Planetary Migration and the Lunar Cataclysm
Weather cancellations left us without much time to talk about Saturn this semester, but we did touch on its trademark ring system. Why did some of the solid stuff that condensed near Saturn form moons, and some form ring particles? Because close to Saturn (closer than the so-called "Roche limit" for all you jargon-lovers out there) the tidal distortion induced by Saturn is too great for a moon to be held together by its own gravity. More precisely, the small bits of solid stuff within the Roche limit were never able to stick together (coalesce) to form moons in the first place, while further from Saturn this wasn't a problem. The same thing goes for Jupiter, Uranus, and Neptune.
Saturn's rings are so bright because they reflect visible sunlight very well, and this in turn is because the individual ring particles are made of ice and (perhaps) ice-coated rock. Spacecraft missions have revealed ringlets, spokes, and a braided ring, leading to theoretical explanations involving (for example) density waves and shepherd moons. A few really faint rings have also turned up. But we already knew about the A, B, and C rings, plus the major gaps such as the Cassini division.
Comets are the most primitive material in the solar system, being cold chunks of ice mixed with rocky material. These "dirty snowballs" are comet nuclei: faint, nondescript leftovers from the formation of the solar system. But their highly elongated orbits occasionally bring them close to the Sun, at which point the surface ice sublimates and produces the lovely glowing head and tails that everyone thinks of when comets are mentioned -- and, for some comets, meteors ("shooting stars") as a side effect. (A meteor is simply a bit of rocky comet "dirt," released from the comet when the ice surrounding it sublimated, that gets in the way of Earth and heats up and glows due to atmospheric drag.) You should know something about a comet's structure (nucleus, coma, tails).
Each time a comet passes close to the Sun, it loses some material, so it only takes a few hundred orbits before there's no more surface ice left to sublimate. Hence the comets we see today can't possibly be the same comets that were around 4.5 billion years ago: There must be at least one "reservoir" of icy leftover planetesimals that constantly replenish our supply of comets.
Actually there are at least two such reservoirs. Short-period comets like Halley, which take less than 200 years to complete a trip about the Sun, have orbits that are roughly coplanar with the planets' orbits. These objects are thought to originate in the Kuiper Belt, the outer disk of our solar system beyond Neptune. More than 1000 Kuiper Belt Objects (KBOs) have been discovered since 1992, including one, Eris, that's more massive than Pluto. This discovery led to a year of debate on what constitutes a planet, culminating in the decision that Eris, Pluto, and Ceres (the largest asteroid) are "dwarf planets" rather than planets.
(Yes, I know: grammatically speaking, all dwarf planets must be planets, just as all foreign cars must be cars and all carnivorous mammals must be mammals. The International Astronomical Union ignored grammar and defined "planet" and "dwarf planet" to be two mutually exclusive categories. Sigh.)
Remember, when Pluto was announced as the ninth planet in 1930, it was thought to be about as massive as Earth, whereas we now know that it only has about 1/6 the mass of our Moon. It also used to be thought that the presence of Pluto's moon -- Charon, discovered in 1978 -- was evidence that Pluto is a planet, but today we know that many asteroids and KBOs have satellites.
What are the new, official definitions of the terms "planet" and "dwarf planet"?
When an icy body -- a leftover planetesimal -- from the Kuiper Belt is gravitationally pulled into an elongated orbit that passes into the inner solar system, a short-period comet is born. The several dozen known Centaur objects are thought to be in the middle of this transformation, as their orbits cross those of the outer planets.
Long-period comets like Hyakutake and Hale-Bopp typically have highly tilted orbits, so they must originate in a different reservoir: a spherical "cloud" of icy bodies that surrounds the Sun and planets, rather than a flattened belt. This is the Oort Cloud, whose trillion or so citizens are so distant that none of them has ever been observed.
What are some results of the Rosetta mission to comet Churyumov-Gerasimenko?
