Prospects for the Detection of Protoplanets [Review]

Sebastian Wolf

Emmy Noether Research Group “Evolution of Circumstellar Dust Disks to Planetary Systems” Max Planck Institute for Astronomy

discs06 – Cambridge, UK [July 20, 2006] Planet Formation in a Nutshell Theory

Star Formation Process Circumstellar Disks Planets

(sub)µm particles • Brownian Motion, Sedimentation, Drift • Inelastic Collision => Coagulation

cm/dm grains • Agglomeration; Fragmentation

Planetesimales

• Gravitational Interaction: Oligarchic Growth

Planets (cores) • Gas Accretion

(Waelkens 2001) Alternativ: Gravitational Instability Giant Planet how to identify

IRAS 04302+2247 Grain Growth?

• SED: (sub)mm slope • Shape of ~10μm silicate feature • Scattered light polarization (e.g., spectro-polarimetry)

HK Tau • Multi-wavelength imaging

+ Radiative Transfer Modelling The “Butterfly Star” in Taurus Wolf, Padgett, & Stapelfeldt (2003)

IRAS 04302+2247

HK Tau

IRAS 04302+2247 The “Butterfly Star” in Taurus

submm-sized grains in the disk midplane,

IRAS 04302+2247 instellar-like grain size in the circumstellar envelope

” .3 4 ~ HK Tau U A 0 0 6

IRAS 04302+2247 Wolf, Padgett, & Stapelfeldt (2003) Size Scales

Solar System

Angular diameter of the orbit of solar system planets in a distance of the Taurus star-forming region (140pc)

Neptune - 429 mas Jupiter - 74 mas Earth - 14 mas Mid-infrared interferometric spectroscopy - Dust processing in the innermost regions

MIDI setup

[Leinert et al. 2004] Mid-infrared interferometric spectroscopy - Dust processing in the innermost regions

Silicate Feature

[Leinert et al. 2004] Effect also found in disks around TTauri Stars - (Schegerer & Wolf, in prep.) Finding Planets – In Disks?!

UV – (N)IR Scattering Additional Problems IR – mm Thermal Reemission (Dust...) Young disks Extinction (Inclination-dependent)

Dust parameters, T(r,θ,φ), ρ(r,θ,φ) Response of a gaseous, viscous protoplanetary disk to an embedded planet

(Bate et al. 2003) Disk surface densities for planets with masses 1, 0.1, and 0.01MJ orbiting a 1Msun star

[see also Bryden et al. 1999, Kley et al. 2001, Lubow et al. 1999, Ogilvie & Lubow 2002, D‘Angelo et al. 2003, Winters et al. 2003] 0.01 MJ

0.03 MJ

0.3 MJ

1 MJ

(Bate et al. 2003) Is this what we have to look for?

[ G. Bryden ] Jupiter in a 0.05 Msun disk around a solar-mass star as seen with ALMA

d=140pc Baseline: 10km

λ=700μm, tint=4h

(Wolf et al. 2002) inclination = 5° Scattered light images

λ = 1.14μm Gap width (FWHM): ~ 1 AU. [7 AU × 7 AU].

based on 2D density distributions resulting from hydrodynamics simulations vertical structure is assumed to be Gaussian with a scale height that varies as a power law with radius (i.e., a flared disk) inclination = 70° no gap gap [ 2M planet at 1 AU ] [ no planet ] J [Varniere et al. 2006] Log( Disk surface brightness) Azimuthally averaged optical surface brightness profiles for a disk with / without an embedded planet. Decrease in the surface brightness profile near the planet 2MJ planet at 1 AU Bright bump in the profile at the outer edge of the gap

[ Varniere et al. 2006 ]

Richling (in prep.): Observability of gaps depending on wavelength + viewing angle (based on irradiated α-disk models)

Steinacker & Henning (2003): Analysis of the spectral appearance of gaps

Wilner et al. 2004: Observations of the inner disk structure with the Square Kilometer Array (Science Case Study) Gaps Small-scale spirals encircling the planet (detached from the global spiral) in young disks = - In the vicinity Feature of a circumplanetary disk of the planet

Zoom in onto the planet: Disk surface densities for a planet with a mass of 0.5MJ orbiting a 1Msun star. Plus signs: Lagrange points. Overplotted curve: Roche lobe. (D‘Angelo et al. 2002) Is this what we have to look for?

