≡ Menu

Andrew Gould on the Science of Microlensing

A recent query about how astronomers work out the mass and radius of planets found through microlensing — such as the so-called ‘super-Earth’ recently discovered 9000 light years from our Sun — prompted Centauri Dreams to query one of the principals in that discovery. Andrew Gould, leader of the MicroFUN collaboration and professor of astronomy at Ohio State University, was kind enough to clarify how this fascinating science proceeds. Herewith his response:

We obtain a planet-star mass ratio from a fit to the light curve. This parameter (q) is a standard output from fits to microlensing curves generated by binary lenses (two point masses, e.g., star and planet). Now, for low-magnification microlensing events, such as OGLE-2005-BLG-390Lb (which was announced by the PLANET team in January), it is generally possible to estimate the mass just by looking at the light curve. And in those cases it can be explained easily to the non-expert (although PLANET did not do this in their article). However, in high-magnification events like OGLE-2005-BLG-169Lb, we get a mass ratio out of our computer codes, but there is no known simple way to make an estimate directly by looking at the light curve.

We estimate the star mass M based on some characteristics of the event plus a Galactic model. Thus we derive a planet mass, m = qM. We will be able to do better on the star mass when we get images of the host star as it moves away from the source (assuming we can get Hubble time for this).

We have no direct information about the radius. In the article, we put forward two ideas about the nature of the planet. One is that it is like Neptune, with a rock and ice core surrounded by a thick padding of gas. Then it would presumably have a radius like Uranus and Neptune, about 4 earth-radii. However, because it is in the region of its solar system that is just out past the “snow line” (where Jupiter and Saturn are in our solar system), we prefer the idea that it is a failed core of a Jupiter-like planet. Then it would be all rock and ice, probably roughly 2 gm/cm³ uncompressed (and so about 2.5 gm/cm³) compressed. This would give it a radius of about 3 earth-radii.

On the matter of terrestrial-size worlds, Gould added that he was hopeful that such planets could be detected within a few years, but noted that they would still be 10,000 to 20,000 light years away. Thus the beauty, and frustration, of microlensing. It seems the most likely way we’ll find our first Earth-mass planets, but the physics involved all but guarantees they’ll be distant indeed.

Comments on this entry are closed.

  • qraal March 19, 2006, 7:06

    Hi Paul

    Bit of a scoop getting Gould’s comment, though I doubt the density he gives for a half-ice/half-terrestrial planet. Basically it’s an ocean planet since on both the ice will be in high pressure phases for much of the planet’s radius. A six Earth mass ocean planet has a radius of 2 Earths according to the modelling by researchers for the ESO’s COROT project, so 13 masses won’t be much bigger. I get a density of about 4.5. But considering the huge uncertainty in these things who knows?

    I’ve been reading old “Analog” magazines, back when they were still “Astounding” – there’s some quite passable stories set on the old cold Jupiter model. Astrophysicists had just publicised (c.1935) their latest thoughts about Jupiter being metallic hydrogen or hydrogen ice – most SF writers picked up quickly that it was mostly hydrogen with nasty chemicals like ammonia thrown into the mix. Just a few years before people were still portraying Jupiter as hot from primordial heat – the Red Spot was a volcano etc. Then it became hydrogen and very cold, the surface under immense pressure. A 1957 story by Poul Anderson, “Call Me Joe”, has the main character – and artificially breed Jovian humanoid – finding seams of metallic hydrogen. In another, James Blish’s “Bridge”, humans build an immense suspension bridge on the surface out of Ice VII.

    Of course now we know Jupiter is too hot. The planet formed quick enough to keep its heat of formation. But what if exoplanets can form slow? There’s a big cored Saturn mass out there – it’s too hot, but perhaps colder versions exist. Perhaps a slow formed, big-cored Jovian can be home to some seriously alien lifeforms – rendered familiar by a lot of old SF tales.


  • Administrator March 19, 2006, 21:52

    Talk about a tale that takes me back! “Call Me Joe” remains a favorite after all these years, and I still have a bunch of those old Astoundings around here both from that era and earlier. Fond memories, and wonderful to imagine how Poul Anderson would have reacted to all the new exoplanetary discoveries.

  • ljk April 5, 2007, 13:05

    An Extrasolar Planet Census with a Space-based Microlensing Survey

    Authors: D.P. Bennett, J. Anderson, J.-P. Beaulieu, I. Bond, E. Cheng, K. Cook, S. Friedman, B.S. Gaudi, A. Gould, J. Jenkins, R. Kimble, D. Lin, M. Rich, K. Sahu, D. Tenerelli, A. Udalski, P. Yock

    (Submitted on 3 Apr 2007)

    Abstract: A space-based gravitational microlensing exoplanet survey will provide a statistical census of exoplanets with masses down to 0.1 Earth-masses and orbital separations ranging from 0.5AU to infinity. This includes analogs to all the Solar System’s planets except for Mercury, as well as most types of planets predicted by planet formation theories. Such a survey will provide results on the frequency of planets around all types of stars except those with short lifetimes. Close-in planets with separations