Chapter 2

Chapter 2, The Science Case

Section 2.5: High Dynamic Range Science

NOAO Logo    Gemini Logo

2.5.1 SCIENCE OVERVIEW Extra-solar Planets

As described in Planet Formation Environments, precision radial velocity studies have revealed the existence of planets outside the solar system and have given us an inkling of the diversity in the properties of extra-solar planetary systems based on their rough demographics (i.e., their a, e, and M sin I distributions). Even so, we still know relatively little about the fundamental properties-such as mass, radius, and composition-of any individual planet.

One exception is the transiting planet HD209458b, which illustrates the next level of detail with which we might hope to characterize extra-solar planets. In the case of HD209458b (a = 0.05), the detailed information revealed by the transit constrains not only some fundamental planetary properties1,2, but also the evolutionary history of the planet. For example, the transit establishes the inclination of the system (86.7 degrees), and therefore the mass of the planet (~ 0.69 MJ), as well as the planetary radius (1.32 RJ). The large radius establishes that the planet is a gas giant rather than a rocky or icy planet; that is, planets composed of pure olivine or water ice with a mass of 0.69 MJ would have radii of 0.31 RJ or 0.45 RJ3.

As described by Burrows et al.,3 the radius of HD209458b does not simply reflect the expansion of the planet's atmosphere due to recent stellar irradiation, but more significantly, it requires the planet to have experienced prolonged stellar irradiation that has retarded the (evolutionary) contraction of the planet. Thus, the large size of HD209458b at the present epoch is a tight constraint on the amount of time (<107 yr) it could have spent far (> 0.5 AU) from its primary. HD209458b either formed close to its present orbital distance, or it migrated to its present distance within 107 yr of its formation.

GSMT and Planetary Masses and Radii: Although transits are expected to be relatively rare, Burrows et al.3 estimate that planetary transits may be observed in perhaps 1/10 of extra-solar planets with a < 0.1 AU. Directly imaging extra-solar planets would open up the possibility of obtaining similarly detailed information for a much larger number of planets, including those with orbital distances more comparable to the giant planets in our solar system. For example, measuring the system inclination from the visual binary orbit will establish the mass of the planet. Additional multi-wavelength imaging or spectroscopy of the planet will reveal detailed information about the planetary atmosphere (cloud composition and planetary albedo, etc.). Given a measured albedo, the planetary radius can be deduced from the luminosity of the planet. With synoptic monitoring, we may even be able to chart the weather on planets outside the solar system. With this level of detail, we will be able to expand planetary science to encompass planets outside the solar system. And we will, therefore, be able to examine in greater detail the degree to which extra-solar planets are similar to or different from the planets in our solar system.

Cloud Condensation and Planetary Albedos: The emergent spectrum from the planet is a combination of light reflected from the central star (primarily at lambda < 2.5m) and thermal emission from the planet (primarily at lambda > 2m). As described by Marley et al.4, the reflected spectrum is affected by processes such as rayleigh scattering, scattering by clouds, and molecular absorption. Rayleigh scattering is important in the blue, while molecular absorption (by species such as TiO, VO, NH3, CH4, H2O, and H2) is significant at red and infrared wavelengths. As a result, planets with low albedos (i.e., no clouds) are expected to be dark in reflected light at > 0.6m. However, when planetary atmospheres are cool enough to permit the condensation of clouds, planets can be significantly brighter at near-infrared wavelengths due to the bright reflecting cloud layer (see Figure 1).

Figure 1 Reflected and thermal emission from a Teff                                              = 300 K 13 MJ planet orbiting at 1 AU from its G2V primary, when viewed at a distance of 10 parsecs. (ref 4) Reflected spectra are shown for the case with no clouds (solid line), and at two supersaturation ratios (dashed and dotted lines). The predicted thermal emission (long dashed line) assumes no clouds.

