The last decade has witnessed a revolution in astronomy with the discovery and characterization of substellar objects, brown dwarfs and extrasolar planets. The next 10 years promise to be even more rewarding. Observational capabilities are now on the horizon for direct detection of these objects and hence for determining the diversity of planetary systems around other stars.
The Keck Telescope has played a major role in these discoveries, primarily through radial velocity surveys. While these studies have been very successful at finding extrasolar planets, such discoveries are inherently limited since they measure only the orbital properties and lower limits on the masses. Direct imaging would allow us to measure colors, luminosities and spectra, thereby probing temperatures and compositions. At relatively young ages (<~1 Gyr), Jupiter-mass planets continue to radiate their internal heat at near-IR wavelengths far above their expected blackbody emission, and thus are amenable to direct imaging.
By virtue of its unparalleled 10-meter primary mirror, Keck adaptive optics imaging can produce the highest angular resolution images of any telescope in existence. The next generation of discoveries require with AO systems that produce much higher contrast (the ability to detect faint
objects next to bright ones), greater PSF stability, and broader wavelength coverage. Both the Gemini and VLT observatories are developing ``extreme AO systems'' (ExAO) to achieve very high contrast images for direct imaging of planets around nearby, young solar-type stars. While very powerful, such systems are by their nature restricted to bright stars (I <~ 8-9 mags) and thus address only a portion of the physical parameter space.
3.3.3.2 Proposed observations and targets
3.3.3.2.1 Planets around low-mass stars and brown dwarfs
Direct imaging of substellar companions (brown dwarfs and extrasolar planets) is substantially easier around lower mass primaries, since the required contrast ratios are smaller for a given companion mass. Indeed, the first bona fide L dwarf and T dwarfs were discovered as companions to low-mass stars (Becklin & Zuckerman 1988, Nakajima et al 1995). Thus, searching for low- mass stars and brown dwarfs is an appealing avenue for planet detection and characterization.
Given that low-mass stars are so much more abundant than higher mass stars, they might constitute the most common hosts of planetary systems.
Keck NGAO will be a significant advance from previous imaging surveys, reaching much lower companion masses and correspondingly much cooler temperatures. These targets are optically faint, and thus unobservable with current or future NGS/ExAO systems. Current single-LGS systems are only able to reach modest contrast ratios at K-band. Direct detection is more favorable at J and H-bands, where planetary mass companions are brighter and higher angular resolution can be achieved; good performance at these wavelengths is only possible with Keck NGAO.
One very low-mass companion, 2MASS 1207-39B, has recently been directly imaged around a young (~10 Myr) field brown dwarf (~25 MJup), with an estimated mass of ~5 MJup and a projected separation of 60 AU (Chauvin et al 2005). The incidence of similar systems is unknown; given that 2M1207~B was found in a search of only two objects, it is promising that many more wide, planetary companions to brown dwarfs. Discovery of Jovian-mass companions around brown dwarfs would be difficult to explain in conventional theories where planets form in circumstellar disk. While disks around common around young brown dwarfs (e.g., Liu et al 2003), they are unlikely to be massive enough to form such companions (Klein et al 2003). But regardless of their origin, such planetary-mass companions would constitute relatively easily observed systems for studying the spectral characteristics of planetary atmospheres.
Figure 19 JHK color image of the 2MASS 1207-3932 system.
As observed with the VLT NGS system equipped with a near-IR wavefront sensor (Chauvin et al 2005). The primary is a young brown dwarf, with an estimated age of ~12 Myr and ~25 MJup.
The companion has an estimated mass of only ~5 MJup. Only a small number of brown dwarfs can be imaged with sufficient sensitivity and angular resolution with current LGS AO to detect Jovian-mass companions. Keck NGAO will be a major advance for detection and characterization of planets around low-mass
stars and brown dwarfs.
Spectroscopic follow-up of the coldest companions will be an important path in characterizing the atmospheres of objects in the planetary domain. Strong molecular absorption features from water and methane provide diagnostics of temperature and surface gravity at modest (R~100) resolution.
