Eyes on Alpha Centauri: The Hunt for a Pandora
Michael here. Could there be habitable planets orbiting amidst the three stars of the Alpha Centauri system? What are the chances that a Polyphemus, or even a Pandora, really exists? If you know anything of James Cameron’s passion for science, you should know that the answer to the question, “could there be planets in the Alpha Centauri system?” is a resounding “YES!”
Picture (click to enlarge): Alpha Centauri A and B hang over the rings of Saturn. Click here to read more about this picture
Scientists are putting money on it. While Cameron is busy creating a moon orbiting a gas giant circling Alpha Centauri A, Professor Debra Fischer of Yale University, working at the Cerro Tololo Inter-American Observatory (CTIO) in Chile, is aiming the CTIO 1.5 meter telescope at both Alpha Centauri “A” and “B” as a part of a 5 year observation that will hopefully reveal planets as small as Mars.
She is using the radial velocity method, which uses spectral measurements to detect variations in the speed that a star moves towards or away from Earth (any star with planets will move in its own small orbit around a common center of gravity).
With our solar system largely explored, the stars beckon us ever more forcefully. And among the stars, the Alpha Centauri triple star system, the closest star system at only 4.37 light years from our Sun, is calling out the loudest for exploration.
Think of how awe-inspiring it would be to know that Alpha Centauri harbors potentially habitable planets.
Read below for my conversation with Prof. Fischer, in which we discuss the current state of thought regarding the chances of habitable planets in the Alpha Centauri system, her observational program that is working to discover these planets, and the possibility of sending a probe to explore the star system directly:
Michael: I appreciate you taking the time to talk about Alpha Centauri. I’m interested in the state of thinking regarding the possibility of planets or planetary systems existing around either Alpha Centauri "A", Alpha Centauri "B", or even around the entire system, if that is even possible.
Prof. Fischer: Yes, in fact it would be possible. There could be a stable orbit around both stars. But, we’re searching for planets orbiting either alpha Cen A or alpha Cen B individually.
Michael: So, is the Cerro Tololo telescope that you’re using focused primarily on Centauri B?
Prof. Fischer: Actually, we’re looking at both “A” and “B”.
Michael: I remember reading that "B" would be a better target because “B” would be less active, with less solar activity that would interfere with your spectral measurements. Is that still true? Are you still hopeful that you might find something around Alpha Centauri “A” as well?
Prof. Fischer: As stars go, alpha Cen A and B are inactive. But, we don't know if there are velocity fields at the sub-meter-per-second level. The stars are not completely smooth cue balls in space, like pool balls on your table. They have atmospheres, and those atmospheres have flares. In addition, the stars themselves pulsate. We know that Alpha Centauri "A" has a dominant pulsation period that’s about 5 minutes, just like the sun. I’m not really worried about periodic variations of minutes. We'll be able to average right over that kind of noise. But no one knows whether or not there are long period variations in "A" or "B". We do know that "A" is probably not as stable as "B". People have measured pulsation periods in "A", and in general we find that more massive stars have more active atmospheres. So yeah, I agree that "B" is a better bet. And if you told me that I could only observe one star, I’d choose B. But we’re studying both stars. And it turns out that this strategy of looking at both stars is pretty critical in ensuring a solid set of data.
Michael: Do you have two telescopes, one constantly trained on “A” and one on “B”?
Prof. Fischer: No. We use one telescope, but we nod back and forth. The exposure times are 15 seconds for A, and 30 seconds for B. So we take ten observations of A, and then we move over to B and take ten observations there, then nod back to A and so on. So it’s just one telescope going back and forth. This is helpful because if we see the velocities of both stars just jump up or down for some reason, this will be a clue that there’s some kind of systematic error, and I hope that we’ll be able to track it down.
Michael: Error due to stellar activity?
