Signatures of Planets in Debris Disks A. Moro-Martin 1,2,3, S. Wolf 2, R. Malhotra 4 & G. Rieke 1 1. Steward Observatory (University of Arizona); 2. MPIA (Heidelberg); 3. Princeton University; 4. LPL (University if Arizona) 1. Steward Observatory (University of Arizona); 2. MPIA (Heidelberg); 3. Princeton University; 4. LPL (University if Arizona)

Outline Introduction: What is a debris disk? How its structure is created? What can it tell us about massive planets? What can we do with Spitzer: …and what we cannot do: SEDs are degenerated …and what we cannot do: SEDs are degenerated Why we need ALMA and Herschel: We need high resolution imaging: - To interpret debris disk structure. - To interpret debris disk structure. We need high sensitivity observations: - To understand how debris disk are created (stochastic collisional events? or slow and constant grinding down of planetesimals?) - To understand how debris disk are created (stochastic collisional events? or slow and constant grinding down of planetesimals?) - To study timescales and frequency of terrestrial planet formation. - To study timescales and frequency of terrestrial planet formation.

Dust is not primordial but must be “continuously” (?) replenished by a reservoir of undetected planetesimals (of unknown mass) producing dust by mutual collisions Dust is not primordial but must be “continuously” (?) replenished by a reservoir of undetected planetesimals (of unknown mass) producing dust by mutual collisions Many (>15%) MS stars are surrounded by debris disks: cold far-IR emitting dust (1-10M ) that reprocesses star light and emits at longer ’s. Debris disks are indirect evidence of planetary formation: Dust Removal Time Scales: Age of Star Poynting-Robertson drag ~ 10 5 yrs Poynting-Robertson drag ~ 10 5 yrs > 10 7 yrs Introduction << KB dust disk = M

As dust particles spiral inward (due to PR drag), they can get trapped in Mean Motion Resonances with the planets. I.e. massive planets shepherds the dust grains in the disks. Radial and azimuthal structure Massive planets may scatter and eject dust particles out of a planetary system Gaps Gaps To induce frequent mutual collisions the planetesimals’ orbits must be dynamically perturbed by massive planetary bodies. Do debris disks harbor massive planets? Log[Number] Semimajor Axis (AU)

- ring-like structure along Neptune’s orbit (MMRs) - minimum at Neptune’s position (avoid resonant planet) clearing of dust from inner 10 AU (due to gravitational scattering by Jupiter and Saturn) Neptune with Solar System planets Example: Kuiper Belt Dust Disk Effect 1 without planets Effect 2 24  m 70  m 24  m

 -Eri 850  m (emitted light; Greaves et al. 98) HR4796A 1.6  m (scattered light; Schneider et al. 99) H  m (scattered light; Weinberger et al. 99) Gaps and asymmetries observed in high-resolution observations suggest giant planets may be present. Structure is sensitive to long period planets complementary to radial complementary to radial velocity and transit surveys. velocity and transit surveys. We can learn about the diversity of planetary systems from the study of debris disks structure! Needed to determine stability of orbits in habitable zones (TPF)

What can we do with Spitzer? Very few systems will be spatially resolved: in most cases we won’t be able to look for planets by studying debris disk structure directly. in most cases we won’t be able to look for planets by studying debris disk structure directly. But the structure carved by the planets can affect the shape of the Spectral Energy Distribution (SED) of the disk maybe we can study the debris disk structure indirectly (FEPS)

Log[F(mJy)] 1 M Jup at 1 AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 3 M Jup at 1 AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 10M Jup at 1 AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] No planet star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 1 M Jup at 5 AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 3 M Jup at 5 AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 10M Jup at 5 AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] No planet star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 1 M Jup at 30AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 3 M Jup at 30AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 10M Jup at 30AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] No planet star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 3 M Jup at 1 AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 3 M Jup at 5 AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] 3 M Jup at 30AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

Log[F(mJy)] No planet star 1AU5AU30AU Log[  m)] 50AU planetesimals Carbonaceous grains Fe-rich silicate grains Fe-poor silicate grains

The SED of a dust disk generated by an outer belt of planetesimals with inner planets is fundamentally different from that of the disk without planets. Significant decrease of the near/mid-IR flux due to the clearing of dust inside the planet’s orbit. It may be possible to diagnose the location of the planet and the absence/presence of planets What can we learn from the SEDs? It is difficult to diagnose the mass of the planet but…

Log[F(mJy)] 3 M Jup at 1 AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Fe-poor silicate grains 70  m 24  m 10AU 1’’ at 10 pc There are degeneracies that can only be solved with high-resolution observations…