Below ~500 K, water clouds are expected to form and may mark the onset of a new spectral class, a.k.a. ``Y dwarfs''. Such objects represent the missing link between the known T dwarfs and Jupiter, but are probably too faint and rare to be detected as free-floating objects in shallow all-sky surveys such as 2MASS and SDSS. Furthermore, the coolest/lowest mass objects may not exist as free-floating objects if there is a low-mass cutoff to the initial mass function of the star formation process, e.g., from opacity-limited fragmentation of molecular clouds (Mmin ~ 5-10 MJup; Silk 1977). Even cooler/lower mass objects might only form via fragmentation, akin to the formation of binary stars, and only be found as companions.
Figure 20 Planet detection sensitivity for Keck NGAO for two different primary masses and ages.
Based on models by Baraffe et al. (1998, 2003) and high contrast simulations described in section 4.3.2.7. NGAO will be able to search for Jovian- mass companions around large numbers of low- mass stars and brown dwarfs. (Most of the detection limits are contrast-limited, but the outer floor seen in the curves is set by the raw sensitivity of the system. The primary is assumed to be at 30 pc with an on-source integration time of 20 minutes.)
3.3.3.2.2 Very young planets in the nearest star-forming regions
Revealing the earliest stages of planet formation, the first few Myr, is a significant observational challenge. The nearest star-forming regions are >~125 pc away, and thus high angular resolution imaging is needed. In addition, young stars and brown dwarfs can be enshrouded by substantial dust extinction, both from the natal molecular cloud and their own circumstellar material. Thus most young (T Tauri) stars are too optically faint for current NGS AO systems or future ExAO systems.
Imaging searches and characterization at the very youngest (T Tauri) stages provide a unique probe of the origin of extrasolar planets, by constraining their formation timescales and orbital separations. Keck NGAO imaging can probe physical separations of >~5-10 AU around these stars. Multiple methods exist for studying disk evolution at such young ages --- direct imaging of
massive outer planets around T Tauri stars can help to understand the co-evolution of young planets and their natal disks.
Most current models indicate that circumstellar disks are not massive or dense enough to form Jovian-mass planets farther than 10-20 AU. However, brown dwarf companions (~15-70 MJup) have been found at >~100 AU around young stars (e.g., TWA-5B; Lowrance et al. 1999), indicating that substellar companions can exist at larger separations than expected in conventional wisdom. And even if the models are correct, angular momentum exchange between giant planets can induce orbital migration, potentially sending some Jupiters spirally inward and propelling others to much larger separations. Likewise, the early luminosity evolution of giant planets as they are forming is highly uncertain (e.g., Fortney et al. 2005), and direct imaging searches with Keck NGAO can provide insight.
It is still an open question whether giant planets form extremely rapidly (<~104 yr) due to disk instabilities (e.g. Boss 1998) or if they first assemble as ~10 Mearth rocky cores and then accrete
~300 Mearth of gaseous material over a total timescale of ~1-10 Myr (e.g. Lissauer 1999).
Potentially both mechanisms may be relevant, depending on the range of orbital separations and circumstellar disk masses. In addition, imaging searches of both young T Tauri stars with disks (classical TTS) and without disks (weak TTS) can help to constrain the formation timescale. In particular, weak T Tauri stars with planetary companions would suggest that planet formation could occur even when disk evolution/dissipation happens rapidly.
3.3.3.3 Comparison of NGAO w/ current LGS AO
Current LGS AO can detect Jovian-mass planets only around the youngest (<~30 Myr) low-mass stars and brown dwarfs, which are quite rare in the solar neighborhood. Keck NGAO will be able to achieve sufficient contrast for planet detection around primaries as old as ~500 Myr, and thus a much larger sample can be surveyed. For T Tauri star searches, by their very nature, these objects are in highly extinction regions, where optical tiptilt star availability is poor.
Figure 21 Schematic comparison of the relative parameter space for direct imaging of planets.
As probed by Keck NGAO and ExAO systems in development by Gemini and VLT. The optical faintness of low-mass stars, brown dwarfs, and the youngest stars makes them inaccessible to ExAO systems, but hundreds of these objects can be imaged with Keck NGAO.
3.3.3.4 AO and instrument requirements
Essential: High contrast near-IR imager with coronagraph, along with means to obtain follow-up low-resolution (R~100) spectroscopy.
Desirable but not absolutely essential: Thermal IR (L-band) photometry and spectroscopy.