Friday, February 12, 2016

A Stranger at Home

An interesting consequence of being asked the same question repeatedly, is that you stop thinking about the answer. Instead, you find yourself reciting a variant of the same prefabricated response that you gave the previous twelve times. Naturally, in this mode of operation, your brain is susceptible to being stumped by otherwise trivial inquiries, simply because you haven’t heard them before and they don’t automatically register in the existing database. Recently, I found myself in exactly this situation. 

A recurrent question about which Mike and I have thought extensively is “what if you’re wrong?” For an extended discussion about this possibility, scroll down to the previous post. But the perplexing question, posed to me by a reporter some time ago, was a different one: “what if you are right?” In all honesty, my first reaction was “huh? What does this even really mean?” Of course, we hope that we are right! We hope that the dynamical mechanism connecting the alignment of the distant Kuiper belt orbits, the detachment of Sedna-type ellipses from Neptune and the mysteriously inclined trajectories of large semi-major axis Centaurs, is a chaotic web of mean-motion resonances facilitated by Planet Nine. Moreover, we hope that Planet Nine will be observationally detected, like, as soon as possible, ya know what I’m sayin’?… 

But on second thought, it is evident that I was being dopey. This question has considerable depth. If we are right, the clockwork of our solar system is about to acquire a very aberrant new gear, and this has profound implications for how our strange cosmic home fits into its extrasolar context. More specifically, the detection of Planet Nine would render our solar system a slightly less abnormal member of the Galactic planetary census.

In order to understand just how unusual the architecture of the known solar system is, it is useful to dial the clock back to late November of 1995 - that is, to the discovery of the first planet around another sun-like star. With a mass slightly larger than that of Saturn, this object (dubbed 51 Peg b) is bonafide giant planet. However, unlike Jupiter and Saturn, that require more than a decade to finish a single revolution around the sun, 51 Peg b completes its orbital trek in a little over four days. Indeed, the first proof that planets around other main-sequence stars are extant also provided the first hint that orbital architectures of extrasolar planets can be very different from that of solar system’s planets.

Observational characterization of more expansive giant planet orbits during the subsequent decade and a half continued to yield surprises. Evidently, long-period giant planets tend to occupy eccentric, rather than circular, orbits. The figure below shows the semi-major axis - eccentricity distribution of well-characterized extrasolar planets. While the solar system giants would be found scraping the bottom of this figure, exoplanets clearly occupy the the entire eccentricity range, with a nearly parabolic orbit of HD 20782 b at the helm of the population. 

Figure 1: semimajor axis - eccentricity distribution of well-characterized extrasolar planets. The predicted orbit of Planet Nine is shown as well.

More recently, the triumphant success of the Kepler mission showed that the default mode of planet formation in the galaxy generates objects that are somewhat smaller than Uranus and Neptune, but are substantially more massive than the Earth. In other words, planetary masses of order ~10 Earth masses are not only prevalent in the exoplanet catalog, they are dominant. Although the transit technique limits the observational window of Kepler to orbits inside ~1AU, there is little reason to suggest that more distant orbits should be devoid of such planets.

Figure 2: the catalog of planetary candidates detected by Kepler. The sizes of the depicted points are representative of the corresponding planetary radii. The semi-major axes are shown on a logarithmic scale. Figure from Batygin & Laughlin 2015.

Cumulatively, a distinct picture of the Galactic planetary census is beginning emerge, wherein the ordered orbits of the known planets of the solar system are starting to appear increasingly abnormal. On the other hand, with a characteristic mass approximately 10 times greater than the Earth and an eccentricity of ~0.6, Planet Nine fits into this extrasolar planetary album seamlessly. Intriguingly, this yet-unseen world may provide the closest link between our solar system, and the extrasolar realm. Indeed, Planet Nine may constitute the closest thing to the solar system’s very own extrasolar planet.

Why I believe in Planet Nine.

In my last post I went into detail on ways in which our Planet Nine hypothesis could be wrong, and I suggested for you, if you’d like to be a Planet Nine skeptic, which you’re encouraged to be, what new observations you should be looking for before you start to believe it yourself. Here, I’m going to tell you why I already am a believer in Planet Nine and why maybe you should be too.