Prof. Fischer: In this case, I’m thinking of errors that we introduce. Perhaps there is an instability in the spectrometer. Or maybe we have the wrong times written down - the Earth is travelling at 30,000 meters per second, and we have to account for and remove that velocity. If that’s not done perfectly, then we would incorrectly attribute the Earth's velocity to the alpha Cen stars.
Michael: The first time I read about your project was March or April of 2008, and at that point I think you were planning on starting in June of 2008. So it’s been going on for about a year now?
Prof. Fischer: We commissioned the spectrometer in May of 2008 and collected observations for a month in May. We received discretionary time from the director, who liked our project and I received a few weeks of telescope time through NSF (National Science Foundation) in August and September. But the project started in earnest on January 1st of this year.
Michael: Two questions. How long would it take to find a Hot Jupiter? (A Jupiter-sized or larger gas giant orbiting very close to the star). A Hot Jupiter would cause a larger Doppler shift that would be detectable more quickly. I would assume that detecting a more Earth-sized planet would take a number of years. So do you have a timeframe in mind – a hopeful timeframe – as to when you may be able to tell that there’s a Hot Jupiter, and then another timeframe in which you would hope to find a smaller, more Earth-sized planet?
Prof. Fischer: Ah, we are quite sure that there are no Jupiters or Saturns orbiting "A" or "B". First of all, people have collected data and those types of planets haven't been found. We think that makes sense, because alpha Centauri is a binary star system [actually, the very distant Proxima Centauri makes it a triple system, but that star is very far away from "A" and "B"], there are only stable orbits out to about 2 times the Earth to Sun distance around either star. Any planets at greater distances would be gravitationally destabilized by the other star in the system. So, we are only looking for objects that have orbital periods out to about 4 or 5 years. That’s when the program ends – in 4 or 5 years we’ll be done.
Michael: Could you go into more detail on why planets cannot exist beyond 2 times the Earth to Sun distance?
Prof. Fischer: The stars themselves are in a very elongated, eccentric, elliptical orbit. At the widest part of their orbit, they are almost 40 AU (40 times the earth to sun distance) apart. That's a comfortable separation for orbiting planets. But, when the stars, A and B, make their closest approach, they are only 11 AU away from each other. The stars are moving into the domain of the planets, and they will gravitationally rip away any planets that aren't closer than 2 AU around "A" or "B". Several computer simulations have demonstrated this. If you start the simulation with a system that has more distant planets, as the other star comes in closer, the planets are gravitationally disrupted, dislodged from their stable orbits. The orbits first become eccentric, and then the planet shoots right out of the system.
Michael: Does the 2 AU maximum distance apply to both Alpha Centauri "A" and "B"? Does the maximum apply for any size planet, or would the maximum change depending on the mass or orbital period?
Prof. Fischer: [The maximum distance] is not quite as sensitive to the mass of the planet as one might guess. However, because star “B” isn’t as big as "A", it’s tugs less on planets that are orbiting “A”. “A” is a more massive star, so it pushes more on any planets that would be orbiting “B”. But the difference in the stable zone is relatively small.
Michael: I’ve heard Chile is a prime location for an observatory. What was the reason that this location was chosen for the observatory?
Prof. Fischer: Well, it’s a great site. We had to observe from the southern hemisphere because the alpha Cen stars can be seen for most of the year. This site in Chile was originally chosen because the atmosphere is quite stable and the weather is really good. We’ve been observing since January and I can count on one hand the number of nights that we’ve been shut down.
Michael: So very little cloud cover, smog, or pollutants?
Prof. Fischer: Yeah, that’s right. We’re far away from any cities, and there’s an ocean breeze that rolls over the top of the mountaintop and it creates what’s called a laminar flow, that’s very stable.
Michael: You are using a 1.5 meter scope, correct?
Prof Fischer: That’s right. It’s a small telescope. I also use the Keck telescope, which is the world’s largest telescope and I use a 3 meter telescope at Lick Observatory. So the telescope is small, but it’s all we need because these stars are so close to us, and they are so bright.
Michael: Why hasn’t there been a dedicated observation of the Alpha Centauri system until now? It would seem the obvious choice.
