The case for planet Pluto

The case for planet Pluto
(under construction)
by Wm. Robert Johnston
last updated 17 October 2006

Seven years after reaffirming Pluto is a planet, the International Astronomical Union (IAU) reversed itself in September 2006 and declared that Pluto is not a planet. Publishers and educators have obediently begun changing their products and curriculum. Some planetary astronomers have harshly criticized the IAU's new definition of a planet, and they have a good case. I have to weigh in with those who would count Pluto and comparable objects as planets.

The public is understandably confused. They have learned that Pluto is a planet for 76 years, and the background of the IAU decision involves a range of lesser known details:

For nearly half a century from its discovery in 1930, Pluto was thought to be as large as Mercury, maybe as large as the Earth. There was then a clear, convenient demarcation between the nine planets and the smaller solar system bodies like asteroids. Only in the period 1978-1986 did astronomers finally determine Pluto's diameter, only half that of Mercury and 2.5 times that of the largest asteroid.
In 1992 astronomers began discovering icy asteroid-like bodies in the region of the solar system past Neptune. Pluto was recognized as the largest of this population of trans-Neptunian objects (TNOs), with 1,127 known so far. Despite a movement by some astronomers to revoke Pluto's designation as a planet, the IAU affirmed in 1999 that Pluto was counted as a planet. In contrast, the new trans-Neptunian objects, all smaller than Pluto so far, were classified as minor planets like asteroids.
In the 1989 astronomers began discovering planets orbiting stars other than the Sun. So far 210 extrasolar planets have been discovered. Although the observation methods are biased towards finding larger planets, we already find a variety not seen in our own solar system.
In 2005 astronomers announced discovery of three large trans-Neptunian objects. The largest, 2003 UB313 (now named Eris), is larger than Pluto. This finally forced the question of distinguishing between small planets and large TNOs.

Any definition of a planet must be recognized as subjective to some degree: distinguishing between small planets and large minor bodies is to some extent arbitrary. Astronomers have not had to deal with this continuum before, but now they must. Further, any planet definition is in some sense just a name. Pluto is the same physical object regardless of which category we put it in.

Despite the limitations on a planet definition, it is still desirable. Such groupings assist us in looking at general characteristics of objects. Astronomers deal with large numbers of objects: currently, astronomers have assigned designations to hundreds of moons, thousands of comets, hundreds of thousands of asteroids, and millions of stars. Astronomers long ago agreed to give the IAU certain roles in regard to nomenclature, such as overseeing the naming of small solar system objects.

Since a planet definition should be useful to the astronomical community, this suggests some tests of a good definition:

A clear scientific basis: we should strive for logical scientific criteria to distinguish planets from non-planets, so as to have a term useful to scientists.
Generalizable to other planetary systems: we are already learning about planets beyond our solar system, including planets very different from those we are familiar with. A planet definition should be general enough to apply, in a meaningful way, to these objects.
Easy to apply: it would be helpful if the criteria do not depend on characteristics that may not be known for years or decades after an object is discovered.

With this in mind, we consider the new IAU definition: "A 'planet' is a celestial body that (a) is in orbit around the Sun, (b) has sufficient mass for its self-gravity to overcome rigid body forces so that it assumes a hydrostatic equilibrium (nearly round) shape, and (c) has cleared the neighbourhood around its orbit." Part (c) is the criterion which disqualified Pluto. This criterion may also fail all of the suggested tests for a good definition:

It is scientifically ambiguous. Pluto along with hundreds of other known TNOs cross the orbit of Neptune, so Neptune fails this definition. Thousands of known asteroids cross the Earth's orbit, so the Earth is apparently not a planet either. Jupiter even shares its orbit with thousands of known Trojan asteroids.
It may not be applicable in other planetary systems. Many of the Jupiter-sized planets known are in highly eccentric orbits. Should their neighborhoods prove to be "uncleared", disqualification of objects larger than Jupiter will erode the significance of the definition. (Such extrasolar planets are a profound objection to one proposal, that planents be in near-circular orbits.)
It is very difficult to apply, since it requires knowledge about smaller bodies in the neighborhood of the object in question. It took 62 years to begin identifying objects in the neighborhood of Pluto's orbit. Suppose Pluto was missed in those photographs taken in 1930, but that 2003 UB313 Eris was not missed in the photographs taken of it in 1954: Eris, in an orbit well beyond Neptune's, would have apparently qualified as a planet. For extrasolar planets as well, such knowledge may be generations away.