The cloud composition, and therefore the spectral appearance and detectability of planets in reflected light, is a strong function of orbital radius. Given the orbital parameters and current models of extra-solar planetary atmospheres, planetary surface temperatures are expected to range from less than approx 200 K for planets at greater than approx 1 AU to ~ 1500 K for some of the close-in planets (~ 0.05 AU). Over this range in temperature, the major cloud condensates are expected to be silicate clouds (e.g., enstatite, MgSiO3) at < 1000 K, water clouds at < 500 K, and NH3 clouds at < 200 K. Not only will clouds increase planetary albedos, they will also partially fill in absorption features. The depth and shape of absorption bands in the near-IR is a powerful diagnostic of the cloud properties of planetary atmospheres.

For the range of orbital separations accessible to the GSMT (greater than approx 0.1"; or 1 AU at 10 pc), we are likely to detect planets with H2O and possibly NH3 clouds, and consequently, high albedos (0.4-0.6) at lambda < 1.5m, dropping off quickly beyond ~ 2m.5 As some specific examples, planets at orbital distances ~ 1.0 AU are expected to have Tsub eff ~ 200 K, the most highly reflective atmospheres (due to water clouds), and gaseous CH4 absorption features. Planets at orbital distances ~ 5.0 AU (Tsub eff less than approx 150 K) are expected to have NH3 clouds above a H2O cloud deck, and gaseous CH4 absorption features.5

Weather and Rotation: Synoptic photometric studies of extra-solar planets may hope to detect the effects of planetary weather, e.g., the coming and going of clouds, as modulated by the rotation of the planet. In this case, we would expect to see a time variability signature on two timescales: the timescale for large scale changes in planetary weather, and the planetary rotation timescale. By monitoring the flux in and out of spectral regions of high albedo, it may be possible to map out the surface filling factor of clouds as a function of rotational phase. Such weather-related modulations have been tentatively reported in recent studies of the higher mass L dwarfs (e.g., Bailer-Jones & Mundt).6

Thermal Emission: In addition to the possibility of detecting planets in reflected light, we may also be able to detect thermal emission from planets at longer wavelengths. Due to strong line blanketing by molecular bands and collisionally broadened lines, strong thermal emission is expected in spectral regions of low atmospheric opacity. These atmospheric opacity holes are (conveniently) typically the same as the ground-based observing bands, due to the common opacity sources in the atmospheres of the Earth and other planets. Because the excess (line blanketed) flux emerges through the opacity holes, planets can be significantly brighter in these spectral regions. For a planetary atmosphere at 200 K, for example, the excess emerging in the 5mm region is ~ 105 above blackbody for a cloudless atmosphere (e.g., Burrows et al.).3 Estimated GSMT Performance: Characterizing Planets via Reflected Light

The discussion in High Dynamic Range Imaging suggests that a 30m diameter GSMT equipped with an AO (Adaptive Optics) system delivering images with Strehl ratio 0.8 at 2 will enable detection of planets fainter by ~ 5x106 times that of their parent stars at angular distances between 0.1" and 0.4"; beyond 0.4", observable contrast ratios will be larger. We note that this estimate assumes that: (1) a high performance AO system with wavefront correction achieved using a deformable mirror with ~ 1000 actuators; (2) light from the central star is blocked by a coronagraphic stop; and (3) it will be possible to average the effects of wavefront fluctuations over a period of 1 hr and achieve a s/n gain of ~ 30 during that integration period.

At a distance of 10 pc, a planet with radius equal to that of Jupiter with an albedo A ~ 0.5 at 2 and located at a distance of 1 AU (0.1") from its parent star will be ~ 107 times fainter, with an apparent magnitude m (2.2) ~ 21 mag. It would therefore be readily detectable by GSMT, in broad-band. Longer integrations will be required in order to diagnose key atmospheric features via low resolution spectroscopy (see Figure 4 in Section

These estimates suggest that, with a plausible AO system, GSMT will have the power to detect and characterize Jupiter-like planets orbiting at distances ~ 1 AU from parent stars located within 10 pc, thus making it a powerful tool for study of extra-solar giant planets. Estimated GSMT Performance: Characterizing Planets via Thermal Emission

Here we consider detecting and characterizing planets at 5, where thermal emission dominates. An AO system similar to that assumed above could in principle deliver images with Strehl ratios approaching 0.99. If so, it should be possible to achieve star/planet contrast ratios approaching 108 in 1 hr integration for separations theta > 0.4", assuming the use of a cooled coronagraphic mask.

