Alpha Centauri DOES have planets!

Ever since I wrote the two posts about the potential for planets at Alpha Centauri, those posts have been the most popular on this blog. With all the exoplanet action, this discovery was only a matter of time, because it seems that more than 90% of stars–maybe substantially more than that–have planets. I’ve written here about why it can be very hard to detect planets, even in stars that are “close”.

So it seems that Alpha Centauri B has a planet, which is now boringly named Alpha Centauri Bb. I think it needs a more interesting name.

The planet in question is about 13% more massive than Earth, which suggests that it is rocky, but orbits the Sun-like star far too closely to harbour life. The distance between it and the star is a mere 6 million kms (vs 150 million for Earth) and the surface temperature a mere 1200-odd degrees which makes it, as scientists wryly remark “unsuitable for life”. Or at least life as we know it. Science Fiction writers can go crazy here.

It is my prediction that if this planet exists this close to the star, others will be found. Already, some scientists say that the signal is too complex to suggest the planet’s existence with certaintly. A complex signal could well mean more planets. Trouble is, finding a Mars or Earth-sized planet possibly orbiting in a plane at an angle to the line of observation, as part of a triple-star system, and orbiting at a fair distance–as in 1 AU or more–is going to be like the proverbial needle in the proverbial haystack.

Meanwhile, get your space ships ready.


Once more on the search for exoplanets, and Alpha Centauri

Excuse me for the absence of regular science posts, but my brain has turned to mush from writing fantasy. Isn’t it easy when you’re allowed to make everything up and as long as it makes sense, no one cares about accuracy? But, yes, I will return to where I left my characters in the Jupiter system, or in a space station orbiting the fictional gas giant Sarasvati, in the not-too-distant future.

This morning I came across this very interesting article on the Centauri Dreams website. By the way, Centauri Dreams, the website of the Tau Zero Foundation, is a very rich source for writers of realistic SF, especially in relation to planetary exploration and interstellar travel.

The article summarises results and speculation arising from new planets discovered by the European Southern Observatory’s HARPS spectrograph, which provides the most accurate Radial Velocity measurements we currently have (see an earlier post on how planets are discovered). Because of its increased sensitivity, HARPS can detect smaller planets. The smallest planet found at this point in time is a mere 1.5 times Earth’s mass. One of the, perhaps expected, outcomes of the spectrograph’s bevy of newly found low-mass planets (super-Earths or near Earth-mass) is that there are many of these smaller planets, a lot more than there are very large planets, and that the previous bias was merely a product of larger planets being easier to detect. The galaxy is swarming with smallish rocky planets. It is quite likely that some will be found inside the habitable zone.

We may already have found some of these planets. Much was made last year of the ‘discovery’ of Gliese 581g, supposedly in the habitable zone of an M class star. However, further analysis has so far failed to confirm the existence of this planet. But the star has two other planets which orbit at the edge of the habitable zone, and out of these, Gliese581d looks the most promising. The width of the habitable zone is not absolute, but varies with the planet’s albedo (basically, how much light it reflects) and composition and (if any) atmosphere composition (see another post on that here). So a newly discovered planet, HD85512b, at 3.6 Earth masses, may also fit the bill. It is a little close to its K class parent star, but could harbour liquid water on its surface if certain conditions of composition and atmosphere are met (see original paper by Kaltenegger et al. here).

Using the HARPS spectrograph, another group of researchers report on the search for planets orbiting sun-like stars within 40 light years from our solar system. (original paper by Pepe et al here). This work has resulted in a the discovery of a number of planets, again, most in the smaller size category. One of the main targets for the hunt is Alpha Centauri B, but there are some problems, one of the main ones being that it is part of a triple star system, and that any model the describes the wobble of the star that is caused by an orbiting planet must take into account that there are two other stars in the system, and as you could understand that is tricky business.

Image depicting an exoplanet system snarfed from NASA JPL

So… does Alpha Centauri have any planets?

A short follow-up on this post I made a while ago (note the tantalising difference in wording of the title).

Just today, I stumbled across a scientific paper, thanks to a tweet by @b0yle, that outlines the priority of stars to which to send an as-yet-hypothetical interstellar mission (link here; the paper is not as jargon-laden as a lot of other scientific publications, so go read if this subject interests you).

Short answer: we still don’t know. And with our current techniques, we can’t be certain. We’re getting very close, though.

Some figures: there are 56 stars within 15 light years of us (give or take a few that may or may not be within this range and possibly very dim ones we haven’t yet discovered).

Most of those are M-class stars, ‘red dwarfs’. There are also 2 G-class stars, like the sun (Alpha Centauri A at 4.4 ly and Tau Ceti, at 11.9 ly). According to current evidence, up to 30% of all stars may have planets. At this point in time, most discovered planets have been very large.

The fact that we haven’t detected any in the Alpha Centauri system doesn’t mean that there are none.

Planet size is expressed in number of times the planet is heavier than Jupiter. It is still fairly hard to detect anything significantly smaller than that. Yet evidence suggests that smaller planets could be more common than larger ones.

Alpha Centauri is a star system of three stars (one K-class, one G-class–like the sun–and one M class). What is the statistical chance that none of them will have planets?

Exoplanets: could Alpha Centauri have any?

My fiction recently gave me cause to examine interstellar travel. Many writers tend to shy away from the reality that we’re a long way from anywhere. It’s too hard, too intimidating, too depressing. I, too, have done the wormhole thing, you know, where your characters can zip between worlds, but fun as it is, and I won’t stop writing space opera, it always feels like cheating to me. In my next project, I wanted a more realistic approach. How realistic? I haven’t answered that yet, but I’ve started out by looking at the facts.

