Discussion and news about the modern effort to understand the nature of life on Earth, finding planets around other stars, and the search for life elsewhere in the universe

Friday, October 1, 2010

A distant cousin

It seems worth following up on the previous post with a little more on the planet GL 581g. With media attention temporarily swirling around this announcement and numerous opinions being offered it can be a little difficult to locate the core truths. Now that Vogt et al.'s actual scientific report is live we can see the basis for their public commentary. It's a very nice piece of work. It's also a remarkably, and refreshingly, chatty report - not something that scientific papers are particularly known for. The incredibly tricky and slippery nature of extracting planet detections from radial velocity (or Doppler 'wobble') data on the star is nicely laid out. Boy did they have to work hard on this. 11 years of data, including some from a 2nd instrument. Although there were over 200 discrete measurements of the star's motion these were, of course, spaced across a decade in time. This kind of sparse sampling - a necessary evil given the nature of telescope time allocation, weather, and competition - presents many challenges when you're looking for numerous overlaid time-varying signals.

Nonetheless, they pulled out the best solutions they could, checking against gravitational simulations of the system to make sure these answers resulted in a real, stable, system of planets and observing the star for signs of luminosity variation - sunspots and the like - that could dupe us into seeing things. All seems good. It is interesting too that they present an age for the system of 4.3 billion years, based on spectral analysis of the star - rather younger than the 7-12 billion years previous measurements had given, and the basis of some of my previous comments. Dating stars is tricky, so I'll pause on any extrapolations from that.

All of which brings us back to asking whether this really is Earth 2.0 as so many headlines have been suggesting. It's not, is the simple answer. Two big tick boxes get filled in - close to Earth mass and in the 'habitable' zone of a normal star. This alone does not make for tropical islands, lush forests, or highway systems.  The real reason for being excited about this planet is that despite being extraordinarily alien, it nonetheless exhibits characteristics that place it firmly as a distant cousin. Imagine you were a hugely pampered but not overly prejudiced western explorer in the 1500's, and that no one from your neck of the woods had ever set foot beyond Lisbon. Setting off across the oceans you arrive at Papua New Guinea. You would immediately recognize other humans, however they would be engaged in complex and utterly alien work and social customs, like nothing you'd seen or experienced before. Their completely different lifestyle would confound you - but it would be obvious that you shared essential biology and characteristics. I think that's a fair analogy here. GL 581g and Earth are distant relatives, products of a universal set of mechanisms that build planets. The key is that GL 581g is the least distant relative we've come across so far.


CoffeeCupContrails said...

Of course. This could just as well be one big mercury/mars style rocky mass just floating there in circles. Any scientific possibility - or physical limits- on being able to detect chemical signatures from these earth sized planets?

Caleb Scharf said...

It's going to be tough, but future instruments might be able to do it. Lee Billings has written a great piece at Seed Magazine that includes some ruminations on precisely this point.

lbillings said...

Thanks for the shout-out, Caleb!!! I really like your extension of the "distant relative" analogy here. I was struggling with how to properly extend this same analogy while writing the piece yesterday, and ended up avoiding it. I'm glad I did -- I think your version is tough to beat.



Anastasia said...

Interesting article and discovery - too bad it doesn't seem like any more answers about its habitability will come soon - if it is tidally locked, Ibillings mentioned in his article that one side will permanently develop an ice cap - is there any way to determine the probablity of this computationally or is there just not enough observational data?

Anastasia said...
This comment has been removed by the author.
Caleb Scharf said...

Not enough data at the moment. We have climate models that can give us an idea of what happens on a locked planet - but we need as input things like albedo, water content, atmospheric composition/thickness, ocean/continent arrangement etc etc. We might be able to place some broad constraints on the above - i.e. habitability will require X to be between A and B - but it's a pretty messy phase space, and computationally it's demanding to run enough sims to cover this space.