This award will support an observational program to analyze high-contrast, high spatial resolution images of young stars with circumstellar disks. The principal goals will be to find planets directly or to look for morphological structures in the disks caused by the presence of planets. Much of the work will be to analyze the results of an extensive coronographic survey, but there will also be new observations.
The results of the analysis will explore the processes involved in the birth, early evolution, and architecture of exoplanets located in the outer regions of both protoplanetary and debris disks. Members of this research team also are engaged in a program which involves beginning students in research, especially students from underrepresented groups, at the University of Washington.
In this collaborative NSF program, obtained high-contrast imagery of circumstellar disks to search for indirect signatures of giant planets within the disks and to image the planets directly with a goal of increasing our understanding of when, where in the disk, and with what frequency giant planets are formed. Grady has focussed on disks in the 1-10 million year range where available sub-millimeter continuum observations sampling larger dust grains in the disk indicated there are developing cavities or wide gaps in the disk, extending up to 70 times the earth-sun (au) distance from the star. This latter class of object, or transitional disks, can also be identified by missing warm thermal emission in the 5-25 micron range, and from our survey, in the presence of striking structural features in near infrared (IR) images of the disks in scattered light. At the start of the study transitional disks were known for young solar or sub-solar pre-main sequence stars (T Tauri stars), but were expected to also be found in association with slightly more massive stars. In addition to imaging transitional disks around T Tauri stars, we identified transitional disks around 1.5 to 3 solar mass pre-main sequence stars (Herbig Be, Ae, and Fe stars) as those with Meeus Group I IR spectral energy distributions. We find that >92% of these objects have resolved disks detected in scattered light at 1.6 microns. The T Tauri transitional disks exhibit a range of morphologies in the near-IR including systems with distinct cavities or gaps, smooth, filled disks, and those with a break in their radial surface brightness distributions. The intermediate-mass star disks add additional phenomena such as spiral arms, asymmetric shadowing of the outer disk, and offsets of the center of the outer disk relative to the star. For these stars we can exclude the structure being due to grain growth and settling in the cavity region, since the large dust grains are located at the inner edge of the outer disk and not within the gap/cavity region, and can also exclude clearing processes tied to the stellar radiation field. By process of elimination, this leaves dynamical clearing by 1 or more companions. We exclude stellar-mass companions in the disks associated with the A and B spectral type stars since such bodies would produce more hard (E>1keV) x-rays than are observed using either the Chandra or XMM missions. Some of the structural features in the disks can be produced either by instabilities in the disk or can be excited by companions. There are disks in our sample with more spiral arms than known instabilities, suggesting that at least some of the arms are associated with companions. In the latter case, spiral arms associated with one disk can be produced by ~Saturn-mass bodies. Direct detection of young giant planets embedded in their disks and located tens of au from their star has proven more challenging, and appears to be most successful at wavelengths longward of where the instrument we used (the High Contrast Imager with Adaptive Optics at the Subaru 8.2m telescope) operates. We are currently proposing to surveying transitional disks at 3.8 microns (L' band) beginning in 2015. We have also searched for wide planets in older disks, and have detected a few such objects, preferentially in younger (t<50 million years) disks. The majority of these planets have cloudy atmospheres, with one, cooler and older object, GJ 504b, resembling Jupiter. In addition to indicating that giant planets at r>30 au are comparatively rare, these data suggest that young giant planets a) have spectra which differ from brown dwarfs at similar ages, and b) may be shrouded in disk material in their earliest stages. These features may complicate understanding how the luminosity of these objects relates to their fundamental properties such as mass, how it evolves with time, and to understanding their atmospheres. By combining our measurements with data from other efforts, we find that the portion of circumstellar disks which can host young, giant planet candidates can be significantly larger than seen in the Solar System. Age estimates for transitional disks are uncertain by a factor of 2 (or up to +/-5 million years), but the available data, including the fact that after ~5 million years, all of the intermediate mass PMS stars with disks are transitional disks, suggests that the time needed to produce giant planets and to have them modify their disk sufficiently that it can be remotely detected is a few million years. This is consistent with age estimates for Saturn. This study has produced numerous papers, a refereed book chapter, an article in Sky and Telescope (August 2012) , and a US PhD in fall 2014.