Prof. Fischer: It’s because we learned early-on that finding planets was a "numbers game." We found planets similar to Jupiter, a gas giant planet, around 5 to 10% of the stars. So we had to look at a lot of stars to find planets and publish results. Now that we’ve found a lot of planets, we're taking a different strategy: choosing two stars and trying to average down our precision so we can detect an Earth. That’s something that we haven’t been able to do yet. So this strategy of "high cadence" observations is a new idea. Our rival team, the Swiss team, lead by Michel Mayor, is working on HARPs, a spectrometer on a telescope at La Silla in Chile (which I can see from CTIO). They’re also looking at alpha Centauri. So I think we’re in a sort of friendly race. And it’s exactly the right thing to do; we need to have confirmation. It's important that both teams see this tiny whisper of a signal.
Michael: I know that if NASA gets enough funding, there could be a few space-based planet hunting missions. The SIM Lite mission is an astrometry mission which would directly observe the circular motion of the star based on the existence of a potential planet.
Prof. Fischer: That’s right. And that would be the most important mission for this particular project.
Michael: Any updates on potential funding for SIM or other planet hunting missions?
Prof. Fischer: As a community astronomers try to set priorities in the form of strategic roadmaps and the "Decadal Survey" to help advise both NASA and the National Science Foundation. I think there’s a good plan in place. As the first priority NASA has just repaired the Hubble Space Telescope. And Kepler was recently launched. The next priority for a major mission is the JWST (James Webb Space Telescope) which will replace Hubble. Then, the plan is for a flagship exoplanet mission. The two competing ideas right now are an astrometric mission like the Space Interferometry Mission, and an imaging mission to take pictures of planets orbiting nearby stars, conceptually, like the Terrestrial Planet Finder.
They’re both fantastic ideas - there's no question that eventually we want to image the planets. But I think that an imaging mission may have greater technical risk associated with it. There’s still a lot of work to be done to demonstrate that technology is in hand to image close-in planets. In contrast, an astrometric mission like SIM has had extensive technology development.
Michael: If you detect a planet, that would certainly help NASA get the funding it needs for SIM. I’m just crossing my fingers – I want everything to happen.
Prof. Fischer: Exactly, me too!
Prof. Fischer: It would have been possible, but we’ve looked and they’re not there. This kind of planet would be fairly easy to detect. Alpha Centauri has been observed by folks at the Anglo Australian telescope, and by our Swiss colleagues with HARPS. In both cases giant planets have not been detected. This is good news for prospective "Earths." If there were giant planets around alpha Centauri "A" or "B," they would have destabilized orbits of small rocky planets.
Michael: One thing that I read on another site was that you were observing other stars as well, including Tau Ceti.
Prof. Fischer: That’s right. These are our standard stars. They help us to debug our analysis and the instrument. [They are] our calibrators.
Michael: Kepler, which uses the transit method, recently launched, and it was placed in an orbit in which it could view a patch of stars without that portion of the sky being obscured by the sun during any portion of its orbit. Would it be possible to use the transit method to detect planets around Alpha Centauri?
Prof. Fischer: The probability of a planet transiting in that system is pretty low. We think that the planets that formed around “A” or “B” would have been dynamically forced into the same orbital plane as the binary star system. That’s been demonstrated through computer simulations. And we know that the tilt of the plane is close to edge-on, but probably not close enough for a transit to be seen.
Michael: A similar problem could have occurred with radial velocity measurements like the ones you are doing. If you are looking at a star system top down, obviously you won’t be seeing anything going towards you or away from you. So how is the Centauri system oriented in respect to your observatory?
Prof. Fischer: Right. Right. What we measure is the velocity coming towards us and away from us. So we wouldn’t see velocities perpendicular to our line of sight – so the velocity in the plane of the sky. We're helped again because the orbital plane of Alpha Centauri “A” and “B” is 79 degrees (90 degrees is perfectly edge-on). That's important; since the planets will inherit this same orbital tilt, most of the reflex velocity of the star is along the line of site. That's one of the reasons we chose this system. We knew ahead of time that any planets would have an optimum orbit for detection with our technique.