Density distribution in the midplane Can we map young giant of the circumstellar disk with an embedded massive planet. planets? Close-up view: Planetary region

Procedure Density Structure (2D Hydrosimulation)

Stellar heating (3D Radiative transfer)

Planetary heating (3D Radiative transfer)

Prediction of Observations

Wolf & D’Angelo (2005) Close-up view: Planetary region

Mplanet / Mstar = 1MJup / 0.5 Msun

Orbital radius: 5 AU 50 pc Disk mass as in the circumstellar disk as around the Butterfly Star in Taurus

Maximum baseline: 10km,

900GHz, tint=8h

100 pc Random pointing error during the observation: (max. 0.6”) ; Amplitude error, “Anomalous” refraction; Continuous observations centered on the meridian transit; Zenith (opacity: 0.15); 30o phase noise; Wolf & D’Angelo (2005) Bandwidth: 8 GHz Close-up view: Planetary region

1. The resolution of the images to be obtained with ALMA will allow detection of the warm dust in the vicinity of the planet only if the object is at a distance of not more than about 100 pc. 50 pc For larger distances, the contrast between the planetary region and the adjacent disk in all of the considered planet/star/disk configurations will be too low to be detectable.

2. Even at a distance of 50 pc, a sufficient resolution to allow a study of the circumplanetary region can be obtained only for those configurations with the planet on a Jupiter-like orbit but not when it is as close as 1 AU to the central star.

3. The observation of the emission from the dust in the vicinity of the planet will be possible only in the case of the most massive, young circumstellar disks we analyzed. 100 pc

Wolf & D’Angelo (2005) Strong spiral shocks near the planet are able to decouple the larger particles (>0.1mm) from the gas => Gaps in young disks formation of an annular gap in the dust, - even if there is no gap in the gas density two-fluid simulations (example: gap in 1mm grains opened by a 0.05MJup planet)

PaardeKooper & Mellema (2004) Imaging in the Mid-infrared (~10micron)

Hot Accretion Region around the Planet

10μm surface brightness profile of a T Tauri disk with an embedded planet ( inner 40AUx40AU, distance: 140pc) [Wolf & Klahr, in prep.] i=0deg i=60deg

Science Case Study for T-OWL: Thermal Infrared Camera for OWL (Lenzen et al. 2005) Justification of the Observability in the Mid-IR for nearby objects (d<100pc) T-OWL Thermal Infrared Camera for OWL

5 AU

10 pc 30 pc 70 pc

Wolf, Klahr, Egner, et al. 2005 in Lenzen et al. 2005 MATISSE Multi AperTure Mid-Infrared SpectroScopic Experiment

High-Resolution Multi-Band Image Reconstruction + Spectroscopy in the Mid-IR

Proposed 2nd Generation VLTI Instrument

Specifications:

• L, M, N, Q band: ~2.7 – 25 μm • Spectral resolutions: 30 / 100-300 / 500-1000 • Simultaneous observations in 2 spectral bands

What’s new?

• Image reconstruction on size scales of 3 / 6 mas (L band) 10 / 20mas (N band) using ATs / UTs • Multi-wavelength approach in the mid-infrared 3 new mid-IR observing windows for interferometry (L,M,Q) • Improved Spectroscopic Capabilities MATISSE

MATISSE

Precursor of Darwin in terms of image reconstruction; Experience (MIDI + AMBER) What is the status of “disk clearing” in the inner few AU?

Sublimation radius ~ 0.1-1AU (TTauri HAe/Be stars) but: Observations: Significant dust depletion >> Sublimation Radii TW Hydrae (10Myr): ~ 4 AU (Calvet et al. 2002) GM Aur: ~ 4 AU (Rice et al. 2003) CoKu Tau/4: ~10 AU (D’Alessio et al. 2005, Quillen et al. 2004)

10μm image of a circumstellar disk with an inner hole, radius 4AU (inclination: 60deg; distance 140pc; inner 60AU x 60AU) Constraints on a planetary origin for the gap in the protoplanetary disc of GM Aurigae

increasing planetary mass

Azimuthally averaged mid-plane density The SED of GM Aur computed using azimuthally profiles for substellar objects (planets). averaged density profiles.

•A ~ 2 MJ planet, orbiting at 2.5 AU in a disk with mass 0.047 M and radius 300 AU, provides a good match both to the SED and to CO observations which constrain the velocity field in the disc. • A range of planet masses is allowed by current data, but could in principle be distinguished with further observations between 3 and 20 μm. Rice et al. (2003) Imaging in the Near- infrared

Jupiter @ 5AU Solar-type central star

2.2 micron (scattered AB Aurigae - Spiral arm structure stellar light) (Herbig Ae star; H band; Fukagawa, 2004) Young disks Inner disk Which disks to study? (< a few AU)

Preparatory studies, Etc. ... concentrating on face-on disks Clearly identified disks, well studied, but … potentially ”planet-building sites” well hidden… Useful techniques: Coronography; Differential polarimetric imaging;

