A Starshade for JWST: Science Goals and Optimization

Rémi Soummer (STScI) November 12th, 2009

and the New Worlds Probe team W.Cash et al.

Thursday, November 12, 2009 etnsaentdi re.TFpbiain r hw ndr le h hnigPoetfnigi hw ythe by shown is funding Important Project red. changing http://olbin.jpl.nasa.gov/news/tpf in The at highlighted available is blue. Also technology dark ﬁgure. and in the ideas shown of New are background lower publications black. the TPF in in shown green. histogram are blue in events noted Programmatic are meetings (left). top the at 1. Figure

Bill Clinton George W. Bush Barack Obama h itr fTFfo 99t 09i hw satmln.Peiet n AAAmnsrtr r shown are Administrators NASA and Presidents timeline. a as shown is 2009 to 1999 from TPF of history The Dan Goldin Sean O’Keefe Mike Gri!n Ed Weiler Al Diaz Mary Cleave Alan Stern Ed Weiler Anne Kinney Rick Howard Jon Morse President War in Afghanistan War in Iraq G. W. Bush announces Hurricane Katrina Sub-prime October 2001 March 2003 August 2005 New Vision for NASA Mortgage Meltdown August 2008 TPF Foundation Science 14 January 2004 Adv. Cryocooler Technology Develop. Program TPF-C Instrument Concept Studies NASA Releases NASA Astrophysics Strategic April 2000 October 2002 May 2005 The Vision for Mission Concept Studies Origins Roadmap 2000 Origins Roadmap 2003 Strategic Roadmap #4 Space Exploration JPL 400-887 JPL 400-1060 The Search for February 2004 TPF/Darwin Planning February 2008 Earth-like Planets Noordwijk Worlds Beyond: NASA Releases NASA announces 19-21 April 2006 Report of the ExoPlanet Columbia Accident TPF-C and TPF-I NRC Panel Review Task Force Prelim. Architecture Final Architecture StarLight Investigation Board as a separate of TPF Science NASA/ESA Letter of ESA Cosmic Vision Decadal Review TPF Book Review Review Cancelled Report Vol. 1 missions Requirements Agreement Expires Proposals Due RFI Responses Due May 1999 12-14 Dec 2000 11-13 Dec 2001 February 2002 August 2003 April 2004 Sept 2004 Dec 2006 June 2007 April 2009

Decadal Review Concept Studies Interferometer or CoronagraphTPF-C and TPF-I Deferred Inde!nitely Decadal Review 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

Darwin and Astronomy Darwin/TPF Meeting Darwin/TPF TPF/Darwin Meeting IAU Colloq. 200 Spirit of Lyot March 2009 Stockholm, Sweden Heidelberg, Germany Saas-Fee San Diego, USA Villefrance-sur-mer, France Berkeley, USA Exoplanet 17-19 Nov 1999 June 2002 22-25 April 2003 9-12 Feb 2004 26-29 July 2004 3-7 Oct 2005 June 2007 Community Report January 2001 Biosignatures and JPL Pub 09-3 A New Pupil for Planetary Properties April 2003 Coronagraph SPIE Exoplanets II SPIE Exoplanets III Extrasolar Planets to be Investigated Phase-Induced Workshop San Diego, USA San Diego, USA August 2005 August 2007 Feb 2009 D. Spergel by the TPF Mission Leiden Amplitude 2-6 Feb 2004 June 2006 TPF-I Milestone #3 astro-ph/0101142 JPL Pub 01-08 Apodization Oct 2004 Terrestrial Planet March 2007 Broadband Nulling March 2005 O. Guyon Precursor Science Finder Coronagraph Terrestrial Planet Demonstration Technology Plan February 2001 Summary Report on for the Terrestrial Science and Technology Finder Interferometer TPF Science for the Terrestrial Apodized Square Aperture Architecture Studies April 2003 Planet Finder De"nition Team Science Working Tech & Design Planet Finder P. Nisenson & C. Papaliolios for the Terrestrial Adaptive Nulling JPL Pub 04-14 (STDT) Report Group Report Jan 2008 Expo Coronagraph ApJ 548, L201 (2001) Planet Finder O. Lay, JPL Doc 34923 JPL Pub 07-1 TPF-I Milestone #2 Pasadena JPL Pub 05-08 JPL Pub 02-11 M. Jeganathan, 14-16 Oct 2003 Formation Control & R. Peters Demonstration April 2001 January 2005 June 2005 July 2007 Sept 2006 Symmetric Nulling General Astrophysics Technology Plan TPF-I Milestone #1 May 2002 March 2003 Coronagraph Beam Combiners Technology Plan and Comparative for the Terrestrial Adaptive Nuller Band-limited masks Workshop E. Serabyn & M. M. Colavita for the Terrestrial Planetology with the Planet Finder Demonstration M. Kuchner & W. Traub JPL Pub 05-01 Appl. Opt. 40, 1668 (2001) Planet Finder Terrestrial Planet Finder Interferometer ApJ 570, 900 (2002) Mission JPL Pub 05-05 August 2008 JPL Pub 03-07 June 2006 October 2006 JPL Pub 05-01 Earth-Like TPF-C Milestone #2 TPF-C Milestone #1 Kepler August 2003 Exoplanets: 9 x 10!"# contrast Monochromatic contrast Launch SPIE Exoplanets I The Science of NASA’s 10% bandwidth Demonstration March 2009 San Diego, USA Navigator Program July 2006 JPL Pub 06-5 Petal-shaped Occulter Spitzer W. Cash Corot SPIE Meeting SPIE Meeting Launch SPIE Meeting SPIE Meeting Launch Munich, Germany Waikoloa, USA August 2003 Glasgow, UK SPIE Meeting Marseille, France March 2000 August 2002 June 2004 Orlando, Florida Dec 2006 June 2008

