Friday, 3 June 2011

Alien civilisations - less likely?

A couple of years ago I posted a review of a non-fiction book by Stephen Webb: Where is Everybody? Fifty Solutions to the Fermi Paradox . In this book, the author considers the Fermi Paradox; that given the number of stars in this galaxy alone and the length of time it has existed, there should be swarms of high-technology Extra-Terrestrial Civilisations (ETCs) around, so why haven't we detected any? He examines a wide range of possible explanations before concluding that our planet is uniquely fortunate and may host the only technological civilisation in the galaxy.

My own conclusion was slightly different from the author's, in that I speculated that given the age of this galaxy, with the average age of its stars being some two billion years older than our sun, there have probably been plenty of ETCs around but that it could be rare for more than one to be in existence at any one time, since they may not last all that long.

One of the unknowns, until very recently, was how many stars have planets - particularly planets like ours, rocky and in the CHZ (continuously habitable zone): which is to say, at the right distance from its sun for the billions of years needed for not just life (or our understanding of it) but advanced intelligence to evolve. The habitable zone is popularly known as the "Goldilocks" zone: not too hot and not too cold for liquid water to exist on its surface (i.e. average surface temperatures within the 0-100 degrees centigrade range). However, this gap in our knowledge is rapidly being filled by astronomers who, by using highly sensitive instruments and sophisticated data processing techniques, have discovered over 1,200 exoplanets orbiting nearly 1,000 stars, with the numbers steadily growing. What they have discovered so far has been summarised in a couple of recent issues of the New Scientist magazine (Astrobiology supplement by Caleb Scharf, 7th May; and No Place Like Home by Lee Billings, 14th May) and is discouraging to those keen to find ETCs.

First, I had better summarise the three different indirect methods by which exoplanets are detected (even the biggest of them around the closest stars are far too small to observe directly).

The first method used is the Doppler or radial velocity technique. This relies on the fact that planets do not, strictly speaking, orbit their stars. The planets and their stars orbit a common axis whose position is determined by their relative masses; if a star and planet were both of the same mass, the axis would be half-way between them. Generally, stars are vastly more massive than any of their planets so the common axis is within the star, but not in its centre. The star therefore wobbles slightly as the planet moves around it, and this can be detected. The speed of the wobble indicates the period of the planet's orbit and therefore its distance from the star; the size of the wobble indicates the relative mass of the planet. Obviously, if a star has several planets, each exercising its influence on it, then its pattern of wobbles can be very complex and require lots of number-crunching to resolve. This method favours the discovery of large planets orbiting very closely around their stars, as these create the biggest wobbles. This may mask the existence of smaller planets further out.

The second method is known as transit photometry, which is based on measuring the slight dip in a star's brightness as a planet passes in front of it. The degree of the dip indicates the planet's size, the time period involved indicates its speed and therefore distance from the star. This method also has its disadvantages. An obvious one is that the planet's plane of orbit has to be side-on to us, otherwise it wouldn't pass between its star and our planet, so any planets with different orbital planes will fail to be detected. It is also necessary for three transits to be observed to be certain that this is a genuine effect, which in the case of a planet the same distance from its star as ours means that it will take two to three years to confirm. Jupiter's orbit takes twelve years, so confirming the observation of a similar planet would take 24 to 36 years. So once again, bigger planets close to their stars are the easiest to detect.

The third method is called gravitational microlensing, which relies on the fact that massive objects bend the fabric of space. In practical terms, it means that a star exactly in between us and a far more distant star will act as a lens, focusing the light of the distant star. If the nearer star has planets, these can produce a subsidiary focusing effect which can be analysed to determine the planets' masses and orbital distances. However, the opportunities for such observations occur very rarely.

I find it amazing that no only can such miniscule observations result in confident estimates of the size and mass of planets orbiting distant stars but that the nature of the planets can also be deduced: whether they are rocky worlds or gas giants. Data from their stars also allows astronomers to deduce whether or not a particular planet is within the habitable zone.