At ages 0.1, 1.0, and 5 Gyr, and at a distance of 10 pc, a Jupiter-mass planet will have monochromatic brightness at 5, m (5) = 9.5, 13.2, and 18.0 mag, respectively. By contrast, a solar-type parent star will have m = 3.8 mag, resulting in contrast ratios of 200, 5000, and 3x106 between the star and planet. GSMT should thus be able to readily detect Jovian mass planets with ages less than 1-2 Gyr, provided they are located 0.4 "(4 AU) or more from their parent star. Characterization of their atmospheres will require longer integration times as shown in Figure 4 in Section

Hence, GSMT will enable direct imaging search for Jovian (and young, t < 1Gyr, sub-Jovian) mass planets located at r > 4 AU for all solar-like stars within 10 pc. Follow-up moderate resolution spectroscopy will provide detailed characterization of their atmospheric composition, along with an estimate of surface gravity. Its extraordinary sensitivity will also enable detection of thermal emission from similar-temperature Uranus- and Neptune-like planets.


As described in Planet Formation Environments, detailed studies of the physical and dynamical structure of young planet forming disks are needed to measure the physical conditions under which planets form, to constrain the efficiency of various physical processes, and for the indirect detection of forming protoplanets. Possible high angular resolution measurements, which complement the high spectral resolution measurements discussed in Planet Formation Environments, include the following:

  1. The detection of gaps created by large protoplanets in images of disks seen in scattered light (1-2.5m) from the central star or thermal emission from the disk (> 5m). These observations would extend to large samples of stars the detection of gaps created by stellar mass companions (e.g., first reported by Roddier et al.; see Figure 2).7 They also promise detection of gaps opened by lower companion masses as well as sampling gaps created within an AU of the parent star. The scattered light measurement will characterize the spatial distribution of the dust in the disk, which will complement ALMA (Atacama Large Millimeter Array) measurements of the spatial distribution of the gaseous component of the disk (e.g., Guilloteau et al.; see Figure 2).8 Such measurements can also be used to probe the grain size distribution, e.g., as demonstrated by Close et al.9 in their study of the circumstellar disk around the young binary system UY Aur. High angular resolution GSMT measurements of the thermal emission from disks would extend recent thermal emission measurements of disks in systems such as HR4796 (e.g., Telesco et al.; see Figure 3)10 to probe more detailed disk structures, including those possibly created by young protoplanets.

    Figure 2 The young binary system GG Tau (separation 0.3 arcsec) observed by Guilloteau et al. (ref 8) in 13CO J       = 2-1 line emission (left; red, white, and blue contours), 1.4 mm continuum emission (left; orange ring), and by Roddier et al. (ref 7) scattered J-band light (right).

    Figure 3 HR4796 as observed by Telesco et al., (ref 10) 18.2um (left). In the right panel, the 18.2um contours are overlayed on the 1.1um HST/NICMOS (Hubble Space Telescope/Near-Infrared Camera and Multi-Object Spectrometer) (image of Schneider et al.) (ref 11).  The mid-infrared and near-infrared emissions are both sharply peaked at 70 AU radius. The multi-wavelength images indicate a typical grain size of 2-3um for the mid-infrared emission. These grains are too small to be primordial and are probably the result of the collision of larger bodies.

  2. Spatially resolved gas content and dynamics of disks. One of the most exciting possibilities to measure the structure in the gaseous component of disks using the pure rotational lines of H2. As described in Section 2.4, although ISO detected surprisingly strong pure rotational H2 emission from disk systems over a range of ages (<1-10 Myr),12 the lack of spectral or spatial resolution of the measurements makes it difficult to determine whether the gas is located in the traditional planet forming region and available for the formation of planets. Estimated GSMT Performance: Characterizing Scattered Light Disks

For a disk with optical depth tau (2.2) = 3x10-4 surrounding a solar-type star located at 10 pc, the predicted contrast between the disk and parent star is 6x10-6 and 2.5x10-7 at radial distances of 1 AU (0.1") and 5 AU (0.5") respectively; at those same distances, the corresponding K-band magnitudes are 17 and 20.5 mag, thus enabling straightforward detection and characterization of these disks (see Figure 4 in Section