timeline.pdf. May 2006

!""#$%&'()*&$"+*,-&-.-/"('"0/!1*(%(23""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""4(5/)*6/*-"78(*,(),1&8"9!:*(;%/<2/<="""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""""">/-/)"?="@$;,(*"""""""""""""AB"C.%3"DBBE

Thursday, November 12, 2009 Context

• Terrestrial Planet Finder • NASA is interested in new strategies consistent with current budget ‣ Mission concept studies ‣ Exoplanet Community Report ‣ Possibility of a “medium-class” mission (~$700M)

• Discovery proposal 3 years ago (Cash et al.)

• Medium mission ‣ “small coronagraph” e.g. PECO 1.4 m ‣ occulter possible if host telescope available ‣ small interferometer e.g. FKSI

Thursday, November 12, 2009 New Worlds Probe (NWP) Concept

~50,000-70,000km

50-70 m diameter

‣ Science capabilities given JWST instruments (what can we detect? biomarkers?) ‣ Observing time, available stars, Field of Regard, DRM ‣ Operations (alignment, planning & scheduling, target acquisition, overheads) ‣ Starshade itself

Thursday, November 12, 2009 External Occulters: early ideas

• Lyman Spitzer, in American Scientist 1962

- “This method involves the use of a large occulting disk far in front of the telescope to reduce the light from the star.” - “In the same way that the diffracted light from a telescope mirror can be reduced by a smooth reduction of the reﬂectivity towards zero at the edge, the shadow behind an occulting disk can be made much blacker if the transparency of the disk varies smoothly at the edge of the disk rather than abruptly; a reduction of intensity within the shadow by an order of magnitude was achieved with the use of an occulting disk edged with sharp spikes”.

Thursday, November 12, 2009 External Occulters: early ideas

• Gordon Woodcock 1974

Thursday, November 12, 2009 External Occulters: early ideas

Thursday, November 12, 2009 External Occulters

Plain Apodized Starshaped occulter occulter

Shadow is not optimum, Much better Star shaped occulter (approximation spot of Arago-Poisson shadow of the continuous apodization)

Umbras Boss

2000~2002

Web Cash: analytical functions, e.g. ‘hypergaussian’

Bob Vanderbei: optimal numerical solutions

Thursday, November 12, 2009 Parameters for the Starshade

• Deﬁne a geometric inner working angle (IWA)

500 1000

1000 1000

500 500 star ‣ IWA ~ D/z 500 1000 planet

‣ Shadow properties (Fresnel propagation) ~ D2/λz • there is a minimum distance & size for a given set of IWA, contrast, bandwidth and other constraints.

