2011 Annual Science Report

Carnegie Institution of Washington Reporting  |  SEP 2010 – AUG 2011

Project 1: Looking Outward: Studies of the Physical and Chemical Evolution of Planetary Systems

Project Summary

This project five main objectives focused broadly on understand the origin and early evolution of our solar system. First, we have employed a new planet finding spectrometer to aid in detecting planetary systems surrounding neighboring stars. Second, we have begun the Carnegie Astrometric Planet Search project to detect giant planets around nearby loss mass dwarf stars. Third, we focused on understanding of radial transport and mixing of matter in protoplanetary disks. Fourth, we have continued to survey of small planetary size objects in the Kuiper belt. Fifth, we have continued our studies of the composition, structure, and ages of circumstellar disks.

4 Institutions
3 Teams
52 Publications
3 Field Sites
Field Sites

Project Progress

1.1 Radial velocity searches for planetary systems hospitable to life

Over the past fifteen years CoI Butler and our planet search programs have clarified the statistical outlines of the planetary census for giant planets with velocity semi-amplitudes K ≥ 10 m/s and periods P ≤ 10 yr. This work has had a profound impact on astronomy and science as a whole, motivating the new field of astrobiology. The study of extrasolar planets is cited as one of the three pillars of modern astrophysics in the most recent Decadal Survey.

All of the lowest mass planets around nearby stars have emerged from high cadence precision Doppler runs. The two crucial ingredients in detecting terrestrial mass and potentially habitable planets around nearby stars are measurement precision approaching 1 m/s, and long blocks of telescope time. After nearly a decade of work, the Planet Finding Spectrometer on the Magellan telescope is producing 1 m/s precision and finding terrestrial mass planet candidates in and near the habitable zone of M dwarfs, as shown in Figure 1.1.

Next month (November 2011) the Automated Planet Finding telescope at Lick Observatory will enter science mode. Our group, led by Steve Vogt, will get half of the time on this facility. Within the next year we expect low mass and potentially habitable planets will begin spilling out of these systems.

1.2 Astrometric search for planetary systems hospitable to life

Co-I Boss leads the Carnegie Astrometric Planet Search (CAPS) project at LCO, while Co-I Weinberger and NAI Fellow Anglada-Escude are key members of the CAPS team. The CAPS team is undertaking an astrometric search for gas giant planets and brown dwarfs orbiting nearby low mass dwarf stars with the LCO du Pont telescope. The plan is to follow about 100 nearby (primarily within about 10 pc) low mass stars, principally late M, L, and T dwarfs, for 10 years or more, in order to detect very low mass companions 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 red to be included in ground-based Doppler planet surveys, which are often optimized for FGK dwarfs. The smaller masses of late M dwarfs also yield correspondingly larger astrometric signals for a given mass planet. The CAPS search will also help to determine whether gas giant planets form primarily by core accretion or by disk instability around late M dwarf stars, and help with the census of nearby stars, where habitable planets are now known to exist (e.g., Gl 581g).

Our first paper published in 2009 described many of the details of the CAPSCam camera, planet search program, and data analysis pipeline. Our second paper (Anglada-Escude et al. 2011) places an astrometric upper mass limit on the known Doppler exoplanet GJ 317b (Johnson et al. 2007). Our results are presented in Figure 1.2. GJ 317 is a relatively bright M3.5 M dwarf, necessitating the use of the CAPSCam Guide Window (GW) mode with 0.2 sec exposures. The GJ 317 analysis showed that the limiting astrometric accuracy, at least for the brighter targets in our sample, is about 0.6 mas per epoch in Right Ascension (RA) and about 1 mas per epoch in Declination (Dec). The errors in Dec are worse than in RA due to a data count ramp in the CAPSCam Guide Window, which is used for the brighter targets. With 18 epochs, the overall accuracy in fitting the parallax of GJ 317 is about 0.15 mas, and the accuracy should improve for fainter targets. With a similar number of epochs, this accuracy is sufficient to detect a Jupiter-mass companion orbiting 1 AU from a late M dwarf 20 pc away with a signal-to-noise ratio of about 4.

We have now gathered (at least) first epoch CAPSCam data for all of the ~100 targets on the current target list. 42 targets have at least 7 epochs of data, and about a dozen have enough epochs (as many as 23) to confirm known low mass binary companions or to place upper mass limits on known Doppler planets. The CAPSCam effort is on the verge of yielding a steady stream of astrometric measurements of the dynamics of late M dwarfs and their very low mass companions.

1.3 From Disks to Planets: Early mixing and transport in young disks

Co-I Boss is partially supported by the NASA Origins of Solar Systems Program to work on mixing and transport in marginally gravitationally unstable disks. This work is being continued and extended to include an analysis of the time history of a population of individual dust grains, as they traverse high and low temperature regions of the disk. This will allow a determination of the extent to which water is transported as a solid by the dust grains, through a collaborative effort with Prof. Morris Podolak of Tel Aviv University, who has developed a model for silicate dust grains with water ice mantles. By following the extent to which water ice condenses or sublimates on a population of grains being transported around the disk, a better understanding of the distribution of water in the solar nebula will be achieved, with important implications for the delivery of water to the terrestrial planets.

