Analog SFF, January-February 2009 Read online

  At only 12.6 parsecs away, HD69830 is very nearby by galactic standards. It is in the constellation Puppis, so it is very roughly in the direction of Sirius, although about four times as distant. It is also a little farther away from the galactic center than Sirius and is farther galactic north, a direction perpendicular to the plane of the galaxy; the closest bright star to the galactic north pole is Denebola (Beta Leonis), although Arcturus is fairly close also. The stars nearest to HD69830 are typical of those in the solar neighborhood, in that they are the usual collection of the bizarre and the unbelievable. Within about five parsecs of HD69830, there are the following systems, among others. There is one with two Sun-sized stars separated by a few hundred million kilometers whose planets, if they exist, would have a second sun, sometimes by day, sometimes by night. There is an average-sized white-orange star that noticeably waxes and wanes as its huge sunspots grow and decay. And there is a UW Ursa Majoris type variable star, two suns orbiting each other so close that they are almost touching and so fast that they orbit each other in less than one day.

  So as I said, a typical stellar neighborhood. In addition, about three parsecs away from HD69830 is the very Sunlike star HD76151 (alias GL327, HR3358 etc.). Its status as a solar twin has been officially recognized: according to one study, it is the 14th most Sunlike star in the immediate galactic vicinity. It is probably younger than our Sun, and no planets have yet been found orbiting it. What has been discovered is that it is emitting an excess of infrared radiation, more than would be expected from the typical spectrum of a solar-type star. Stars with extra infrared emission have been noted many times before. One of the first to be discovered was Beta Pictoris, a young, hot star about 17 parsecs away. Extra infrared radiation has also been found in the spectra of similar nearby stars, like Vega. The radiation excess is interpreted as coming from a large amount of rocky or icy debris circling the star and being heated as a result. Theories of solar system evolution suggest that this debris disk represents the early, planet-forming stage of a system, when small rock and ice particles were plentiful at first but gradually disappeared as they collided and coalesced into larger, planet-sized bodies.

  The debris disk of HD76151 is different, however. Because of the long wavelength of the detected infrared excess—and therefore its low temperature—it appears to originate several billion kilometers away from the star. If similar measurements were made of our solar system, this cool radiation would originate from the Kuiper belt, the collection of comets, planetary fragments and general wreckage left over from the formation of the solar system, located outside the orbit of Neptune. But the Kuiper belt of HD76151 appears to be about 100 times as bright as the Sun's. It must be truly massive, consisting of hundreds of objects the size of Pluto and millions of bodies over 100 km in diameter. With so much junk floating around in the outer stellar system of HD76151, some of it would probably find its way into the inner part of the system from time to time, maybe a lot more often than comets wander past the Earth. This might have a serious effect on the habitability of any Earth-like planets in this system. It is well known that if large comets strike a world, they can have enough explosive power to disrupt the evolution of life—or end it.

  Like HD76151, the star HD69830 has an infrared excess in its spectrum, but this time the excess is hot, implying that it originates from dust orbiting relatively close to the star, probably within about one hundred million kilometers or so, or closer to the star than the Earth is to the Sun. In fact, the inner part of our solar system is still permeated by a similar dust disk: it is called the zodiacal cloud and is created by the tails of short-period comets and occasional collisions between asteroids. It can be seen from the surface of the Earth as the zodiacal light, a faint glowing band sometimes observed following the Sun near the horizon after the Sun sets. But the dust disk of HD69830 is about a thousand times as bright as the Sun's zodiacal cloud. To create this amount of dust by collisions, an asteroid belt in the HD69830 system would have to contain more than 10 times the number of large objects that there are in the asteroid belt of our own solar system. Because so many collisions are needed to explain the observed density of hot dust in the HD69830 system, the asteroidal objects would also have to be more tightly packed than in our own asteroid belt.

  Hot dust is rare: a survey of 69 Sunlike stars showed that HD69830 was the only one to have it in measurable amounts. More recent work by Meyer and collaborators has clarified the situation a little. By observing a large sample of over three hundred Sunlike stars, they conclude that most very young stars have hot dust, but that it almost entirely disappears with age. This is completely consistent with the usual understanding of how the solar system evolved: the formation of planets in the inner solar system was accompanied by a decline in the collision rate, and this, combined with the Sun's radiation pushing the smaller dust particles farther out, caused a large decline in the amount of dust close to the Sun.

  So one conclusion that can be drawn from Meyer's work is that because hot dust disks around the youngest Sunlike stars seem fairly common, this means that many Sunlike stars are likely to have planets. This result, although obviously important, is not surprising either to working scientists or to readers of speculative fiction. Meyer's study also suggests that the dust surrounding HD69830 is due to its extreme youth. Nevertheless, the star can't be that young: HD69830 probably formed about two billion years ago and so is too old for a planet-forming, hot, dusty disk to be still hanging around in its inner stellar system. The dust could also be created by collisions in a super-massive asteroid belt, but Wyatt and coworkers recently suggested that this is unlikely. Their numerical simulations showed that the amount of dust produced by such a belt actually decreases as the star gets older. This is because collisions in the belt grind the dust down into particles that are small enough to experience something called Poynting-Robertson drag[1] that causes them to fall into the Sun, thus gradually removing dust from the belt. For a star as old as HD69830, simulations show that the observed amount of dust is too large to be produced by a giant asteroid belt that has been around since the formation of the system. So Wyatt concludes that the dust is temporary: some kind of collisions have occurred relatively recently in the HD69830 system, within the past 100 million years, and have produced the hot dust. What type of collisions? To explain that, we first have to describe the planets of HD69830.