Michael: It’s such an obvious target - it’s surprising no one did this before.
Prof. Fischer: Yeah, but astronomers are lucky to get a few nights of telescope time a semester. Sometimes, people get 20 nights, 50 nights, something like that, on smaller telescopes. So, you have to decide: do you take all of that time, and put it into one star, where after four years you may come up empty handed? Or do you mitigate your risk and look at lots of stars so that you’re sure to find planets?
Michael: Now that we have so many results (planets found), we can start focusing more directly on certain star systems…
Prof. Fischer: Yes, I really think it’s the right science to do. And I feel like I can afford to do this risky project because I’ve got other planets rolling out from other projects.
Michael: I have a few astrobiology type questions I was wondering about. I know that Alpha Centauri "B" has a different spectrum – the more orange K1V – as opposed to the Sun and Alpha Centauri "A", which are G2V. Would that spectral type, as far as you know, prohibit the development of life, or change the potential atmospheres in a way that would hurt the chances of a planet evolving some kind of lifeforms?
Prof. Fischer: Maybe the biggest question is whether or not water could survive on planets in that system. Interestingly, we think water was delivered to Earth by water-rich asteroids that orbited the Sun at distances beyond 2 AU. If close binary stars disrupt the orbits of water-rich asteroids around alpha Cen A and B, maybe rocky planets there don't have oceans. That would make life as we know it impossible. On the other hand, models of planet formation suggest that the presence of that second star actually helps to assemble the planet faster - maybe that also plays a role in rapidly delivering water to any planets in the system. Our understanding is so immature at this point, that anything is possible and it's best to just look. We are trying to make comparisons based on a single example, our solar system. The alpha Cen stars may be perfectly lovely places for planets with life. The stars emit light that is very much like the radiation from our Sun. The brightest star, "A" is the most massive star, but it is only about 10% bigger than our sun. And “B” is about 30% smaller or less massive than our Sun. Compositionally, the stars are quite similar to our Sun, so the same building blocks for our life existed around these stars.
Michael: That’s sounds very hopeful...
Prof. Fischer: The Alpha Centaurians could be sitting on their planet observing our sun and theorizing that, "there’s no way that poor single star, Sol, has planets with life. It’s a single star, it wouldn’t have made planets fast enough to accrete water."
Michael: I know that “A” and “B” range from about 11 to 36 AU. Would that cause major climate changes during the 79 year orbital period?
Prof. Fischer: Certainly that would impact the atmosphere on the planets but I don't know how that would play out. There have been so many climate cycles on our planet and it's complicated and tied to the ratio of water to land mass. 11 AU is still quite a distance from either planet. It's an interesting question!
Michael: So let’s say we had another sun, equivalent to our sun, that was 11 AU from us right now. How would that affect us in terms of heat, radiation, solar wind…
Prof. Fischer: I haven’t seen any work on this, though it would be a very interesting investigation. It’s a great idea. The second star would be further away from us than Saturn is in our solar system. So it’s pretty far away. But a second star would impact the planets – you’re right!
Michael: With the new Star Trek movie, there’s been an influx of articles about NASA’s advanced propulsion research.
Prof. Fischer: It's fascinating! If we do find rocky planets around Alpha Centauri “A” or “B”, then NASA may think about how they could send a probe there - maybe a probe the size of a cell phone now. If the probe accelerates to 10% of the speed of light, it would take about 40 years to reach Alpha Centauri, and then about 4 years to radio the signal back. Both of those things (accelerating to 10% the speed of light and phoning home from alpha Centauri) are hard to do.
Michael: The timeframes, if you think about it, aren’t actually much bigger than some current missions. There are plans on the drawing boards for missions that won't produce results until potentially well into the 2030's.