As we’ve discussed, the Planet Nine hypothesis was initially developed to explain one simple phenomenon: the alignment of the most distant objects in the Kuiper belt. The existence of that alignment looks pretty compelling, but even when you calculate things like a 0.007% chance that it could happen due to chance you still worry about the fact that there are only 6 objects that you’re talking about. Still, Konstantin and I worked on this for about a year until, by about late last summer, we had a nice comprehensive theory which could explain how a massive planet on an elongated orbit could capture equally orbitally-elongated Kuiper belt objects into protected mean-motion resonances. It was a fun result with some cute physics to it, as no one had really considered the effect of such extreme planetary eccentricities on populations of small objects before. It’s always a good day when you learn something new about the ways in which planetary physics can work.
The whole point of Planet Nine was to explain the orbital alignment of these six objects. The number of other phenomena that Planet Nine also explains -- essentially by accident -- is astonishing.

A particularly satisfying aspect of the hypothesis was that it neatly and eloquently explained the peculiar orbit of Sedna.  I have written elsewhere on what is peculiar about Sedna’s orbit and why it demands an explanation, and I have spent 12 years searching for solutions to Sedna’s peculiar orbit, and here was an explanation where we hadn’t even been looking for one. In short, Sedna is peculiar because it has been pulled away from the Kuiper belt by something. And to be pulled away from the Kuiper belt there needs to be something beyond the Kuiper belt to do the pulling. Back when we discovered Sedna, we proposed that perhaps that something was a planet! Or a passing star! Or the cluster of stars that the sun was born in! We didn’t really know. With only a single object there were more possibilities than answers. But as we continued surveying the outer solar system and found no new bright planets out there, we gradually settled into the view point that the most likely explanation was that Sedna had been pulled away from the Kuiper belt by the combined effect of the nearby stars that formed along with the Sun 4.5 billion years ago. This proposition was exciting: Sedna would be a fossil record of the birth of the Sun itself, and finding more of them which teach us about that time period.

Now, however, we have a simpler explanation. If a planet is forcing the most distant objects into alignment, it will also take these most distant objects and periodically pull them away from the Kuiper belt before pushing them back in. In fact, the Planet Nine hypothesis demands that objects like Sedna, and also 2012 VP113, a more recently discovered by similarly odd object, exist. After 12 years of searching for the explanation for Sedna we found it by trying to explain something else entirely.  

Interesting side note: As I was writing this post I noticed something that I hadn’t before. It’s not just Sedna and 2012 VP113: all of the distant objects which are pulled even a little bit away from the Kuiper are in our cluster (specifically, if you look at all objects with semimajor axis>100 AU and perihelion > 42 AU).  Wow. 

That’s not bad. As a scientist, you would love to form a hypothesis that makes predictions that turn out to be true. That makes you begin to believe in your hypothesis. In this case, we didn’t predict the existence of Sedna and then go find it, but rather we knew about Sedna and accidentally came up with a solution. That’s more of a two-birds-with-one-stone situation than a prediction, I think. Still, we were quite pleased.  While previous speculation about planets beyond Neptune had struggled to find viable explanations for even single phenomena, we had come up with a relatively rigorous theory which naturally explained two seemingly unrelated phenomena.

At this point I think that Konstantin and I were mentally ready to publish a paper with a conclusion something like “here’s a nice theory which explains two different things and hey it’s even quite plausible!”

What happened next is what made me go from finding the explanation plausible to finding the explanation likely. While sitting in my office looking at the outputs of our gravitationally simulations, Konstantin and I realized that Planet Nine had another major effect that we hadn’t anticipated. Some of the objects with very distant elongated orbits had their orbits twisted so that instead of being more or less oriented along with the disk of the rest of the solar system, they were essentially perpendicular to it. And, when they happened, instead of being lined up with the other distant objects, their orbits swung off to the left or to the right by nearly 90 degrees. I described these orbits as “wings” because that’s how they looked in the simulations.