Along with Co-I Conel Alexander, Boss and Podolak are currently preparing a paper for submission to Science, presenting the results shown by Boss at the August 2011 Meteoritical Society meeting in London. The protosun is likely to have experienced multiple FU Orionis outbursts caused by rapid mass accretion from a gravitationally unstable solar nebula. 3D hydrodynamical models of the trajectories of particles in the solar nebula during such a phase show that cm-sized particles can traverse distances of 10 AU or more, both inward and outward, in the midplane of the nebula, in less than 1000 yrs. The particles are subjected to extreme environmental variations during these journeys, such as temperatures ranging from 60 K to 1400 K, pressures from ~10-9 to ~10-7 bar, and oxygen isotope ratios ranging from planetary to solar composition. The models predict that many calcium, aluminum-rich inclusions (CAIs) could have survived FU Orionis phases and been transported across the nebula, from near the protosun, out to the comet-forming regions, and back again, multiple times, acquiring along the way rims with mineral phases indicative of their thermal histories, and oxygen isotope variations derived from the local dispersion of these isotopes.

1.4 Late mixing and migration in the Solar System: Completing the Inventory of the Outer Solar System

CoI Sheppard has surveyed the southern sky and Galactic plane for bright Trans-Neptunian objects. This survey is one of the first large-scale southern sky and Galactic Plane surveys to detect dwarf planets and other bright Kuiper Belt Objects in the trans-Neptunian region. A total of 2500 square degrees was searched in the survey. About 2200 square degrees of the survey were south of declination -25 degrees, where northern KBO surveys cannot efficiently observe. The surveyed area includes almost all of the southern sky within about 20 degrees of the ecliptic. Another 300 square degrees of sky was surveyed in the northern Galactic Plane near the ecliptic using optimal image subtraction techniques to remove the stellar background. Eighteen bright trans-Neptunian objects were discovered, including some of the most southern outer solar system objects ever detected as well as the intrinsically brightest solar system objects discovered in several years (2010 EK139, 2010 FX86 and 2010 KZ39). Based on 2010 KZ39’s orbit, it is a possible Haumea family member candidate, the only known family in the Kuiper Belt, but no significant water ice was detected on 2010 KZ39’s surface which is unlike the other known Haumea family members.

Assuming moderate albedos, several of the new discoveries from this survey could be in hydrostatic equilibrium and thus could be considered dwarf planets. Combining this survey with previous surveys from the northern hemisphere suggests that the Kuiper Belt is nearly complete to around 21st magnitude in the R band, as shown in Figure 1.4. The corresponding size of an object at 21st magnitude depends on the distance and albedo of the object. Assuming a moderate albedo of 15 percent, at 30 AU, 21st magnitude corresponds to a radius of 80 km while at 50 AU it is 225 km. Through looking at the cumulative luminosity function of the KBOs, significant incompleteness in the main Kuiper Belt probably starts around a radius of 100 km and becomes drastic around a radius of 60 km.

All the main dynamical classes in the Kuiper Belt are occupied by at least one dwarf-planet-sized object. The 3:2 Neptune resonance, which is the innermost well-populated Neptune resonance, has several large objects while the main outer Neptune resonances such as the 5:3, 7:4, 2:1, and 5:2 do not appear to have any large objects. This indicates that the outer resonances are either significantly depleted in objects relative to the 3:2 resonance or have a significantly different assortment of objects than the 3:2 resonance. For the largest objects, the scattered disk population appears to have a few times more objects than the main Kuiper Belt population, while the Sedna population could be several times more than that of the main Kuiper Belt population.

Beyond the Kuiper Belt edge, at a few hundred AU or so, there could easily be more Pluto, Mercury, or even larger sized objects in Sedna-like orbits. No new Sedna-like objects were detected even though the survey was sensitive to objects up to about 300 AU. Sedna is likely one of the larger and thus one of the brighter members of its population. Any further Sedna-like object detections will likely require significantly fainter magnitudes while still covering large areas of sky. Pan-STARRS has a chance to detect some Sedna-like objects since it will survey large areas of sky to around a magnitude fainter than this survey, but LSST will be needed to find significant numbers of Sedna-like objects since sensitivity and large areas of sky are needed to probe this distant, faint population.