At first only the largest planets were observed, but more recently (and especially with the use of the Kepler telescope launched into orbit in 2009) it has become possible in some cases to start building up a picture of entire solar systems, identifying the number, size and orbits of several planets orbiting the same star. The results are demolishing some long-held beliefs.

The theories of solar system formation which have developed over the centuries have of course all been based on a sample of one: ours. They tended to conclude that all of the planets will be more or less in the same orbital plane with close-to-spherical orbits, and that planets in close orbits will be small and rocky, with gas giants further out. All of these conclusions have turned out to be flawed.

What astronomers have observed so far might be summarised as follows: planets and panetary systems are the norm, but while some systems have an even more regular structure than ours, others can best be described as chaotic. Gas giants are found in close orbits, the most spectacular example being Upsilon Andromedae which has a planet 1.4 times the mass of Jupiter so close to the star that its orbital period is just 4.5 days! Furthermore, that same star has a super-massive gas giant, 14 times the mass of Jupiter, with its orbital plane at a 30 degree angle to the first one. And there is a third giant in that system, 10 times the mass of Jupiter, in an extremely elliptical orbit with a different orbital plane again. This kind of chaotic structure would have a huge effect on any smaller planets in the system, wildly disturbing their orbits and making it impossible for them to remain in the habitable zone for any length of time.

One possible consequence of such gravitational instability is that planets can end up being flung out of their solar systems altogether, presumably accounting for the recent discovery of many such homeless planets floating around our galaxy, only detectable via their gravitational lensing effect. In fact, a later New Scientist news item suggests that so many of these loose planets have now been discovered that they must be considerably more common than planets which are still orbiting stars.

This is really significant since in order for life to evolve to a human level of intelligence on any particular planet, that planet has to remain within the Goldilocks zone for billions of years. And that means above all that stability is required. Not only does the orbit of the planet have to be fairly circular and the star itself be stable, but other planets in that system have to be in stable, near-circular orbits in more or less the same orbital plane.

Of the 1,200 planets detected so far, only 366 are rocky and of Earth or super-Earth size (the initial requirement for supporting Life As We Know It). Of these, just six are in the habitable zone. While most seem to be in reasonably stable orbits at present, that does not mean that they have been, or will remain, in that zone for the length of time required to develop intelligent life.

Exploration continues and conclusions are sure to change as more data comes in, but initial indications are that our Earth's characteristics and history have been unusually favourable to the development of intelligent life. Which suggests that it is unlikely that other civilisations exist anywhere near us.

5 comments:

Chimeradave said...

Great article Tony. Depressing though, isn't it?

Anthony G Williams said...

It is, rather.

On the other hand, there are a heck of a lot of stars in this galaxy...

WCG said...

Interesting, Tony. But lest we get too pessimistic, I should point out that our initial sample is skewed, since the most easily detected planets are those which are least like Earth.

As you note, huge planets which are very close to their sun are the easiest to discover, yet that doesn't match up to our own solar system at all. But is that because we're unique, or just because those are the solar systems easiest to detect at a great distance?

We already suspected that Earth-like planets - or sapient life, at least - must be rare, so this doesn't change that. And I just think we should be wary of making any conclusions based on such a skewed sample.

It's not that I'm optimistic - certainly not about advanced alien life - but that I don't think we need be too pessimistic just yet, not based on this data.

(Note that I tend to agree with your conclusion on the Fermi Paradox, that intelligent species just don't last too long. Look at what we're doing to our planet.)

Anthony G Williams said...

Well, the Kepler telescope is allowing them to tease out smaller planets now - they are getting down to Earth-size.

Not that a planet has to be Earth-sized, of course. As long as it's rocky, with a comparable atmosphere and surface water, that'll do nicely...

WCG said...

Sure, Tony. I'm just saying that it's not a random selection of planets. It's skewed somewhat - how much, I don't know - by which planets have been easier to discover.

So we need to be cautious about what we conclude from it, at least so far.