Friday, 7 March 2014
Background: M-type dwarf stars have only recently been recognized as reasonable hosts for habitable planets (Tarter et al. 2007). In fact, because of their small masses and the close-in location of their habitable zones, the first habitable worlds are likely to be found around M dwarfs. For this reason, the Exoplanet Task Force recommends a short-term strategy concentrated on them (Lunine et al. 2008). Planets ranging in mass from gas giants to super-Earths (e.g., Endl et al. 2008) have been discovered around M dwarfs by both Doppler spectroscopy and gravitational microlensing. One hot super-Earth (GJ 436 b) has been seen to transit, producing a mean density estimate consistent with either an ice giant or a large rocky planet with a minor gaseous envelope (Adams et al. 2008).
There are only 21 known G stars and 45 K stars within 10 pc of the sun, but at least 236 M dwarfs (Henry et al. 1997). M dwarfs are thus a natural choice for astrometric planet searches, where closeness is the primary virtue. Such close stars are also likely to be the best targets for further investigation of habitable planets by future NASA telescopes designed to search for biomarkers in the spectra of the planets' atmospheres.
Research To Date: Co-I Alan Boss received NSF funds in 2004 to help build the Carnegie Astrometric Planet Search (CAPS) Camera (Figure 2), a specialized camera that yields astrometric accuracies of 0.25 millarcsec per epoch. CAPSCam was first mounted on Carnegie’s 2.5 m du Pont telescope at Las Campanas Observatory in March 2007, and has been used extensively since. Data reduction is now underway, and the first parallax has been obtained.
Proposed Work: Boss proposes to lead a ground-based astrometric planet detection program (CAPS), with participation by Co-I Weinberger. CAPS will follow at least 100 nearby (within 10 pc) low mass stars, principally late M, L, and T dwarfs, for 10 years or more. It will detect planets with orbital periods long enough to permit the existence of habitable, Earth-like planets on shorter-period orbits. These stars are generally too faint and too red to be searched by Doppler spectroscopy, and their small masses yield correspondingly larger astrometric signals for a given mass planet.
The demonstrated astrometric accuracy of 0.25 milliarcsec per epoch will permit detections with signal to noise ratios greater than 4 for long-period companions with masses as low as 0.2 Jupiter masses. A 1 Jupiter-mass planet orbiting at 1 AU from a late M dwarf located 10 pc away would be detected by our survey within about three years. We thus expect our planet search to begin to bear fruit within the first few years of CAPS observations.