Radio Interferometric Planet Search


Geoffrey C. Bower, Alberto Bolatto, Eric Ford, Paul Kalas


Anna Treaster, Vince Viscomi, Adam Fries, Phoebe Sanderbeck

RIPL is an astrometric search for planets around nearby low mass stars that uses the Very Long Baseline Array and the 100-meter Green Bank Telescope. 
Our ultimate goal is to learn whether or not M dwarfs, the largest stellar constituent of the Galaxy, are surrounded by planetary systems as frequently as FGK stars and how the planet mass-period relation varies with stellar type. The population of gas giants at a few AU around low mass stars is an important discriminant between planet formation models.

RIPL has a number of unique qualities:
  • Opportunity to discover planets around nearby active M dwarfs at large radii;
  • Ability to fully characterize orbits of detected planets, without degeneracies in mass, inclination, and longitude of ascending node;
  • Sensitivity to long-period planets with sub-Jovian masses;
  • Complementary with existing planet searching techniques:  most targets cannot be explored through other methods;
  • Ability to follow-up detected planets with imaging and spectroscopy; and,
  • Absolute astrometric positions within the radio reference frame for stars and planets.
The quality and uniqueness of VLBA+GBT astrometry for planet searches are the result of two factors:
  • High precision of VLBA astrometry: The VLBA can routinely ~0.1 mas accuracy through relative astrometry.  The astrometric signature of an M dwarf with an orbiting Jupiter at 1 AU at a distance of 5 pc is 1 mas: easily detectable.
  • Active stars are difficult to study in optical programs: Our target stars are active M dwarfs, which have radio fluxes on the order of 1 mJy.  These radio stars are difficult to study through optical radial velocity techniques because they are faint and because the activity in these stars distorts line profiles, reducing the accuracy of radial velocity measurements.  RIPL is complementary to other approaches for planet detection and characterization.
planet astrometric sensitivity
ripl sensitivity
Sensitivity of different methods in planet mass and semi-major axis space for radio astrometric surveys and other methods.
``Exp. VLBA'' refers to the upgraded VLBA with 16 Gb/sec recording rates. The semi-major axis at the minimum in the astrometric search curves is determined by the search duration, which is 3 years for RIPL and the Exp. VLBA campaign.

Observing and Technical Details

RIPL is a 3-year survey of 29 M dwarf stars, consisting of 4 epochs per star per year.  Observations will be performed with the VLBA and the GBT, observing as a single array.  Each epoch will have a duration of 4 hours;  approximately half of the time will be spent on the target star.  The remainder of the time will be spent on phase-referencing and delay-correction calibrators that are necessary to achieve astrometric accuracy.  Data will be recorded at the maximum rate of 512 Mb/s.  We will achieve ~25 microJy rms sensitivity in the image for each epoch.  The observing wavelength is 3.6 cm.

Astrometric accuracy is set by statistical errors and variable delay caused by changing water vapor content in the troposphere.  Statistical error in astrometry is limited by the synthesized beam size, which is approximately 1 milliarcsecond, and the signal to noise ratio of the detection.  For a star with a flux density of 1 mJy, then the statistical contribution to the error will be 25 microarcsec.  Tropospheric fluctuations are corrected by fast-switching between nearby calibrators and the target star.  These errors are typically less than 100 microarcsec but depend on proximity of the calibrators and weather conditions.

Variability in the stellar position due to activity has been shown to be less than 200 microarcsec for a small sample of stars that we have already observed with the VLBA.


RIPL is a large project making use of NRAO facilities.  The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.  The RIPL project is supported by a grant from the NASA Origins program.