Next we review microlensing surveys for planets and then direct imaging surveys.

Scott Gaudi up first on microlensing.
Interesting statistics on preponderance of solar like planetary systems.
Hints of free floating planets seen.

Paul Kalas on direct detections.
Also Graham and Kadsin.

Plus bonus “structure of giant planets” review at the end.

Rapid fire overview and list of surveys.
If you want a summary of what it is and who is doing it, read the opening slides.
Get statistics of planets in otherwise inaccessible mass-orbit parameter space, but no or limited followup for most sources.

Nice summary of the double planet system around 0.5 solar mass star from OGLE in 2008.

Budget and manpower limited. Good pro/am collaboration.
Should be significant improvement with Los Cumbres and other large area transient surveys coming online.
Followups hard as more data gets in – this is a common theme – followups are expensive and backlog is piling up. For everything interesting.

6 planet events out of 13 high magnification events.
Completeness well characterised. Typical stellar mass ~ 1/2 solar mass.

Infer very common to find giant planets/neptunes out beyond the snow line.
Weak inference, that is probably correct, that many or most outer giants do not migrate.

Tens of percent of jupiter/saturn/neptune mass planets at several AU around sub-solar stars.
Small fraction of that mass-ratio migrates.
Infer ~ 15% of stars have solar system analog planetary configuration.
ie near circular giants outside snow line and terrestrial planets further in – modulo caveats on selection, statistics etc.

So NOT rare, but not ubiquitous either, which sounds about right.

7 new microlensing events in ’09-’10 – pending publication.
Current detection rate ~ 4 per year. Sensitivity adequate to reach earth mass iff fairly lucky in configuration and timing.

Preliminary high cadence observations from MOA suggest anomalous number of short duration lensing events – could be very wide planets, or free floaters. (Kamiya et al in prep).

If free-floating planets there are ~ 3 per star in the inner galaxy.
That is not implausible.


Direct Detection: ooh, we are promised new unpublished results…

9 current imaging candidates – masses are up there, many or most could technically be brown dwarfs. All of course are wide separation – 10s to 100s of AU.
Which enhances argument for them being brown dwarfs, depending on what you believe about planet formation etc.

Nice set of slides showing the candidates.

GJ578 – aka HD 182488 – G star at 15 pc, Z ~ +0.2 – Thalmann et al 2009
10-20 M_Jup at about 30 AU projected separation – expect a ~ 55 AU with e ~ 0.7 +/- 0.2
common motion confirmed with 2nd epoch, hint of a lower mass inner companion.
Mass uncertainty depends on age, could be young as Z is high.

β Pic – something seen in 2003 VLT archive – consistent with ~ 10 M_Jup and 8 AU separation
nothing seen in 2009 followup. Could be planet has moved in projection – or it was background and not a companion.

HR 8799 – the triple.
I still think the darned thing is a blue straggler, in which case the planet estimates are completely off.
Planets could then be very old, as progenitor could have been several Gyrs old, or they could be younger than estimated, if they formed in an excretion disk during merger.
Latter would be more consistent with the dynamics.

Hm, there is new stuff on HR8799
arXiv.1003.5121 claims the age used in the planet papers is in fact completely wrong and it is over a Gyr
Here is the primary paper – arXiv.1003.3340 Moya et al 2010 MNRAS in press

I’m personally strongly inclined to HR 8799 being a “type I” blue straggler, formed from solar mass star, which was originally old, with low mass companion that got engulfed in a common enevelope phase and the planet formed in a debris disk 50-100 Myrs ago – time since mass transfer phase ought to be ~< half of estimated main sequence age. PS: hmm, Phil points out that the angular momentum in such a merger may not be enough to make the planets - (someone) needs to look at that properly. Something like a Curry & Hansen model for an excretion disk. Fomalhaut: planet predicted by Alice Quillen seen by Kalas et al. Spectacular detection from ACS. Planet is too blue and too bright given mass constraints. Consistent with reflected Fomalhaut spectrum - requires much larger area, ie 20-35 planetary radius ring system. Implies planetary mass could be less than maximum constraint, could be sub-Jovian. WFC3 3rd epoch imaging in Nov 2009. This is news. ( ACS HRC coronograph is broken) 1.1 micron observing - diffraction structure from camera right over where planet should be, trying STIS instead, have first image, going to use our old trick of rolling this summer to self-subtract. Graham: did a good review on high contrast imaging from the ground. Nice figure from Guyon '09 showing the different techniques for getting high contrast imaging being explored and where we are. Kasdin: high contrast imaging from space Terrestrial Planet Finder (c) concepts... Detailed discussion of internal coronographs and external occulters. Damn: a Bell Labs guy did the optimal apodizing pupul solution and pointed out its use for finding planets over 40 years ago. Bloomin' phone company... I personally am fond of the "starshade" concept as per New Worlds Explorer mission concept.
Not only is it just tres cool engineering, but you also get to use the telescope for secondary science while the starshade moves to the next target.
Y’know, like stars and galaxies and shit.

Nice slide on the tolerances for shape, distortion and alignment for a starshade,
good summary

Mission concepts to remember:
THEIA – Kasdin and Spergel – 4 m telescope + occulter – sounds interesting
JWST + occulter – Kasdin discusses why he thinks this won’t work well, he makes a good case
O3 – 1-1.5 m telescope + occulter – no spectra, detects super-earths

PPS: Stevenson and Guillot wrap up the day with reviews of giant planet structure.

We don’t know the equation of state for hydrogen… that is a bit embarrassing.
Must remember the Argon Problem!

Why can’t we figure out Puffy Planets?

Marginal anti-correlation between metallicity and radial bloating.
Also maybe a Teff correlation – ie semi-major axis anti-correlation

At least CoRoT-9b is Just Right.


  1. #1 Birger Johansson
    March 30, 2010

    Microlensing: “Current detection rate ~ 4 per year. Sensitivity adequate to reach earth mass if fairly lucky in configuration and timing.”
    Earth-based instruments are likely to catch only the edge of any lensing event.
    To get information from closer to the line of maximum lensing, we should have dedicated space probes scattered across the solar system, with cameras capable of astrometry down to magnitude 18. And we could let appropriate instruments ride piggyback on future planetary probes. Compared to the cost for Hubble, it would be a bargain.

  2. #2 andy
    March 30, 2010

    Out of interest, do you reckon such a merger scenario for HR 8799 can explain the Lambda Boötis abundances?

  3. #3 Steinn Sigurdsson
    March 30, 2010

    @Birger – there is a lot of interest in a space based microlensing mission, or a joint mission with ulensing capabilities.
    Piggybacking probes is hard, mostly because of control issues – the camera would have to be able to point, and therefore either independently pointable, or able to seize the whole spacecraft – this does not make primary science people happy

    @andy – yes, I think such a scenario is just what you want for lambda Boo, in particular if the astroseismology results of Moya et al hold up and the lambda Boo abdundance peculiarities are actually bulk composition and not just surface anomalies
    in fact then you really need something like a Blue Straggler scenario

    It is interesting how the need to understand the primary host of exoplanets is now driving stellar astronomy at a sharply increased pace – partly just better date coming along for the ride

  4. #4 Lab Lemming
    April 4, 2010

    Were the lensing people able to offer stats on what proportion of stars have no planets at all, based on geometric expectations vs. discovery rate?

New comments have been temporarily disabled. Please check back soon.