We now know of over 3900 planets around other stars -- including as many as nine planets orbiting the same star! Until 2010 or so the great majority of extrasolar planets (or "exoplanets") were discovered via the Doppler method. The idea is that you look for tiny, changing Doppler shifts in the absorption lines present in the star's spectrum: first blueshift, then redshift, then blue, then red, then.... This implies that the star is moving slightly back and forth as a planet orbits it and pulls on it. Just as the Sun pulls on Jupiter, Jupiter pulls equally hard on the Sun; but because the Sun is so large it doesn't respond much to Jupiter's pull. So it is for other stars with planets.
(NOTE: Just because a method involves the Doppler effect doesn't mean that it's radar! The Doppler technique for finding extrasolar planets has nothing whatsoever to do with radar.)
We discussed at length precisely how we can use these data -- Doppler shift as a function of time -- to figure out the semimajor axis and the eccentricity (degree of flattening) of the planet's orbit; we also get a lower bound (why?) on the planet's mass. In some cases the data are too complex to be explained by a single planet, so we instead infer a planetary system with two or more members.
The Doppler method is biased towards massive planets orbiting close to their parent stars, because those are the planets that pull hardest on the parent stars and hence produce large, easily measured Doppler shifts. And indeed we've found a fair number of "hot Jupiters." Many of them are so absurdly close, however, that we've had to rethink our ideas on planetary system formation, which held that giant gas planets would have to form far enough away from the parent star that ices could solidify. Ice would add to the rock and metal to give the forming planet's solid core enough mass (and hence gravitational pull) to gather in a massive Jupiter-style atmosphere. This theoretical idea still makes sense to us, so we suspect that the hot Jupiters do indeed form at Jupiter-like distances and then "migrate" inward due to gravitational interactions with other planets, spiral density waves in the gas disk surrounding the newborn star, whatever.
The prospects for life evolving and thriving on an Earthlike planet would appear to be dim in any planetary system in which a giant planet comes migrating through the neighborhood: The gravitational influence will make it hard for the Earthlike planet to maintain a stable orbit at a reasonable distance from the star. The massive planets found to be in highly elliptical orbits would have equally ugly influences on terrestrial planets, threatening to make a close approach and gravitationally fling them into the deep freeze of interplanetary space.
We have used the Doppler method to find extrasolar planets that are "only" the mass of Uranus and Neptune, but for planets much smaller than that it becomes difficult to make reliable measurements of the small Doppler shifts produced by small planets tugging on their parent star. The exception would be an Earth-mass planet that's extremely close to its parent star; but of course the scalding-hot climate would rule out life as we know it -- supposing, that is, that the star is similar to our Sun. Thus we need other techniques to detect Earth-like planets in Earth-like orbits around Sun-like stars.
For example, in the early part of this decade the highly successful Kepler mission detected many exoplanets via the "transit" method -- so how does that method work? What extra information does it give us that we wouldn't get from Doppler data alone? What can we learn if we can apply both methods to the same exoplanet? Why does the transit method, useful though it is, fail to detect most exoplanets? Can you say a few words on what "gravitational microlensing" is and what we've found using this method? Finally, what makes direct imaging of exoplanets so difficult?
It may be no accident that we're discussing all of this on a planet orbiting a star like our Sun, because other kinds of stars may not be able to provide a healthy environment for life to evolve. Very massive stars are extremely bright and hot and thus produce a lot of UV light, which is lethal to life; even worse, they only shine for a relatively brief time before running out of nuclear fuel to burn, so there's probably no time for complex life like us to evolve on their planets. Very small stars (the majority of stars) don't have these problems, but these "red dwarfs" are faint and emit mostly low-energy (red and IR) photons, so a planet would need to be orbiting very close to such a star for water to be liquid and life to get the energy it needs. And at such small distances tides would torque the planet into a state of synchronous rotation, with one side too hot for life because it always faces the star, the other side too cold because it always faces away. Still, as you saw in a video, for these "tidally locked" planets perhaps global wind circulation could moderate the climate and prevent the atmosphere from turning into ice on the cold side, in which case life might be able to exist in the region of perpetual twilight. (Incidentally, yet another possibility is that life could evolve on satellites of planets: "exomoons.")