Very distant … high-resolution mm maps AB Aurigae HD 100546

(Grady 2001 / 2003) Influence on the inner ~12 AU Net - SED

Inner Disk

Wolf & D’Angelo (2005) Influence on the inner ~12 AU Net - SED

Planet

Wolf & D’Angelo (2005) Influence on the inner ~12 AU Net - SED

Planetary Environment

Wolf & D’Angelo (2005) Influence on the inner ~12 AU Net - SED

Inner Disk + Planet + Planetary Environment

No significant effect on the Net SED Wolf & D’Angelo (2005) Influence on the Net - SED

Planetary Contribution (direct or scattered radiation, dust reemission) (depending < 0.4% on the particular Disk reemission (inner 12 AU) model)

Planetary radiation significantly affects the dust reemission SED only in the near to mid-infrared wavelength range. This spectral region is influenced also by the warm upper layers of the disk and the inner disk structure, the planetary contribution. => The presence of a planet + the temperature / luminosity of the planet cannot be derived from the SED alone.

In the case of a more massive planet / star the influence of the planet is even less pronounced in the mid-infrared wavelength range (lower luminosity ratio LP / L*).

see also Varniere et al. (2006) High-Spatial Resolution Spectroscopy

Spectroscopic verification of gaps through their dynamics R ~ 105, λ ~ 5-30μm

GSMT Science Case Study [ www.aura-nio.noao.edu ] MHD simulations: GapsGaps Internal Stress arises self-consistently from turbulence generated by magnetorotational inin youngyoung disksdisks instability (‚MHD turbulence‘) -- >>> gaps are shallower and asymmetrically wider MMHDHD >>> rate of gap formation is slowed simulationssimulations

(Winters et al. 2003; see also Nelson & Papaloizou 2003) Sources of Astrometric Wobble

1. Planet’s Gravitational Pull

2. Disk’s Gravitational Pull

3. Disk’s Photospheric Signal (center-of-light wobble)

(G. Bryden, priv. comm.) Center-of-Light-Wobble

(G. Bryden, priv. comm.) Protoplanetary Disks evolve …

• Near-infrared photometric studies: sensitive to the inner ~ 0.1 AU around solar-type stars: • Excess rate decreases from ~80% at an age of ~1 Myr to about 50% by an age of ~3 Myr (Haisch et al.~2001) • By ages of ~10-15 Myr, the inner disk has diminished to nearly zero (Mamajek et al.2002).

• Far-infrared / millimeter continuum observations probe the colder dust and thus the global dust content in disks: • Beckwith et al. (1990): no evidence of temporal evolution in the mass of cold, small (<1mm) dust particles between ages of 0.1 and 10Myr • By an age of 300 Myr the dust masses were found to by decreased by at least 2 orders of magnitude (Zuckerman & Becklin 1993). • Based on studies with the Infrared Space Observatory (ISO), the disk fraction amounts to much less than 10% for stars with ages > 1 Gyr (e.g., Spangler et al. 2001; Habing et al. 2001; Greaves et al. 2004; Dominik & Decin 2003). … but still the disk may outshine the planet.

• The exozodiacal dust disk around a target star, even at solar level, will likely be the dominant signal originating from the extrasolar system: • Solar system twin: overall flux over the first 5 AU is about 400 times larger than the emission of the Earth at 10μm

• Zodiacal light of our own solar system: • potential serious impact on the ability of space-born observations (e.g. DARWIN) • attributed to the scattering of sunlight in the UV to near-IR, and the thermal dust reemission in the mid to far-IR • > 1micron: signal from the zodiacal light is a major contributor to the diffuse sky brightness and dominates the mid-IR sky in nearly all directions, except for very low galactic latitudes (Gurfil et al. 2002). betaPic Young disks / Debris disks Planet Ù Disk interaction IRAS 04302+2247

AUMic

Young circumstellar Debris disks disks around T Tauri / HAe/Be stars

optically thick optically thin

HK Tau

Density structure dominated by BD+31643 Gravitation +

Radiation Pressure Gas dynamics Poynting-Robertson Scattered light images (optical) effect CharacteristicCharacteristic DebrisDebris DiskDisk DensityDensity PatternsPatterns

A planet, via resonances and gravitational scattering produces

[1] An asymmetric resonant dust belt with one or more clumps, intermittent with one or a few off- center cavities, and

[2] A central cavity void of dust. Simulated surface density of circumstellar dust captured into particular mean motion resonances (Ozernoy et al. 2000) [1] Location Resonant structures can serve as indicators of a planet in a [2] Major orbital parameters circumstellar disk [3] Mass of the planet

(static) equilibrium density distribution Scattered Light Image 104 particles 107 particles dynamical simulation •~ 104 massless particles • gravitational intercation with planet + star Rodmann, Wolf, • Poynting-Robertson effect + Radiation Pressure []Spurzem, Henning, in prep. [1] Location Resonant structures can serve as indicators of a planet in a [2] Major orbital parameters circumstellar disk [3] Mass of the planet

(static) equilibrium density distribution Scattered Light Image 104 particles 107 particles

