EPICS, exoplanet imaging with the E-ELT

1
EPICS, exoplanet imaging with the E-ELT

• Raffaele G. Gratton,Markus Kasper (PI), Jean-Luc
  Beuzit, Christophe Verinaud, Emmanuel
  Aller-Carpentier, Pierre Baudoz, Anthony
  Boccaletti, Mariangela Bonavita, Kjetil Dohlen,,
  Norbert Hubin, Florian Kerber, Visa Korkiaskoski,
  Patrice Martinez, Patrick Rabou, Ronald
  Roelfsema, Hans Martin Schmid, Niranjan Thatte,
  Lars Venema, Natalia Yaitskova
• ESO, LAOG, LESIA, FIZEAU, INAF-Osservatorio
  Astronomico di Padova, ASTRON, ETH Zürich,
  University of Oxford, LAM, NOVA

1
2
Outline

• Science goals
• Instrument and AO concept
• Science Output prediction

3
Exoplanets observations 2009

• Close to 400 Exoplanets detected, gt80 by radial
  velocities, mostly gas giants, a dozen Neptunes
  and a handful of Super-Earths

• Constraints on Mass function, orbit distribution,
  metallicity
• Some spectral information from transiting planets

3
4
(Some) open issues

• Planet formation (core accretion vs gravitational
  disk instability)
• Planet evolution (accretion shock vs spherical
  contraction / hot start)
• Orbit architecture (Where do planets form?, role
  of migration and scattering)
• Abundance of low-mass and rocky planets
• Giant planet atmospheres

4
5
Object Class 1, young self-lumPlanet formation
5
6
Object Class 2, within 20 pcOrbit architecture,
low-mass planet abundance
500 stars from Paranal 30 deg, 60-70
M-dwarfs

• Requirements
• High contrasts10-9 at 250 mas (Jupiter at
  20pc)
• spatial resolution 10-8 at 40 mas (Gl
  581d,8 M?)

6
7
Object Class 3, already known onesPlanet
evolution and atmospheres

• discovered by RV, 8-m direct imaging (SPHERE,
  GPI) or astrometric methods (GAIA, PRIMA)

7
8
Contrast requirements summary
9
Concept
9
10
Concept Achieve very high contrast

• Highest contrast observations require multiple
  correction stages to correct for
• Atmospheric turbulence
• Diffraction Pattern
• Quasi-static instrumental aberrations

NIR diffraction suppression
x 1000 !
11
XAO concept

• Main parameters (baseline)
• Serial SCAO, M4 / internal WFS, XAO
• XAO roof PWS at 825 nm, 3 kHz
• 200x200 actuators (20 cm pupil spacing)

Advanced RTC algorithms studied in parallel to
EPICS phase-A
Simulation by Visa Korkiakoski
11
AO coro
12
High Order Testbench (HOT)Demonstrate XAO / high
contrast concepts

• Developed at ESO in collaboration with Arcetri
  and Durham Univ.
• Turb. simulator, 32x32 DM, SHS, PWS,
  coronagraphy, NIR camera
• H-band Strehl ratios 90 in 0.5? seeing (SPIE
  2008, Esposito et al. Aller-Carpentier et al. )
  correcting 8-m aperture for 600 modes

Aller-Carpentier, proc. AO4ELT
13
HOT XAO with APL coronagraph
Martinez et al., submitted to AAL

• Good agreement with SPHERE simulations
• Additional gain by quasi-static speckle
  calibration (SDI, ADI)

14
Residual PSF calibration

• From systematic PSF residuals (10-6-10-7) to
  10-8-10-9
• Spectral Deconvolution (SparksFord, Thatte et
  al.), Trade-off spectral bandwidth vs inner
  working angle,? IFS (baseline Y-H)
• Polarimetric differential imaging for smallest
  separation, needs planet feature (e.g. CH4
  band, or polarization)? EPOL (600-900 nm)
• Coherence based methods (speckles interfere with
  Airy Pattern, a planet does not) ? Self-Coherent
  camera (Baudoz et al, Proc. AO4ELT)
• Angular Differential Imaging (ADI) ? All

15
Example Spectral Deconvolution
15
16
Speckle chromaticity and Fresnel
SD needs smooth speckle spectrum -gt near-pupil
optics
20 nm rms at 10x Talbot
20 nm rms in pupil plane
17
End-2-end analysis

• Apodizer only leads to improved final contrast

APLC
18
Baseline Concept
All optics near the pupil planeminimize
amplitude errors and speckle irregular
chromaticity
19
Detection limits, incl. photon noise
20
Detection rates, MC simulation
20
21
Predicted Science Output

• MC simulations
• planet population with orbit and mass
  distribution from Mordasini et al. (2009)
• Model planet brightness (thermal, reflected,
  albedo, phase angle,)
• Match statistics with RV results

• Contrast model
• Analytical AO model incl. realistic error budget
• Spectral deconvolution
• No diffraction or static WFE
• Y-H, 10 throughput, 4h obs

22
Detection rates, nearbyyoung stars
Contrast requirements
Simulations by M. Bonavita
23
Predicted EPICS output
Target class targets Self-luminous planets Giant planets Neptunes Rocky planets
1. Young stars 688 100 (100) Dozens Very few (?)
2. Nearby stars 512 Dozen 100 Dozens Dozen
3. Stars w. planets gt100 Some gt100 gtDozen gt2
23
24
Summary

• EPICS is the NIR E-ELT instrument for Exoplanet
  research
• Phase-A to study concept, demonstrate feasibility
  by prototyping, provide feedback to E-ELT and
  come up with a development plan
• Conclusion of Phase-A early 2010
• Exploits E-ELT capabilities (spatial resolution
  and collecting power) in order to greatly advance
  Exoplanet research (discovery and
  characterization)