• occulter diameter (and distance) increase with ‣ decreasing IWA ‣ telescope diameter ‣ bandwidth (easier on the blue side than red side)

‣ contrast 9

Thursday, November 12, 2009 James Webb Space Telescope

• Observatory ‣ 6.5 m, segmented ‣ modest optical quality (diffraction limited at 2 microns, SR=80%) ‣ Imaging and spectroscopy from 0.6 to 25 microns • Near Infrared Camera (NIRCam) ‣ 2 instruments: short arm (0.6-2.3 μm) and long arm (2.4-5.0 μm) ‣ several ﬁlters, broadband and medium band • Near Infrared Spectrograph (NIRSpec) ‣ 0.6-5.0 μm, R=40-100 (prism) and R=1000

• Mid Infrared Imager (MIRI) • Tunable Filter Imager (TFI)

Thursday, November 12, 2009 NWP: Science Goals Summary

• Can we design an occulter to image and characterize and Earth-like planet? ‣ contrast goal 1e-10

‣ characterize habitability & possible biomarkers, near infrared: O2, H2O, CO2, CH4 ‣ possibility of thermal emission as well? (radius and temperature) • Characterization of giant planets is interesting in the near infrared • Imaging & Spectroscopy from 0.6 to 1.7 μm covers large science program

Turnbull et al. Marley et al.

Thursday, November 12, 2009 NWP: Science Goals Summary

• 1) Find and characterize planetary systems including terrestrial planets in the habitable zone ‣ 20 - 30 nearby extrasolar systems can be observed and mapped

‣ If ηearth is >0.5, there is a high probability of observing ~5 terrestrial planets ‣ With ~3 revisits each, we can characterize their atmospheres, and establish the terrestrial planet’s habitable zone residency. Rotation? Biomarkers? Oxygen? • 2) Characterize known RV planets (Jupiters, Neptune, super Earths) ‣ Mass: disentangle m sin(i) with inclination measurement ‣ Atmospheric composition, gravity, temperature & radius if emitted light can be measured • 3) Determine brightness and structures of exozodiacal disks.

NWP Can Perform a Variety of Exoplanet Science

Thursday, November 12, 2009 Table 1. Known radial velocity planets for which the angular separation is larger than 60mas (here the angular separation is calculated using the semi-major axis). The contrast is calculated for a separation corresponding to the semi-major Large-separationaxis, and assuming quadrature. There areRV 18 planets planets for which the separation is larger than 120mas, respectively 23 for 100mas, 31 for 90mas, 38 for 80mas, 44 for 70 mas, and 57 for 60mas. There are 41 RV planets for which the maximum elongation a(1 + e) > 100mas. Data from The Extrasolar Planets Encyclopaedia;28

Planet Name M sin(i) period sma excentricity separation contrast Epsilon Eridani b 1.6 2502 3.4 0.70 1.059 2.97E-09 GJ 832 b 0.6 3416 3.4 0.12 0.688 2.95E-09 55 Cnc d 3.8 5218 5.8 0.03 0.443 1.02E-09 HD 160691 c 3.1 2986 4.2 0.57 0.273 1.96E-09 Gj 849 b 0.8 1890 2.4 0.06 0.267 6.17E-09 HD 190360 b 1.5 2891 3.9 0.36 0.247 2.22E-09 47 Uma c 0.5 2190 3.4 0.22 0.243 2.97E-09 HD 154345 b 0.9 3340 4.2 0.04 0.232 1.94E-09 Ups And d 4.0 1275 2.5 0.24 0.186 5.41E-09 HD 62509 b 2.9 590 1.7 0.02 0.163 1.19E-08 HD 39091 b 10.4 2064 3.3 0.62 0.160 3.15E-09 14 Her b 4.6 1773 2.8 0.37 0.153 4.44E-09 47 Uma b 2.6 1083 2.1 0.05 0.151 7.66E-09 Gamma Cephei b 1.6 903 2.0 0.12 0.148 8.16E-09 HD 217107 c 2.5 4210 5.3 0.52 0.142 1.23E-09 HD 89307 b 2.7 3090 4.2 0.27 0.126 1.98E-09 HD 10647 b 0.9 1040 2.1 0.18 0.121 7.73E-09 HD 117207 b 2.1 2627 3.8 0.16 0.115 2.39E-09 HD 181433 d 0.5 2172 3.0 0.48 0.115 3.79E-09 HD 70642 b 2.0 2231 3.3 0.10 0.114 3.13E-09 HD 128311 c 3.2 919 1.8 0.17 0.106 1.10E-08 GJ 317 b 1.2 693 1.0 0.19 0.104 3.78E-08 HD 216437 b 2.1 1294 2.7 0.34 0.102 4.68E-09