Log[F(mJy)] 3 M Jup at 30AU star 1AU5AU30AU Log[  m)] 50AU planetesimals Fe-rich silicate grains There are degeneracies that can only be solved with high-resolution observations… 10AU 70  m 24  m 1’’ at 10 pc

How ALMA and Herschel can help We need ALMA high resolution imaging to interpret debris disk structure in terms of planetary architectures. 1M Jup 1AU1M Jup 30AU1M Jup 5AU Solar System No planets 70  m

1M Jup 1AU1M Jup 5AU1M Jup 30AU Solar System No planets 24  m

But ALMA alone is not enough because… But ALMA alone is not enough because… Warm dust (i.e. “zodiacal light”) produced by asteroid-like bodies in terrestrial planet region may be invisible for ALMA (low  ; too diffused) Vega observations with Spitzer show that mm and far-IR observations are very different from each other, and both need to be considered in the interpretation of this systems.

We need Herschel high sensitivity to understand the physical processes giving rise to debris disks, whether… Steady production of dust during long periods of time (“standard scenario”). or… or… Stochastic collisional events [new Spitzer results; and observations of debris disks around old stars (few Gyrs)]. The high sensitivity of will allow us to make deeper surveys of stellar clusters of known age, to study the number of stars with recent collisional events as a function of time.

Is the “late bombardment” epoch in the early Solar System common among other stars? Is its intensity below or above average? Consequences for the survival of Life in the terrestrial planets in the habitable zone. Terrestrial planet formation should produce a clear IR signal. High sensitivity surveys will allow us to: Study timescales for terrestrial planet formation. Estimate whether they are common or rare. Observation at different ’s will allow us to tell where the action is taking place. Questions ALMA and Herschel will help to answer….

Conclusions Conclusions Massive planets create structure in debris disks. Structure is sensitive to long period planets. Radial velocity and transit surveys are not enough to study the diversity of planetary systems. Radial velocity and transit surveys are not enough to study the diversity of planetary systems. ALMA high resolution observations are fundamental for the interpretation of structure in terms of planetary architectures Spitzer is not enough because SEDs are degenerated. Herschel high sensitivity observations are needed because mm and far-IR give complementary information (e.g Vega). High sensitivity surveys will allow us to study: Whether dust is produced in stochastic collisional events, or in a slow grinding down of planetesimals. Timescales of terrestrial planet formation. Are terrestrial planets common or rare? Are other planetary systems more hostile to Life?

For details about the modeling: “Study of the Dynamics of Dust from the Kuiper Belt: Spatial Distribution and Spectral Energy Distribution”, Moro-Martin & Malhotra, 2002, AJ, 124, 2305 “Dynamical models of KB Dust in the Inner and Outer Solar System”, Moro-Martin & Malhotra, 2003, AJ, 125, 2255 “Dust outflows from planetary systems”, Moro-Martin & Malhotra, 2004, submitted to ApJ “Model Spectral Energy Distributions of Circumstellar Debris Disks. II. Outer Belt of Planetesimals with Inner Giant Planets”, Moro-Martin, Wolf & Malhotra, 2004, submitted to ApJ Pre-prints at:

Dynamical Models (3D) Radiative Transfer Models (1D) Dust-Producing Planetesimals Orbits Planet’s orbits M star L star Radial Density Distributions Final Disk SED Single Grain Size, Single Composition Disk SED Disk Mass Grain Sublimation Temperature Stellar SED Grain radius Grain Optical Constants 4. Power Law Distribution of Particle Sizes Combine Different Grain Chemical Compositions  grain radius Mie Theory  =5.7x10 -5 Q pr /  s (for Sun)  ratio of radiation pressure force and gravitational force. Q pr : radation pressure coeff. ,s: grain density and radius This is a self-consistent approach, as he disk density distribution can neither be defined independently from the SED of the embedded star nor the dust grain properties This is a self-consistent approach, as the disk density distribution can neither be defined independently from the SED of the embedded star nor the dust grain properties

Radial Density Distributions Log[Surface Density] Log[r(AU)] without planets with planets  =

Radial Density Distributions Log[Surface Density] Log[r(AU)] without planets with planets  =0.4

Log[Number] Semimajor Axis (AU) without planets with planets  = Accumulation of particles in exterior MMRs with planet Clearing inside the planet’s orbit Effect 2 Effect 1

 Grain Radius Relation 2 Log[  ] Log[a(  m)] Models run Amorphous Silicates Crystal. Silicates Amorphous Carbon

Single Grain Size, Single Composition Disk SED C400 MgFeSiO 4 Mg 1.9 Fe 0.1 SiO 4 C1000 Mg 0.6 Fe 0.4 SiO 3 MgSiO 3 Small grains Intermediate grains Large grains

List of planetary systems with distinct Spitzer colors

Final Disk SED