High Contrast Imaging Extreme AO 30 m Telescopes

High Contrast Imaging Extreme AO & 30 -m Telescopes James R. Graham UC Berkeley 2005/02/16 1

High Contrast Imaging • Solar observations with a Lyot coronagraph SOHO C 3 coronagraph • SOHO • Coronal mass ejections & sun-grazing comets • Planet detections! 16° 2 http: //sohowww. nascom. nasa. gov

High Contrast Imaging • Stellar coronagraphs Smith & Terrile 1984 Science 226 1421 • Discovery of scattered light disk— Pictoris • Brown dwarfs—GD 229 B Nakajima et al. 1995 Nature 378 463 3

State of the Art • Fomalhaut debris disk F 606 W + F 814 W HST/ACS coronagraph Kalas Clampin & Graham 2005 Nature, Submitted – µ ≈ 20 mag arc sec-2 – µ/µ 0 ≈ 10 -10 • Hard-edged Lyot coronagraph – Contrast is limited by quasi-static wavefront errors • Speckle noise 4

Utility of High Contrast Imaging • Broad potential scientific application – Exoplanet detection – Circumstellar disks • Proto-planetary & debris disks – Fundamental stellar astrophysics • Stellar binaries – Mass transfer & loss • Cataclysmic variables, symbiotic stars & supergiants – Solar system: icy moons, Titan, & asteroids 5

Exoplanet Science • Doppler surveys have cataloged 137 planets – Indirect searches are hindered by Kepler’s third law • PJupiter = 11 years • PNeptune = 165 years • A census of the outer regions of solar systems (a > 10 AU) is impractical using indirect methods • 1/r 2 dimming of reflected light renders TPF-C insensitive to planets in Neptune orbits • Ex. AO is sensitive to self-luminous planets with semimajor axes 4– 40 AU 6

Architecture of Planetary Systems • 137 Doppler exoplanets – 5% of targeted stars possess massive planets – Lower limit on occurrence of planets – Abundance of solar systems—why isn’t it 15 to 50%? • A diversity of exoplanet systems exist… • ≤ 20% of the solar system’s orbital phase space explored – Is the solar system typical? • Concentric orbits & radial sorting – What are the planetary systems of A & F stars? – How do planets form? What dynamical evolution occurs? • Core accretion vs. gravitational collapse • Planetary migration • Doppler surveys raise new questions – What is the origin of exoplanet dynamical diversity? 7

Architecture of Planetary Systems • Direct imaging is “instant gratification” – Fast alternative to Doppler surveys • Improved statistics (4– 40 AU vs. 0. 4– 4 AU) – Worst case, d. N/d log(a) ~ const. – Oligarchy, d. N/d log(a) ~ a – Searching at large semimajor axis • Sample beyond the snow line • Characterize frequency & orbital geometry > 4 AU – Is the solar system is unique • Reveal the zone where planets form by gravitational instability (30– 100 AU) • Uncover traces of planetary migration – Resolve M sin(i) ambiguity 8

Cooling Planets • Contrast required to detect a cooling planet is much less in the near-IR than in the visible – Radiation escapes in gaps in the CH 4 and H 2 O opacity at J, H, &, K Burrows Sudarsky & Hubeny 2004 Ap. J 609 407 9

What is Ex. AO • How can we achieve contrast Q < 10 -7? • Control of wavefront errors – Wavefront errors, , cause speckles which masquerade as planets • 2 ≈ (Q/16) D 2 [ 22 - 12] on spatial frequencies 1/ < f < 2/ • = 3 nm rms for Q = 10 -7 between 0. ” 1 < < 1” (30 cm to 300 cm) • Control of diffraction – Need AO & a coronagraph because wavefront errors and diffraction couple 10

Wavefront & Diffraction Control 64 /D • Focal plane simulations for Gemini Ex. AO at H – The dark hole shows the control radius /2 d • Increasing contrast due to suppression of speckle pinning Circular pupil Lyot coronagraph APLC Remi Soumier 11

It’s Not About Strehl • 70 nm RMS dynamic wavefront error 0 nm 2 nm – S = 0. 93 • 0 , 2, & 4 nm RMS static wavefront error – Strehl ratios differ by less than 10 -4 – Systematic errors prevent detection of the exoplanet 5 MJ 1 Gyr exoplanet • Atmosphere has ‹ ›=0 – Not crazy to do this from the ground 4 nm 12 Bruce Macintosh

Ex. AO Science on 8 -m Telescopes • Ex. AOC on 8 -m telescopes can yield the first detections of self -luminous exoplanets 13

Ex. AO Science on 8 -m Telescopes • Probe beyond the snow line – Complementary to Doppler & astrometric searches Doppler 8 -m Ex. AO 14

Ex. AO Science on 8 -m Telescopes Ag e H 2 O NH 3 Ma ss • First reconnaissance of planetary atmospheres Jupiter 15 Ex. AO T dwarfs

8 -m vs. 30 -m • Better angular resolution • Better contrast HST – For a given rms wavefront error budget (on fixed spatial scales) Gemini Ex. AOC TMT? Jovian reflected light • TMT can’t lock on fainter guide stars! TPF-C? 2 = 1. 0 arc sec 1 = 0. 1 arc sec 16

TMT Science: What 8 -m’s Can’t Do • Detect Doppler planets – /D is too big to find planets in 5 AU orbits – Inner working distance of TMT is three times smaller • Reflected light Jupiters – Q ≈ 2 x 10 -9 (a/5 AU)-2 – TMT could make old, cold planets a priority – Redundant with TPF-C and indirect searches? 17

TMT Science: What 8 -m’s Can’t Do • Explore star forming regions – Taurus, Ophiuchus &c. are too distant – TMT can work into 5 AU • Intermediate contrast Q ≈ 10 -6 at increased angular resolution (10 mas at H) is valuable – Planet forming environment – Evolved stars and stellar mass loss 18

TMT Science: What 8 -m’s Can’t Do • Astrometry – Detection of exoplanet orbital acceleration requires astrometric precision of about 2 mas (about 1/10 of a pixel for an 8 -m) – Ultimate goal is to measure Keplerian orbital elements, especially e – Angular resolution of TMT is major benefit for TMT • Spectroscopy of exoplanet atmospheres – Rudimentary Teff , log (g) measurements at R ≈ 40 are feasible with an 8 -m – TMT can study composition of exoplanet atmospheres, especially important to understand the condensation of H 2 O and NH 3 clouds 19

The Path to Ex. AO TMTs • 104 actuator deformable mirrors • 5122 fast (k. Hz), low noise (few e-) CCDs • Fast wavefront reconstructors – FFT algorithms • Segment errors & discontinuities must be factored into the wavefront error budget – Discontinuities are OK, so long as the wavefront sensor is band-limited – AO controls wavefront errors, but not diffraction – Unobscured, filled aperture is ideal… • Large gaps render apodization problematic • Uniform reflectivity 20