Prof. Fischer: True. The Voyager spacecraft launched in the 70’s, and that’s 40 years out now. It's within a human lifetime. You can imagine doing something like that.
Michael: I need to read up on my advanced propulsion technologies, but I believe that Project Orion, or nuclear thermal propulsion rockets could achieve speeds close to that. Especially if everything that is not the fuel tank is so incredibly small, I can definitely see that happening.
Prof. Fischer: With a platform in space, the space craft can be assembled outside of the Earth’s gravity. People are thinking about this. It’s so tantalizing, it’s just -- almost -- within reach.
Michael: Do you know whether SETI or anyone else has been beaming signals directly at the Centauri system?
Prof. Fischer: I'm not sure, but I'd bet that this is a prime target. Frank Drake is the astromomer who searched for radio signals around Tau Ceti and Epsilon Eridani. He was working from the northern hemisphere, so the Alpha Centauri system would have been out of reach. Just think - our television signals have swept past the star system. We’ve been transmitting for more than 50 years. They’re seeing five year-old shows. If they're watching, the Alpha Centaurians are probably more tuned in than I am!
Michael: The Hubble recently directly imaged the first extrasolar planet. Is it possible that Hubble could be trained on Centauri to look for planets?
Prof. Fischer: That was around HR 8799. This is interesting because this image is the first example of what astronomers hope to do. I showed [these images] to my class when they came out – they’re very exciting. The planets are many times the mass of Jupiter and at quite a distance from the star. It’s much more difficult for Hubble to image small planets, especially in close orbits around Alpha Centauri “A” and “B”. You really need a mission like the Terrestrial Planet Finder.
Michael: I think it would be such a huge cultural, media event if we could confirm an earthlike planet, or even a super earth, around either of the Centauri stars.
Prof. Fischer: Yes, I agree.
Michael: Do you think you could detect a Mars size planet?
Prof. Fischer: That’s what our simulation showed - that we could reach down that low. But this all hinges on the stability of the stellar atmospheres.
Michael: That's so exciting. Maybe it's just wishful thinking, but I think there's a very, very high chance of planets around either or both stars.
Prof. Fischer: I started this project with Greg Laughlin. He’s a brilliant theoretician and a professor at UC Santa Cruz. One day Greg and I were in a meeting together - we both served on the exoplanet task force commissioned by NASA and the NSF. At one point Greg said to the whole committee, “I’ll bet my career that there are planets around Alpha Centauri A AND B.” And I just thought, wow! His confidence inspired me.
Michael: It's certainly a bet I would make.
Thanks so much to Prof. Debra Fischer for taking the time to talk Alpha Centauri!
So, could a Polyphemus / Pandora pair exist? The answer is that it would depend on where James Cameron placed Polyphemus in relation to Alpha Centauri A.
According to Prof. Fischer:
"If there were gas giants within 2 AU of either star, they definitely would have been discovered by now. [In regards to planets orbiting beyond 2 AU], the theorists would say that it is not possible, but we have 4 counter examples of close binary systems with gas giant planets orbiting at 2.5 AU, where simulations show it would be unlikely that they would even form (once they form, they can survive in a stable orbit). A new idea I heard at Cambridge University is that the planets may migrate outward! Therefore, they form closer in (and faster because of the binary star system) and then they migrate out a bit."
So while there is no Polyphemus within the habitable zone (where liquid water can exist) of Alpha Centauri A, there is a chance - despite theoretic models to the contrary - that gas giants may exist beyond 2 AU. So a Polyphemus / Pandora pair could exist beyond 2AU. But if that's the case, then it would require an alternate source for heating, as liquid water would otherwise not exist at this distance from Alpha Centauri A.
However, a Polyphemus certainly could have formed within 2 AU in an alternate universe, and there certainly could be Pandora-sized planets with liquid water directly orbiting Alpha Centauri A.
In any case, we'll have to wait and learn more about exactly where Cameron placed his planets, and any other explanations he may have come up with!
