, it better enables their analysis. Whereas having intrinsic properties as the top-level classification requires research on the individual objects in order to classify them, which will generally require a prior division into extrinsic groups in order to allow enough analysis to then put them in intrinsic categories.For example, your desire to use mass limits as a divider: If an object lacks a visible moon, then in order to estimate its mass, you need to know its size and composition, which usually means understanding where it formed, which means grouping it with objects that formed in similar areas. You need to do an extrinsic grouping in order to even estimate your intrinsic classification. So why not have the extrinsic grouping at a higher level, since that's what researchers will actually do anyway?It's the same for any measure you try to pick for sphericalness. Use hydrostatic equilibrium (as the IAU did for the nonsensical "dwarf" category) and you exclude objects that share major geophysical properties but fall on different sides of the line. Your own scheme is worse. There's no reason (neither dynamical nor geophysical) for objects just above and just below that mass limit to be grouped separately. It is not a natural grouping. Actual research will inevitably have to create other primary groupings that ignore your top-level classification anyway. Your system doesn't actually add anything useful.I can see internal-differentiation being a useful divider. Trying to create an actual "line" is still too arbitrary to turn into a major classification system, but at least it is scientifically interesting property of the objects being studied. However, as I said earlier, because it depends so much on composition, historical-temperature, collision history, etc, it is a fairly unique trait of each object. Once you've been able to classify an object's degree of differentiation enough to put it in one of two groups, such crude super-categories are no longer useful to the study of that object and those like it.OTOH, extrinsic groups remain useful, KBO/main-belt/trojan/moon's-of-Saturn/etc, even if the intrinsic properties of some members of the group are understood in greater detail. Indeed because some members of the group are understood.Quote from: DRussell on 03/30/2017 10:22 amQuote What is the point in placing an arbitrary intrinsic property first? You separate objects of extrinsic similarity. For what reason? What is the benefit of grouping Ceres with Pluto, Pluto with Eris, instead of grouping them first with their neighbours and siblings. What scientific usefulness is served by such a grouping? What science has been harmed by the previous lack of such "roundness" classification? These questions can all be turned right back at you about the system that you prefer. Why exclude all spherical bodies from the class "planet"?Because they really fall into different groups. It's a thing which actually exists. A large separation between object-types that was found in nature, not an arbitrary line drawn in a continuum. The group inside the category plays a fundamentally different role in the history and structure of the entire solar system than the group outside the category.An example: An apparent systematic bias in the orbits of TNOs has led some researchers to speculate that there is another planet somewhere out there.If it exists, it will be considered a planet because it does that. Hence it belongs to a fundamentally different class of objects than the smaller TNOs like Eris, Pluto and Sedna, because they don't do that. Just as Plutinos are slave to Neptune's orbit, not the other way around. The difference is fundamental.