Relative brightness distribution of individual clumps in optical to near-infrared scattered light images may sensitively depend on the disk inclination. Debris diskaroundVega Debris diskaroundVega

Dust reemission

(Wilner et al. 2002)

(Holland et al. 1998, Wilner et al. 2002) SOFIA, JWST Debris diskaroundVega Debris diskaroundVega

24μm

Su et al. (2005): • No clumpy structure • Inner disk radius: 11”+/-2” • Extrapolated 850μm flux << than observed 70μm • Explanation: Dust(Holland grains et al. of1998, different Wilner et al. 2002)sizes Spitzer are traced by Spitzer/SCUBA Giant Planets in Debris Disks

Characteristic Asymmetric Density Patterns

Rodmann & Wolf (2006)

Decreased Mid-Infrared Spectral Energy Distribution (but dust grain evolution makes detailed SED analysis difficult)

Wolf & Hillenbrand [ aida28.mpia.de/~swolf/dds ] (2003, 2005) Giant Planets in Debris Disks

Characteristic Asymmetric Density Patterns

Rodmann & Wolf (2006)

Decreased Mid-Infrared Spectral Energy Distribution (but dust grain evolution makes detailed SED analysis difficult)

Wolf & Hillenbrand [ aida28.mpia.de/~swolf/dds ] (2003, 2005) Giant Planets in Debris Disks

Characteristic Asymmetric Density Patterns

Rodmann & Wolf (2006)

Decreased Mid-Infrared Spectral Energy Distribution (but dust grain evolution makes detailed SED analysis difficult)

Wolf & Hillenbrand [ aida28.mpia.de/~swolf/dds ] (2003, 2005) Inner cavity in an optically thin disk surrounding a solar-type star Fe content GapsGaps SEDSED

Inner disk radius Grain size T Tauri Disks GM Aurigae, TW Hya [Koerner et al. 1993, Rice et al 2003, Calvet et al. 2002]

Debris Disks β Pic (20AU), HR 4796A (30- 50AU), ε Eri (50AU), Vega (80AU), Fomalhaut (125AU) + „new“ Spitzer Debris Disks [Dent et al. 2000, Greaves et al. 2000, Wilner et al. 2002, Holland et al. 2003]

Wolf & Hillenbrand (2003) TheThe YoungYoung SolarSolar SystemSystem @@ 50pc50pc

-10 Mdust = 10 Msun

with planets

without planets

Moro-Martin, Wolf, & Malhotra (2004) Some problems with SEDs...

Kim et al. 2005 Some problems with SEDs...

Many of the debris disks observed with the Spitzer ST, show no or only very weak emission at wavelengths < 20…30 micron (e.g. Kim et al. 2005) => No / weak constraints on the chemical composition of the dust

Debris disks: Optically thin

- azimuthal and vertical disk structure can not be traced via SED observations / modelling; - only constraints on radial structure can be derived: SED = f ( T(R) )

but even here ambiguities are difficult to resolve … Imaging required

Moro-Martin, Wolf, & Malhotra (2005)

on an inclined orbit Pictoris Disk evolution Warp in β Model includes a Disk of Planetesimals Extending out to 120-150AU, the produced through collisional perturbed gravitationally by a • Source of a distribution grains giant planet (Augereau et al. 2001, see also Mouillet et al. 1997) see also Mouillet (Augereau et al. 2001, •

(Schultz, Heap, NASA 1998)

Holland et al. 1998, Dent et al. 2000, Pantin 1997) few tens ... lunar masses density distribution located between 80AU and 100AU (Zuckerman & Becklin 1993, • Maximum of the dust surface • Orientation nearly edge-on : • mass Total : Pictoris dust disk: β Concluding Remarks

1. SED: (sub)mm slope 2. Shape of 10μm silicate feature 3. Scattered light polarization 4. Multi-wavelength imaging 5. Vertical Disk Structure Concluding Remarks

Vortices => Local Density Enhancements => enhanced grain growth

(e.g., Wolf & Klahr 2003, Klahr & Bodenheimer 2006) Concluding Remarks

1. Gaps 2. Global Spiral Structures 3. Planetary Accretion Region 4. Center-of-light-wobble 5. Inner holes

SIM Concluding Remarks

1. Characteristic Asymmetric Patterns 2. Shape of the mid-infrared Spectral Energy Distribution 3. Warps (β Pic) Concluding Remarks

Theoretical investigations show that the planet-disk interaction causes structures in circumstellar disks, which are usually much larger in size than the planet itself and thus more easily detectable. The specific result of the planet-disk interaction depends on the evolutionary stage of the disk. Numerical simulations convincingly demonstrate that high-resolution imaging performed with observational facilities which are already available or will become available in the near future will allow to trace these signatures of planets. These observations will provide a deep insight into specific phases of the formation and early evolution of planets in circumstellar disks.