Our nearest stellar neighbour is Alpha Centauri (it’s actually a multiple star, but a bit more about that later), the brightest star in the constellation Centaur, the forth-brightest star in the sky (less bright than Sirius, which is more than twice the distance), and mostly visible in the southern hemisphere. At a distance of a mere 4.22 light years, and a theoretically achievable travel speed at 10% of the speed of light, allowing for speeding up and slowing down, it would take roughly 50 years to get there. Wow. I am waiting on the arrival in my real-life mail box of some material about how to get there, and will write more about that later, but let’s assume, for the sake of the argument, we could send a ship there within a human lifetime.

Question is: why would you? What is there? If Alpha Centauri had planets the size of Earth, or Mars in the habitable zone, wouldn’t they already have been detected?

There is a long and a short answer to these questions. Let’s have the long answer first.

What methods do we have to detect planets?

Direct Imaging:
There is no denying that emotionally the best way to determine if something is there is to see it. Humans tend to be visual creatures and have a great ‘I’ll believe it when I see it’ instinct. Surprisingly, telescopes have seen some planets. There is some debate as to whether those objects are planets or brown dwarfs, but something is definitely orbiting those stars. To see a planet, the size of the planet and brightness of the star are going to matter. The reality is that in most cases our telescopes are nowhere near detailed enough, and the above examples are exceptions.

Radial Velocity method (also called Doppler Spectometry):
When a planet moves around a star, the star wobbles a tiny bit. Light waves behave in a manner similar to sound waves with an approaching and passing ambulance. If the star moves away from us, the waves become longer, the light more red; if the star moves towards us, the waves become shorter, more blue. This is called redshift or blueshift and can be picked up with very sensitive instruments. This method tends to detect planets whose weight ratio compares favourably with their stars – relatively large planets orbiting relatively small stars. It also doesn’t take into account any part of the star’s movement that is not towards or away from us. This is where astrometry helps. These two methods combined are the most common in planet-hunting.

Transit Photometry method:
If you’re lucky enough, the plane of the orbit of a planet around a star is exactly the same as our field of vision. In other words, we view the solar system edge-on. In that case, there will be times that the light from the star dims, because the planet passes in front. Because you need more than one pass of the planet, this method obviously favours large planets with a short orbital period.

When a star moves in front of another, the closest star distorts the light of the more distant one, making it appear 1000 times brighter than normal. This effect usually lasts a few weeks, until both stars move on. When, during this time, a planet happens to pass in between, it adds to the effect. You obviously have to be pretty lucky for this to occur at a time you happen to be watching.

This method relies on extremely accurate measurements on how a star moves in the sky. The Radial Velocity method discussed above works best when a solar system is viewed edge-on; astrometry works best when the solar system is viewed face-on. This method has enormous potential, and astronomers predict that we will be able to detect Earth-sized planets. But not yet. At the moment, distortions from the Earth’s atmosphere hamper measurements, but projects like the European Gaia mission, scheduled for 2011 will change that.

Now about Alpha Centauri:
The trouble with Alpha Centauri is that it’s not one star, but three, denoted by the capital letters A, B and C (the small letters are used for any planets discovered). Alpha Centauri A is also called Rigil Kentaurus, the largest and brightest of the group, and is a G star similar to the Sun. Alpha Centauri B is an orange K type star (see here for a cheat sheet on star types). A and B form a close binary, with a distance between them of 23AU (1 AU is the distance from Earth to the Sun), about as far apart as the Sun and Uranus. This is considered to be a piddle of a distance. The star Alpha Centauri C is also called Proxima Centauri and is actually the closest star to us. It’s a red dwarf, and 14,000AU from A. You can imagine that it is hard for planets to form in a system where they are influenced by the gravity from more than one star. However, it is not impossible. Various references that are listed at the bottom of this Wikipedia page suggest that A could possibly have planets in a zone not further than 2.5AU from the Sun-like star. In our solar system, the habitable zone is considered to be roughly between 0.9 to 1.7AU (this includes Mars, but not Venus). Additionally, planets have been detected in close binaries.

The current count of extra-solar planets numbers 418, listed here (be sure to visit the rest of the site, because it is excellent). It’s an entertaining list, which includes a small visual of what the system looks like.

When you go through the list in detail, you will notice a few things. In the first place, most planets are massive. Masses are given in Jupiter masses, so a planetary mass of 1 is equal to Jupiter, and a solar mass of 1 is equal to the Sun. Many, but not all, orbit close to their parent star, and lastly, their parent star is relatively small. Strangely enough, distance from us doesn’t seem to be a factor. The furthest planet detected I could see in the list was more than 20,000ly away. Distance between the star and the planet matters only in that it increases the time necessary to be looking at the planet to reliably detect it. The planet Gliese 581g, the source of the kerfuffle earlier this year, was the result of eleven years’ study of the parent star. Astronomy is not a short-time career. We come back to the planet mass (compared to Jupiter)/star mass (compared to the Sun) ratio. For Earth, the ratio is 0.003. For Mars, 0.0003. For Uranus, it’s 0.04. The smallest ratio I could find in the table was 0.016 for a planet called CoRot 7b. Gliese 581g, which is not in the table, has a ratio of 0.029. It has three times the mass of Earth, and its star is 31% of the Sun’s mass. I expect that new methods will bring this ratio down.

That was the long answer.

The short answer is that there might be planets, but because we haven’t seen them, they are unlikely to be Jupiter-sized, and we don’t yet seem to have the technology to detect planets much smaller than Uranus ratio, not even for our closest stars.