Hence, GSMT should enable detailed study of the structure at ~ 0.2 AU resolution of secondary dust disks surrounding nearby stars in the critical region between 1 AU and 5 AU, and thereby provide the basis for modeling the dynamical interactions between planetary systems and their associated disks. Estimated GSMT Performance: Thermal Emission from Optically Thin Dust Disks

For a disk with optical depth tau (5) = 5 x 10-4 surrounding a solar-type star located at a distance of 10 pc, the predicted contrast ratio between disk and stellar emission is 0.25 and 2.5x104 at 1 AU (0.1") and 5 AU (0.5") respectively; at those distances, disk temperatures are 300 K and 180 K. The M-band magnitudes are 5.5 and 13.0 mag, thus enabling straightforward resolution and detection of normal emission from disks at angular resolutions 0.05" (0.5 AU at 10pc).


These observations require both a high performance AO system and a coronagraph. A design concept for such an instrument is discussed in the Instrumentation section.


  1. Charbonneau, D., Brown, T.M., Latham, D.W., & Mayor, M. "Detection of Planetary Transits Across a Sun-like Star", ApJ, 529, L45, 2000.
  2. Brown, T.M., Charbonneau, D., Gilliland, R.L., Noyes, R.W., & Burrows, A. "Hubble Space Telescope Time-Series Photometry of the Transiting Planet of HD 209458". ApJ, 552, 699, 2001.
  3. Burrows, A., Hubbard, W.B., Lunine, J.I., & Liebert, J. "The theory of brown dwarfs and extrasolar giant planets". Rev. Mod. Phys. 73, 719 (2001).
  4. Marley, M.S., Gelino, C., Stephens, D., Lunine, J.I., & Freedman, R. "Reflected Spectra and Albedos of Extrasolar Giant Planets. I. Clear and Cloudy Atmospheres". ApJ, 513, 879, 1999.
  5. Sudarsky, D., Burrows, A., & Pinto, P. "Albedo and Reflection Spectra of Extrasolar Giant Planets". ApJ, 538, 885, 2000.
  6. Bailer-Jones, C.A.L., & Mundt, R. "Variability in ultra cool dwarfs: Evidence for the evolution of surface features". A&A, 367, 218, 2001.
  7. Roddier, C., Roddier, F., Northcott, J.E., Graves, J.E., Jim, K. "Adaptive Optics Imaging of GG Tauri: Optical Detection of the Circumbinary Ring". ApJ, 463, 326, 1996.
  8. Guilloteau, S., Dutrey, A., & Simon, M. "GG Tauri: the ring world". A&A, 348, 570, 1999.
  9. Close, L.M., Dutrey, A., Roddier, F., Guilloteau, S., Roddier, C., Duvert, G., Northcott, M., Menard, F., Graves, J.E., & Potter, D. "Adaptive Optics Imaging of the Circumbinary Disk around the T Tauri Binary UY Aurigae: Estimates of the Binary Mass and Circumbinary Dust Grain Size Distribution". ApJ, 499, 883, 1998.
  10. Telesco, C.M., Fisher, R.S., Pina, R.K., Knacke, R.F., Dermott, S.F., Wyatt, M.C., Grogan, K., Holmes, E.K., Ghez, A.M., Prato, L., Hartmann, L.W., & Jayawardhana, R. "Deep 10 and 18 Micron Imaging of the HR 4796A Circumstellar Disk: Transient Dust Particles and Tentative Evidence for a Brightness Asymmetry". ApJ, 530, 329, 2000.
  11. Schneider et al. "NICMOS Imaging of the HR 4796A Circumstellar Disk". ApJ, 513, L127, 1999.
  12. Thi, W.F., van Dishoeck, E.F., Blake, G.A., van Zadelhoff, G.J., Hogerheijde, M.R. "Detection of H_2 Pure Rotational Line Emission from the GG Tauri Binary System". ApJ, 521, L63, 1999.

November 2002