1.5 Compositions of circumstellar disks and delivery of volatiles to terrestrial planets

CoI Weinberger continues to study the compositions, structures, and ages of circumstellar disks so as to learn about the birth environments of planets. She completed her Spitzer Space Telescope study of the extremely dust star BD+20 307, which has orders of magnitude more warm dust than any other star its age (>1 Gyr). Warm dust, arising in this case within 1 AU of the star could be the signature of a giant collision. A remarkable feature of this star is that no additional cold dust is needed to fit the infrared excess as shown in Figure 1.5. Peaks in the 10 and 20 μm spectrum are well fit with small silicates that should be removed on a timescale of years from so close to the star. Only four stars out of 600,000 in the IRAS catalog have substantial 24 micron excess. As we wrote in our paper, “This fraction of host dust stars implies a giant impact rate greater than 0.2 impacts/star during its main sequence lifetime, i.e., implies that such impacts are common. Obviously, these are small number statistics. The WISE mission currently surveying the sky at mid-infrared wavelengths will greatly improve the statistics.” (Weinberger et al. 2011)

Weinberger has been working with a graduate student, Joey Rodriguez, from George Mason University to study the composition of the dust around Beta Pictoris using archival Hubble Space Telescope observations taken with the STIS instrument. A long slit with a bar was placed over the star in order to suppress scattered light, and a point spread function star of the same spectral type was also observed. We spectrally scale and subtract this PSF from the 2D spectrum of Beta Pic to reveal the spectrum of light scattered off of the disk dust. There are hints of a change in the color of scattered light between 380 and 550 nm as a function of distance from the star. At 3.5 arcsec way, the scattered light appears redder, but more work remains on analyzing these data.

To study the evolution of disks amongst young and co-eval stars, it is necessary to identify members of nearby associations. Nearby young stars of age 5 – 10 Myr provide our best opportunity to study the late stages of star and planet formation. During this time period, the last gas-rich disks dissipate and the onset of the debris disk phase occurs. Weinberger worked with postdoctoral researcher Evgenya Shkolnik and others to find new M star members of the TW Hya association. Of the two new members found, one has broad Halpha emission indicative of accretion (Shkolnik et al. 2011). TWA disks are bimodal in mass — some are primordial but most are almost entirely dissipated. Weinberger presented her work on the distances to the TW Hya Association members at the January 2011 American Astronomical Society Meeting. We have observed 14 TWA primary members with the CAPSCam instrument at the 2.5m DuPont Telescope at Las Campanas Observatory in order to measure parallaxes. Our final precisions are 1 mas total from statistical and systematic uncertainties. To trace the stars back in time, we take their present positions and now available three-dimensional space velocities and compute their locations in Galactic coordinates. Surprisingly, there is no time when the stars are substantially more congregated than they are today, bringing into question whether this Association is truly an association.

Figure 1.1 Radial velocity measurements of a nearby M dwarf taken this year with the Planet Finding Spectrometer by co-I Butler. This new spectrometer was built by Butler and colleagues for Carnegie’s Magellan (Clay) Telescope. The instrument is producing 1 m/s precision and will allow the detection of planets in and near the habitable zones of M stars. The data shown here reveal a super-earth in a 19 day orbit, which places the planet slightly interior to the habitable zone of the star.

Figure 1.2 The Carnegie Astrometric Planet Search (CAPS) program has placed its first constraint on the mass of a gas giant exoplanet, orbiting outside the habitable zone of its host star (Anglada-Escude et al. 2011). The red histogram shows the distribution of the possible masses of GJ 317b, based on both Doppler and CAPSCam astrometric data. The cumulative probability for the maximum mass of GJ 317b is shown in grey. The Doppler data acts as a strong prior (abrupt cut below 1.8 MJup), effectively suppressing solutions below the minimum mass. The upper limit on the mass is set by CAPS astrometry at 3.6 MJup.

Figure 1.3 Combining the large scale southern sky survey of Sheppard et al. (2011) with other surveys, the known objects in the Kuiper Belt region are complete to about 21st magnitude, shown by the shaded region below the dashed line. It is likely that everything under the dashed line at 21 mag is known. Radius of an object is shown assuming a moderate albedo of 0.15 for various heliocentric distances and apparent red magnitudes. The radii used for the named objects in the figure are Pluto (1161 km), Mercury (2440 km), Mars (3396 km), Earth (6371 km), and an arbitrary lower limit on a hypothetical eccentric giant planet or companion to our Sun, sometimes called Nemesis or Tyche, with 10,000 km radius. In the distant solar system very large objects would easily be undetected to date.

Figure 1.5 The remarkable disk-bearing star BD+20 307 has tremendous infrared excess (Spitzer Space Telescope photometry, red circles) and prominent silicate emission (Spitzer spectroscopy, blue line). Flux from the star is shown in black. Our best-fit model to the spectrum, composed of a simple set of amorphous and crystalline silicates as well as large blackbody grains, is shown in yellow. All components are situated at the best-fit distance of 0.85 AU. We do not detect any cold dust, but we find a high fraction of very small grains. The lack of detectable cold dust rules out a cold belt of planetesimals that can lose mass at a rate sufficient to feed the hot dust, and suggests a giant collision such as the Moon forming event, has recently occurred around this binary star.