Ground-based Astrometric Detection of Exoplanets with CAPSCam: Current Status
Abstract
We began using the Carnegie Astrometric Planet Search Camera (CAPSCam) in 2007 on the 2.5-m du Pont telescope at the Las Campanas Observatory, Chile. Our targets are 100 nearby (within 20 pc) late M, L, and T dwarfs. Nearby, low mass stars maximize the possibility of discovering exoplanets by ground-based astrometry. Being intrinsically more suited to detecting long-period exoplanets than Doppler spectroscopy or transits, astrometry has the potential to join microlensing and direct imaging in determining long-period exoplanet demographics, a key constraint on theoretical models of gas giant planet formation. With an observing allocation of about five nights every two months over the last twelve years, CAPSCam has the potential to make the first astrometric discovery of a gas giant exoplanet orbiting a single star. CAPSCam was anticipated to have an astrometric accuracy of 0.25 miliarcsecond (mas), allowing the detection of a Jupiter-mass exoplanet orbiting at 5 AU around a 0.4-solar-mass star with a signal-to-noise ratio of four. With 18 epochs, CAPSCam determined the parallax of GJ 317 to within 0.15 mas (Anglada-Escude et al. 2012). Our determination of the parallax of the TRAPPIST-1 star of 80.09 +/- 1.17 mas (Boss et al. 2017) agrees with the 80.45 +/- 0.12 mas of Gaia DR2 and placed limits on the presence of any long-period gas giants in that system. One of us (TLA) has developed a new data reduction pipeline. The primary goal is to correct for atmospheric distortion and optical distortion of the telescope and instrument. A model using polynomials for the optical distortions and atmospheric differential chromatic refraction has been developed using the combined Gaia DR2 (for positions) and NOMAD (for colors) catalogs. The data processed with the new pipeline can then be fit with a 5-parameter astrometric model (positions, proper motions, and parallax) or a 12-parameter model that includes an orbital solution (5 astrometric parameters, 6 orbital elements, and a mass ratio). By comparing the probabilities of these two models for all of our targets, it is implied that roughly half of our targets show no evidence for companions.
- Publication:
-
American Astronomical Society Meeting Abstracts #233
- Pub Date:
- January 2019
- Bibcode:
- 2019AAS...23340808B