13 outer working angle, i.e. absence of ﬁeld of view limit for the high-contrast zone, NWP will provide a complete Thursday, Novemberpicture 12, 2009 of the architecture of planetary systems with dust and planets. Observing exozodi is crucial, both for its science return and as a source of background noise for future exoplanet exploration.17 The distribution of the exozodiacal dust is a tracer of the systems orbital dynamics, where planetary orbital resonances create gaps and enhancements in the dust. Tiny planets, too small to be seen directly, should leave distinct marks as well. Imaging the exozodi gives us the inclination of the systems ecliptic plane, which can provide clues to the planets orbit from a single image. Currently known debris disks are 102 to 104 times brighter than the level of an equivalent solar system zodiacal disk.29 Zodiacal and exozodiacal dust create background ﬂux that is mixed with the planet signal in both images and spectra. Even if nearby systems have exozodi levels no greater than the Solar System level, zodiacal and exozodiacal background will be the largest source of noise for most targets. Unfortunately, we know very little about exozodi levels in other systems. Measuring them is crucial to the future of direct exoplanet observations.17 The surface brightness of the exozodi is the main determinant in how long it takes to detect an exoplanet buried in that system; the exposure time required to detect a planet is proportional to the exozodi brightness. Such background-limited observations strongly favor telescope diameter, where the signal to noise ratio is proportional to D4.

3. INSTRUMENT CAPABILITIES AND CONSTRAINTS Both science goals and starshade design depend on existing instrument capabilities and constraints from the observatory, and it is diﬀerent than designing a mission from the beginning. We develop this analysis under the NWP: Imaging

• Imaging capabilities with NIRCam ‣ Several Broadband ﬁlters and medium band ﬁlters ‣ SNR=10 detection of Earth at 10pc (solar-system zodiacal disk) in: - 23 hours with F070W - 7.3 hours with F090W - 5.8 hours with F115W - 11 hours with F200W ‣ Super Earth (5 Earth-mass, same density & albedo) - 2.7 hours with F070W - 0.85 hours with F090W - 0.7 hours with F115W - 1.3 hours with F200W ‣ Emitted light detection at 4-5 microns (with relaxed IWA 200-250 mas and closer occulter - If thermal emission from occulter is acceptable) 14

Thursday, November 12, 2009 NWP: Spectroscopy

• Spectroscopic capabilities with NIRSpec ‣ Use a target acquisition ﬁlter to reduce detector sensitivity range. - F140W (0.8-2.0 μm) most interesting for science - F110W (1.0-1.2 μm) for small starshade

‣ SNR=5 spectrum of Earth at 10pc (solar-system zodiacal disk) in: - 3x105 between 1.0 and 1.7 micron - 1.3x106 below 1.0 micron - resolution R=30-50

‣ SNR=5 spectrum of Super-Earth at 10pc (5ME, solar-system zodiacal disk) in: - 4x104 between 1.0 and 1.7 micron

6 ‣ Super Earth O2 detection in 10 + s at R~100 with grating ‣ Giant planet R=1000 spectrum in 5x105 to 106s ‣ R=2700 possible on bright young giants? 15

Thursday, November 12, 2009 Red Leak

• NIRCam’s detector is sensitive up to 2.5 μm ‣ smaller range but imaging • NIRSpec’s detector is sensitive up to 5.0 μm, ﬁlter substrate up to 2.7 μm ‣ larger range but dispersed spectrum 0

