Starlight Measurements to Size Up a Planet

Overcoat writes "NASA scientists have used a technique called 'astrometry' to determine the size of a planet orbiting Gliese 876, a star 15 light years away from our solar system. By measuring tiny changes in the 'tilt' of light emitted from the star, changes which were caused by the force exterted by the orbiting planet known as Gliese 876b, the scientists were able to determine that the planet is the size of a golfball. Just kidding: the planet's a whopper, coming in at between 1.89 and 2.4 times the size of Jupiter. This marks the first time astrometry, usually used to measure the distance between stars, has been used to measure a planet."

Comments

More information here

[Simon Field]

Some other astrometry uses of the Fine Guidance Sensors can be found here: HST Astrometry Science Team.

I was looking for something a little less lame, something that didn't talk about the "tilt" of the light. Are they measuring polarization? Refraction? Diffraction?

Does anyone have a pointer to the article that UPI has so badly dumbed-down?

What's going on

[Jason T. Wright]

Most extrasolar planets are found by the precision radial velocity technique. The orbiting planet induces a reflex motion in the star (like in track & field when the hammer thrower leans back and revolves about the center of mass a little while the hammer moves a lot). We can detect this reflex motion as a change in the star's radial velocity. These velocities have magnitudes of hundreds or even tens of meters per second.

A limitation of this technique is that if a planet orbits its star in the plane of the sky, there will be no radial component to the star's reflex velocity, so we won't detect it. Further, unless the planet orbits with an inclination such that it passes nearly in front of the star, we will measure only a fraction of the total reflex motion.

This means that when we detect a planet, we can only put lower limits on the mass of the planet, since the signal could be from a massive planet in a nearly face-on orbit, or a tiny planet in an edge-on orbit. This ambiguity is proportional to the sine of the inclination (the "tilt"), so what we measure to be the mass of the planet is actually M*sin(i), where M is the true mass of the planet.

What these folks have done is use an instrument on HST to make extremely accurate measurements of the position on the sky of a star known to have planets, and used these measurements to measue the path of the star in the plane of the sky as it wobbles under the influence of the orbiting planet. This measures the missing tangential component of the reflex velocity, resolving the sin(i) ambiguity, and determining M itself. This is only the second time anyone has precisely determined the inclination of one of these planets.