  [FOOTNOTE 1: A result of the absorption of radiation by small particles, causing them to gain mass-energy. To conserve angular momentum, they must fall into a smaller orbit. It affects particles a few micrometers in diameter. For smaller particles, radiation pressure is larger than drag and so they move outwards.]

  * * * *

  The planets

  The three Neptune-sized planets in this system have been designated HD69830b, HD69830c and HD69830d, and have distances from the primary of 0.08, 0.19 and 0.63 AU (11, 28 and 94 million kilometers), respectively. They were discovered by the radial velocity method, where the gravitational attraction of orbiting planets causes slight back-and-forth movements in the star that are detectable in its spectrum because of the Doppler effect, like the change in pitch of a racing car's engine as it flies past. Examination of the radial velocity variations has ruled out the presence of planets of Saturn mass or larger within four AU (600 million kilometers) of the star.

  Strictly speaking, the radial velocity technique only measures the minimum mass, not the actual mass. To explain this, we first note that the radial direction is the direction along the line of sight from the Earth to the star. Let's assume that the orbit of a large planet around a star is seen exactly edge on from the Earth, so that if the planet were observed in a telescope, it would appear to wander from side to side, traveling in front of and behind the star as it circled around. This orbit would give good radial velocity variations, as the planet would tug on the star from both in front and behind, causing changes in the star's velocity along the radial line-of-sight direction. On the other hand, assume tha
t the orbit is seen exactly face-on, so the planet appears to be traveling in a circle around the star, sometimes above, sometime below, sometimes to one side. This orbit would give no radial velocity variations at all, as the tug of the planet would always be exactly at right angles to the radial direction. So a big extrasolar planet in an orbit that is seen almost face on from Earth could cause the same radial velocity variations as a small planet in an orbit that is seen edge-on, a simple fact of geometry. The masses given for the planets of HD69830 assume that their orbits are seen exactly edge-on, so their actual masses could be somewhat bigger.

  Even if they are really about the same mass as Neptune, though, their compositions may be quite different from Neptune and from each other. We can speculate about the makeup of these planets based on our knowledge of planetary formation. Neptune has a gas fraction, or hydrogen and helium content, of only about 10% of its total mass. The rest of Neptune is made of rock and “ices,” what planetary scientists say when they mean water, methane, ammonia and so on, compounds that are normally frozen at temperatures typical of the outer solar system. For this reason, the term “ice giant” has come to be used for bodies like Neptune, to distinguish them from the real gas giants like Jupiter and Saturn that are mostly composed of hydrogen and helium. The composition and structure of Neptune is reasonably well known and so may tell us something about the makeup of the three new Neptune-sized planets around HD69830. Neptune has a hot interior, so its “icy” component is really a superheated ocean made mostly of water. Above the ocean is a crushing atmosphere consisting mostly of hydrogen and helium, although it is likely that the density of the gaseous atmosphere just above the ocean is similar to that of the liquid ocean just beneath it, so it would be hard to tell where the atmosphere ends and the ocean begins. Temperatures at the “surface” of the ocean are about 2,000K and atmospheric pressures are about 100,000 times that at the surface of the Earth, or about 100 times the pressure at the bottom of the deepest regions of Earth's ocean. Neptune is an extreme example of a giant water world. Smaller water worlds are likely to be very common around stars in the solar neighborhood, as simulations of planetary formation produce them routinely.

  These simulations also show that there are often big changes in a stellar system over geologic time. There is always a tendency to assume that large objects like planets and stellar systems have always been as they are at present. But this is not the case, and like all stellar systems, the system of HD69830 has evolved since its formation. Firstly, the three planets probably were not always in their present-day orbits. One of the most shocking discoveries of the early period of extrasolar planet exploration has been that Jupiter-sized planets are often found very close to their parent stars, sometimes with orbits of days rather than years. Simulations of planetary formation find it difficult to produce large planets in these close-in locations. So the hypothesis is that they formed farther out but then migrated inwards over a period of millions of years, due to their orbital speed being slowed down by the drag of the remaining dust and gas left over from planetary formation. Some of them would have migrated all the way into their stars and been quickly vaporized; others stopped migrating agonizingly close to the stellar surface and so are being slowly vaporized, their original gaseous envelopes being blasted away by the intense radiation, the molecules boiling into space like the tail of a huge comet. The three planets around HD69830 are close enough to the star to be strongly heated; the evaporation process may have affected the innermost planet HD69830b significantly, with a considerable fraction of its original gas disappearing.