!2 1e-2 @ 2 μm " !4 1e-10 !6 λ<1.1μm Average Intensity ! !8 log !10

!12 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 16 Wavelength Λ in Μm

Thursday, November 12, 2009 Available ﬁlters

• NIRSpec “Target Acquisition” ﬁlters ‣ F140X : 0.8 μm < λ < 2.0 μm ‣ F110W: 1.0 μm < λ < 1.2 μm - F140X has a red bump of about 7-8% at ~3 micron • NIRCam ﬁlters ‣ F070W, F090W, F115W, F150W

• Baseline design: ‣ 0.6/0.8 to 2.0 micron (overlap NIRcam & NIRSpec range) ‣ core science up to 1.7/1.8 μm (relax contrast beyond) ‣ constraints at 2.5 μm & 5.0 μm for NIRcam & NIRSpec

Thursday, November 12, 2009 Starshade optimization

• Example of NIRCam filter F090W + detector + dichroic ‣ similar transmissions for other filters (not final parts)

Starshade needs to 0.01 perform at ~1e-6 in this region Typical out-of-band

transmission 4 10 rejection 1e-4 to 1e-5 F090W

106

1.0 1.5 2.0 2.5 18 Wavelength Μm Thursday, November 12, 2009

Starshade optimization

• weighted optimization ‣ NIRcam constraint at 2.5 micron (overall suppression of 1e-10 ‣ Contrast relaxed beyond 1.7 μm (science) up to 1e-6 at 2.5 μm (red leak)

Thursday, November 12, 2009 Starshade optimization

• Overall on-axis transmission including starshade + OTE + detector + dichroic + actual ﬁlter • The contribution form the star leakage is negligible in the error budget (perfect starshade)

10!10

10!12

10!14 axis suppression ! On 10!16

1.0 1.5 2.0 2.5 Wavelength Μm 20

Thursday, November 12, 2009 ! " NWP Simulated Solar System Detection

• Sun-Earth at 10pc with zodiacal disk ‣ Simulation for F090W for a 7h exposure ‣ Includes perfect starshade 70m (tip-tip) at 72193km ‣ Actual F090W transmission + starshade leak up to 2.5 μm

Earth at 10 pc, NIRCam F090W, ExpTime7h Earth at 10 pc, NIRCam F090W Nyquist

Thursday, November 12, 2009 Earth Spectrum with NIRSpec

• SNR=5 spectrum of Earth at 10pc in 3x105s with prism (R=30-50) ‣ does not include effects of out-of-band ﬁlter & varying sensitivity

1.2 " 10!10

1. " 10!10

8. " 10!11

6. " 10!11

!11 Flux ratio 4. " 10

2. " 10!11

!2. " 10!11

0.8 1.0 1.2 1.4 1.6 1.8 2.0 22 Wavelength Μm Thursday, November 12, 2009

! " Sample NWP Science Exposure Time Allocation

• Plausible total exposure time: 7-9% (~107s) - Total slews ~60-80 ‣ Planetary system architecture: deep search of 10 best RV systems - Detect outer planets, terrestrial planets, Probe habitable zone - Detect zodiacal dust and planet(s)-dust interaction and structures ‣ Deep survey of 20 nearby stars - Probe habitable zone for terrestrial planets - Exo-Zodiacal dust brightness, structure and color/spectroscopy ‣ Shallow revisit of known 20 best RV systems - Two visits to disambiguate inclination in RV detection (M.Sin i), test coplanarity ‣ Low-resolution spectroscopy of 20-30 giants and Neptunes ‣ High-resolution spectroscopy 10 giant planets (2-3 mature, 6-9 young) ‣ Low-resolution spectroscopy of 3-5 terrestrial planets

Thursday, November 12, 2009 Sample NWP Science Exposure Time Allocation

• Total exposure time 7-9% of mission time (107s) ‣ 1/3 imaging ‣ 2/3 spectroscopy

Thursday, November 12, 2009 Conclusions

• JWST + starshade excellent characterizer, short exposure times, high- resolution possible on brighter planets • Limited for detection (max observing time ~ 7-9%) • not limited by imperfect detector in most observing modes, near infrared is interesting • better estimation of spectroscopic capabilities in progress

Thursday, November 12, 2009