Objects with perpendicular orbits? I remember when one was discovered a few years ago. It was so unusual that it was nicknamed “Drac,” in honor of Dracula’s ability to climb on walls. Or something like that. I was quite excited to quickly look up the orbital parameters of Drac and see if its orbit corresponded to the location of the wings, but, to my chagrin, Drac was the wrong sort of object. I had remembered correctly that Drac was perpendicular, but its orbit did not go nearly far enough from the sun to be affected by Planet Nine. And it was not even pointing in the right direction. The origin of Drac was still a mystery, but it didn’t seem connected to Planet Nine (oh but it is; more later!).

While Konstantin and I were still sitting in my office, disappointed by Drac, I thought to look at the complete database of all of the object discovered in the outer solar system, and, to my surprise, there was a collection of objects that were not part of the Kuiper belt that we had overlooked. These were object which, though though were quite elongated and went to great distances, traveled far inside the orbit of Neptune – coming nearly to the orbit of Jupiter in some cases – before swinging back out to the distant reaches of Planet Nine. We had ignored these objects previously because we knew that when objects came into the giant planet region their orbits would be modified by interactions with the planets. What we hadn’t anticipated is that objects coming in on perpendicular orbits would have much less of a chance to have their orbits modified. Our simulations showed that objects with distant elongated perpendicular orbits which came close to the giant planets still maintain their alignment to the wings.

When we realized this, Konstantin stay riveted in his chair in my office while I plotted the locations of these objects which we had overlooked. There are 5 of them. I told him, “If these are right where we predict they should be my head is going to explode.” I plotted them. Four are on one of the wings, the fifth is on the other wing. Right as predicted. My head did not actually explode, I think, but it is possible that my jaw hit the floor. We were both silent for a minute, and Konstantin said, in a semi-amazed voice, “This is actually real, isn’t it?”
The distant objects with orbits perpendicular to the solar system were predicted by the Planet Nine hypothesis. And then found 5 minutes later.

Yeah. I think it’s real. As Konstantin later said, “It’s like killing two birds with one stone and not even realizing there was a third in the tree and killing it, too.” The existence of the elongated perpendicular Centaurs – as those objects are called – was a pure prediction that was dramatically confirmed. Sadly, the rest of the world didn’t get to participate in the drama, as it all took place over the course of about five minutes in my office last fall, but trust me on this one: the drama was there.

And Drac, which had been such a disappointment? Once we started looking we realized that our gravitational simulations create Drac, too. Sometimes, when the elongated perpendicular Centaurs do get too close to giant planet, that planet pulls their orbit a little close, and also swings the orbit around randomly. Another Drac is born. The Planet Nine hypothesis requires the existence of objects with orbits like Drac, which otherwise had no plausible explanation.

Does that make four (five?) birds yet? Hard to keep count.

Here, then, is the summary of my reactions to each of the four (now five) things explained by Planet Nine
  1. A distant massive eccentric planet can capture eccentric Kuiper belt objects into elongated anti-aligned orbits like the ones we see: Hey, that’s cool!
  2. The Planet Nine hypothesis explains Sedna, and requires Sedna to exist: Wow. That’s a really nice hypothesis that sounds pretty plausible!
  3.  The existence of Planet Nine predicts the existence of elongated distant perpendicular Centaurs in specific locations and they are then found to exist. Holy cow. Planet Nine is real!?!?!
  4. The Planet Nine hypothesis explains the unusual orbit of Drac and requires that objects with orbits like that will exist: Of course it does.
  5. The Planet Nine hypothesis explains why all of the distant objects which have been pulled away from the Kuiper belt are equally clustered: Any vestigial doubts have vanished.
 At this point my main question is “what unusual phenomenon in the Kuiper belt does Planet Nine not explain?” (We have, regretfully, come to the conclusion that Planet Nine cannot account for the parting of the Red Sea or the waning of the ice ages, though both of those possibilities have been suggested to us multiple times).

So I believe. But it’s OK if you’re not ready to believe. Unlike some hypotheses, this one has a definite proof. We have to go find it. We will. I have very little doubt that we will.