A criterion of similar spirit but more scientifically robust than the IAU "cleared neighourhood" criterion is that a planet be the dominant object gravitationally in its neighborhood. This would clarifies why Neptune is a planet and Pluto is not: Pluto's orbit is a consequence of resonances driven by perturbations from Neptune. But in some sense the inner planets are dominated by Jupiter; the Earth's "clear" neighborhood is partly due to Jupiter. This definition remains difficult to apply in extrasolar planetary systems, dependent on observations decades away.

The definition adopted by the IAU was hastily offered at the end of the 2006 conference (after most attendees had departed) to replace the controversial definition produced by an IAU-designated committee after lengthy deliberation. In general, this definition used only parts (a) and (b) of the adopted defintion. By counting any Sun-orbiting body nearly in hydrostatic equilibrium, it would tend to include asteroids and TNOs down to diameters of 300-800 km. This would likely have included 2-4 asteroids and 10-40 known TNOs, even apart from yet-to-be-discovered TNOs. The "inclusive" nature of the original proposal was its downfall. The significance is clearly that it did not capture the essence of what astronomers have in mind with the term "planet".

There are significant differences between rocky or icy bodies with sufficient self-gravity to be round, and the larger objects traditionally counted as planets--perhaps including Pluto. Basri and Brown recently reviewed several such differences. Around a size of 3000 km, gravitational energy is sufficient to significantly modify the object's internal chemistry. A bit larger and solid state convection becomes important. Internal pressures significantly compress the internal material of a rocky body larger than 6000 km diameter, or an icy body larger than 1000 km diameter. Gravitational differentiation into a denser core and less dense mantle/crust occurs for objects as small as Pluto, perhaps as small as 400 km. Ice-rock bodies like Pluto and even smaller may have subsurface oceans, according to some models.

Pluto is the best studied of objects that some would lump together as less than planets. These studies show a variety of phenomena more associated with planets than with small solar system bodies. Pluto has an atmosphere, apparently particular to when Pluto is near perihelion, but at which time it drives surface changes: frost deposits form and evaporate over large regions. When Pluto is finally examined close up in 2014, astronomers would not be surprised to find similarities to Neptune's moon Triton, which shows geologic activity including active geysers or cryovolcanoes.

The manner in which the new IAU definition has been implemented is arguably clumsy. Following opposition to the committee proposal, the IAU instead hastily adopted a less than fully robust definition that excluded Pluto. Within weeks, after 76 years as a planet, Pluto was relegated to minor planet number 134,340, sandwiched in the MPC list between two main belt asteroids each smaller than New York's Central Park. The newly created "dwarf planet" category, explicitly not a planet despite the grammatically construction, in practice overlaps with minor planets; since the distinction between dwarf planets and small solar system bodies is difficult to apply and not particularly useful at this time, it is likely this term will not make it into the astronomical lexicon.

This comes to a defect in the IAU's adoption of a planet definition: it is a term, not a name, and was imposed by a minority of the scientific community. The scientific community has established usage of the term "planet", and in fact many in the planetary science community have signed a petition rejection the new IAU definition and promised to produce a better one.

I've been asked what definition I would offer, given my issues with the IAU definition. I concede that no definition is perfect, and that it will likely prove necessary to revisit the issue in the near future. All things considered, I would suggest simply using a size threshold to discriminate between planets and smaller solar system bodies. A diameter cutoff of 2000 km would count Pluto and Eris as planets; a cutoff of 1,300 km would include three more TNOs that already appear distinct from the other 1,122 smaller ones (there is in fact a fundamental argument in favor of this smaller threshold). This threshold is approximately the size range where some of the structural and surface pheonomena we identify with planets start to come into play.

A threshold-based definition has the disadvantage of not having as clear a scientific premise as one based on hydrodynamic equilibruim or cleared orbital neighborhood. It is significantly more straightforward to apply, comes closer to the spirit of what astronomers are trying to capture with the term planet, and is generalizable to other planetary systems. It is also somewhat intended as a stop-gap measure: we are attempting to impose a definition on objects we have yet to examine as more than points of light. The adoption of a definition excluding Pluto is ironic given that a NASA mission is en route to Pluto to provide our first closeup look in 2014; this look might have a bearing on how we think about this object, and we may want to keep an open mind.

Related information on this web site: