Uncertain Principles

Some interesting astrophysics news this week, from Nature: scientists have used "microlensing" to discover a extrasolar planet only five times Earth's mass:

Planet OGLE-2005-BLG-390Lb looks much more like home. It lies about 390 million kilometres from its star: if it were inside our Solar System, the planet would sit between Mars and Jupiter.

It takes ten years for the planet to orbit its parent star, a common-or-garden red dwarf that lies about 28,000 light years from Earth, close to the centre of our Galaxy.

Of course, it's not quite time to start buying tickets for the colony ships: at that orbit, around a red dwarf star, the surface temperature is a bit below the boiling point of liquid nitrogen.

The important thing about this is not so much the fact of the discovery, but the technique that was used. Previous extrasolar planet detections have used the Doppler effect to measure a slight change in the star's velocity as it's tugged back and forth by an orbiting planet. As you might expect, this is most effective at detecting very heavy planets in fairly fast orbits, and that's mostly what they've found.

The new detection used a technique based on the fact that gravity can bend light. The researchers working on the project monitored a large number of stars, waiting for something small and heavy to pass between us and them. A heavy object in the path that light takes from a distant star can act like a lens, bending some rays that otherwise would've missed the Earth into our line of sight, and increasing the brightness of the star.

The lensing effect can be used to detect planets bacause it's a pretty short-range effect, and things orbiting around some distance from the star will be magnified at different times, sort of like the effect you get as you slide a lens across a line of text, with each letter being magnified in turn. A star with no planets should show a smooth, symmetric profile, getting brighter then dimmer as the lensing object moves across it, while a star with an orbiting planet should show a characteristic "bump" on one side or another, as light from the planet gets bent into view. What they're reporting in Nature is the observation of such a "bump."

The "microlensing" technique has the potential to be much more sensitive than the Doppler method-- there's a nice graph in the Nature news story showing the sensitivity comparison. There isn't the same sort of size limit that you get with the Doppler detection, where the planet has to be relatively big to move the star enough to detect (there's probably some limit, but the news story doesn't go into that). The down side is, it's kind of a crapshoot-- you have to look at a bunch of stars (millions, according to the New York Times story), and hope that something heavy passes in the path while you're looking. There's an active research effort underway to do this, though (I've heard talks mentioning the idea for a couple of years now), and a fair number of heavy things wandering around the galaxy, so expect more such discoveries to be announced in the future.

(Disclaimers: I am not an astronomer, so this is strictly a layman-level description of what little I retain from hearing a few astronomers discuss gravitational lensing. Clarifications from people who actually know stuff are, of course, welcome in comments. I'm also not usually this quick to jump on a news story, but like Razib, I was really into astronomy as a kid, and what SF fan wouldn't be excited by hearing about distant planets?)