  Simulations of the migration process also tell us something about the interior compositions of the three planets and their structure. As the planet closest to its star, HD69830b would have originated inside the so-called “iceline,” the distance from the primary where water ice is stable indefinitely in full sunlight. Inside this line, ice evaporates instead of accumulating into larger bodies. This means that ices are not a major constituent of HD69830b, so it is likely mostly rocky. Its remaining atmosphere after evaporation is probably still quite thick, with a size of a couple of Earth masses, giving a surface pressure of about 100,000 times the surface pressure of Earth. Just like Earth, it might have a global ocean generated by the emission of water from volcanoes, and it might have some land areas as well. The outer two planets, though, likely formed outside the iceline and thus have a substantial component of ice in their makeup, making it very likely that they are completely oceanic, more like Neptune. Due to their greater distances from their star, they did not undergo substantial evaporation and so have retained thicker atmospheres.

  All three planets, though, must have substantially different climates than the Earth's. Calculations suggest that the habitable zone of this system, the region where the climate of a planet is hospitable for life over periods of billions of years, is centered approximately on 0.8 AU from the star. All three planets are closer to their star than this distance. The innermost two planets are far too hot to be habitable: the surface of HD69830b would be a boiling cauldron of superheated water or glowing rock, and HD69830c likely has no real solid surface. HD69830d is closer to the habitable zone but again is Neptune-like: if there is any life there, it is airborne and hydrogen breathing. But the three planets discovered so far are unlikely to be the only ones in the system: there are almost certainly a number of smaller bodies as well. Current observational techniques cannot resolve Earth-sized planets around stars as big as HD69830, although this will change in the near future as new observing systems come online. In the meantime, simulations can give us an idea of where to look for smaller planets. They tell us that planetary orbits are stable between 0.3 and 0.5 AU from the star, and again between 0.8 and 1.2 AU, or within the habitable zone. Thus there is room in this system for a terrestrial planet with a reasonable climate. There is also another stable location in the so-called “Trojan” region of HD69830d, two zones in the same orbit as that planet but ahead of it and behind it. These areas are named after small asteroid belts in similar stable regions of Jupiter's orbit in our system. Whether a terrestrial planet in this part of the HD69830 system would have a good climate or not is highly debatable, as a planet at this distance from the primary would receive 50% more radiation than the Earth gets from the Sun, or about the same as Venus, and Venus is not a pleasant place.

  Whether potentially habitable worlds exist in this system remains to be seen. Nevertheless, their presence or absence depends to a large degree on the history of the system, particularly if the migration of the three Neptunes has affected planetary formation. Planetary migration also provides a possible way to resolve the mystery of the hot dust, a phenomenon that has implications for the habitability of planets around Sunlike stars, including our own.

  * * * *

  Bombardment

  The habitability of our own solar system has varied over geologic time. For example, the distribution of the age and numbers of craters on the rocky planets and moons in our system indicates that there was a period lasting for about 100 million years, ending about 3.85 billion years ago, when there was a sudden jump in the number of impacts in the inner solar system. This period of time is known as the Late Heavy Bombardment (LHB): it was the last gasp of planetary formation before the cratering rate settled down to roughly its present value. One possibility is that the LHB was caused by the migration of Jupiter and Saturn. Migration could be caused by two possible processes: in addition to the gas drag mechanism already mentioned, it can also be caused by the gravitational interaction between a planet and the millions of smaller bodies still lurking about after the formation of the solar system. Because of Newton's laws, if one of these fragments were tossed out of the solar system by a large planet, there would be a countervailing change in the planet's orbit. This is a small effect, but if it happened many times there would be a cumulative impact on its orbit that would be substantial. This may have caused Jupiter and Saturn to migrate until the periods of their orbits became exa
ct multiples of each other, in this case 1:2. This so-called “resonance” effect, combined with the mysterious phenomenon of chaotic motion, then made the orbits of Uranus and Neptune unstable and tossed them into the Kuiper belt. At that time, the Sun's Kuiper belt was much more massive than it is today, more like the present-day giant Kuiper belt of the very Sunlike star HD76151. The gravitational attraction of Uranus and Neptune subsequently tossed a large number of Kuiper belt objects into the inner solar system. The collisions of these bodies in the inner solar system would have temporarily raised the dust level, creating hot dust just like that observed around HD69830. So it is proposed that we are today witnessing an LHB in the HD69830 system, caused by the migration of a planet into the Kuiper belt of that system.

  If this planet exists, it hasn't been discovered yet. Calculations suggest that the three Neptunes now have stable orbits but that there is a region between 0.3 and 0.5 AU from the star that is much less stable. Perhaps in the relatively recent past a large body located at this distance from its star was perturbed by the three Neptunes into a wandering orbit, tossing comets and asteroids about and creating an LHB event. A disturbing thought is that this is happening to a star that is about two billion years old. The implication is that LHB events could occur quite late in the evolution of a stellar system. Simulations suggest that the timing of an LHB event depends on the rate of migration of the planetary orbits, the masses of the planets and the distances between them and their proximity to the system's Kuiper belt. So it is conceivable that LHB events could be delayed until a very long time after the birth of a stellar system.