You're not, of course, responsible for memorizing everything about every exoplanet we've read about or discussed in class, but you should know something about the most prominent examples -- most notably, Proxima Centauri B and the TRAPPIST-1 system. As yet another example, what's a "superearth" such as Kepler 22b like?
OK, so some day, perhaps a decade from now, we find an Earthlike planet orbiting at an Earthlike distance from a Sunlike star: What do we look for to check if there's life on that planet? A promising idea (albeit very, very difficult in practice!) is to use spectroscopy: take the faint starlight that's reflecting off the planet -- or, for a transiting planet, the starlight that's passing through the planet's tenuous outer atmosphere -- and spread it into its spectrum. Then look for the absorption bands produced by particular gases in that planet's atmosphere. Two gases that exist in Earth's atmosphere because of living organisms are oxygen, O2 (and its cousin ozone, O3), and methane, CH4. Now, if you find absorption bands due to just one of these gases, perhaps you could come up with some other way of producing them that doesn't involve life -- as may be the case, for example, for the methane that's been detected in Mars' atmosphere. But finding both oxygen and methane in the same atmosphere would be a sure sign that they're being pumped into that atmosphere on a daily basis. Why? Because oxygen and methane rapidly react with each other to produce carbon dioxide and water. Why then, if they're so good at destroying each other, are both oxygen and methane present in Earth's atmosphere? Because life keeps producing them to replenish the supply.
(The practical difficulty of this technique is due to the extreme faintness of the planet relative to its parent star: the "firefly in the spotlight glare" effect. Yet here we're proposing not merely to see the firefly but to analyze its light in detail! Thus for non-transiting exoplanets -- which, recall, means most exoplanets -- the technique relies on finding some advanced method of blocking the star's light without blocking the light of the nearby planet, such as a starshade.)
The Galileo spacecraft actually tried this out in the early 1990s when it passed close to Earth (to get a gravity assist) on its way to Jupiter: it carried out spectroscopy on the sunlight reflecting from Earth in order to verify that there's life on Earth. In case you were wondering.
Now that we have the concept of planetary migration under our belts, let's note that a version of this process operated in the early days of our own solar system. Gravitational interactions between the four large outer planets and the many icy/rocky planetesimals in the outer solar system caused Saturn, Uranus, and Neptune to migrate outward -- for example, this is when Pluto and many other "plutino" KBOs got caught into a 3:2 orbit-orbit resonance with Neptune -- and Jupiter to migrate inward. Not drastically inward, as with the "hot Jupiter" exoplanets, but inward nonetheless. Jupiter behaved differently from the other three gas giants because, being so large and having such a strong gravitational field, it was able to fling ("scatter") nearby planetesimals far outward from the Sun; this may well be how the Oort Cloud formed.
But Jupiter would also have flung some planetesimals inward to our neck of the woods. Many of these objects would have plunged directly into the Sun, but some might have stuck around for a few million years in elongated orbits that crossed our own orbit: near-Earth objects. And this would have led to many, many large impacts on our planet.
Interesting enough. We don't have a video record of what went on back then, but in recent years extensive computer simulations have been run to demonstrate that these ideas are plausible. What one group's simulations indicate is that the migration wouldn't have occurred at a steady rate: instead, much of the migration and planetesimal scattering would have occurred during a relatively brief spurt of time. This spurt would have been triggered by Jupiter and Saturn reaching a 2:1 orbit-orbit resonance, at which point they would have strongly influenced each other's orbits, which in turn would strongly influence Uranus and Neptune, which in turn would influence the belt of planetesimals beyond Neptune.
What if that time were 600 My after the formation of the solar system, that is, 3.9 Gy ago? That could explain the Lunar Cataclysm that we discussed earlier in the semester. Much more work needs to be done to flesh out this hypothesis -- for example, it may be that it happened earlier or later than 600 My, in which case it doesn't work as an explanation -- but it's a promising start.