new worlds imager - university of colorado boulderwcash/planets/new_worlds.pdf · new worlds imager...
TRANSCRIPT
W. Cash – University of Colorado 1New Worlds Imager
New Worlds Imager
Webster Cash, University of Colorado
W. Cash – University of Colorado 2New Worlds Imager
New Worlds ImagerAn Alternative to TPF
Webster Cash University of ColoradoJim GreenEric SchindhelmNishanth RajanJeremy Kasdin Princeton UniversityBob VanderbeiDavid SpergelEd TurnerSara Seager Carnegie Institution – WashingtonAlan Stern Southwest Research Institute – BoulderSteve Kilston Ball AerospaceErik WilkinsonMike LeiberJim LeitchJon Arenberg Northrop GrummanRon PolidanChuck LillieWillard Simmons MIT
and growing…
W. Cash – University of Colorado 3New Worlds Imager
W. Cash – University of Colorado 4New Worlds Imager
Exo-Planets
CExo-planets are the planets that circle stars other than our Sun.
CThere are probably 10,000 exo-planets within 10pc (30 light years) of the Earth.
CPlanets are lost in the glare of parent star.
CThe Earth as viewed from light years is 10 billion times fainter than the Sun.
W. Cash – University of Colorado 5New Worlds Imager
Planet Finding: Extinguish the Star
Courtesy of N-G
W. Cash – University of Colorado 6New Worlds Imager
Terrestrial Planet Finder
CTelescopes must be PERFECT to suppress scatter:λ/5000 surface, 99.999% reflection uniformity
CTPF is very difficult
CIs there any easier way?
W. Cash – University of Colorado 7New Worlds Imager
New Worlds Imager vs. New Worlds Observer
CTwo Levels of DifficultyCNew Worlds Observer
–Two Spacecraft–Goal is Finding Planets–Science from Photometry and Spectroscopy–Technology is In-Hand Today
CNew Worlds Imager–Five Spacecraft–Goal is True Imaging of Earth-like Planets–MUCH Tougher – Technology 10-15 years out
W. Cash – University of Colorado 8New Worlds Imager
Initially New Worlds was a Pinhole CameraPerfect TransmissionNo Phase ErrorsScatter only from edges – can be very low
Large Distance Set by 0.01 arcsec requirementdiffraction: λ/D = .01”à D = 10m @500nmgeometric: F = D/tan(.01”) = 180,000km
W. Cash – University of Colorado 9New Worlds Imager
Diffraction Still a Major Problem for Pinhole
Developed by PrincetonGroup for Apertures
Answer: Shape the Aperture (Binary Apodization)
W. Cash – University of Colorado 10New Worlds Imager
CSmaller Starshade–Create null zone, image around occulter
CObserve entire planetary system at once
The Occulter Option
W. Cash – University of Colorado 11New Worlds Imager
The Diffraction Problem Returns
CSeveral previous programs have looked at occultersCUsed simple geometric shapes
– Achieved only 10-2 suppression across a broad spectral band
CWith transmissive shades– Achieved only 10-4 suppression despite scatter problem
http://umbras.org/BOSS Starkman (TRW ca 2000)
W. Cash – University of Colorado 12New Worlds Imager
Extinguishing Poisson’s Spot
COcculters Have Very Poor Diffraction Performance– The 1818 Prediction of Fresnel led to the famous episode of:– Poisson’s Spot (variously Arago’s Spot)– Occulters Often Concentrate Light!
CMust satisfy Fresnel Equation, Not Just the Fraunhoffer Equation
CMust Create a Zone That Is:– Deep Below 10-10 diffraction– Wide A couple meters minimum– Broad Suppress across at least one octave of spectrum
CMust Be Practical– Binary Non-transmitting to avoid scatter– Size Below 150m Diameter– Tolerance Insensitive to microscopic errors
W. Cash – University of Colorado 13New Worlds Imager
The Vanderbei Flower
CDeveloped for Aperture in TPF focal planeCWas to be only 25µ acrossCVanderbei had determined it would work for the
pinhole camera but did not work for occulter.
W. Cash – University of Colorado 14New Worlds Imager
The Apodization Function
( ) 0A ρ = 1rρ <
1rρ >( )
21
21
nr
rA eρ
ρ −
− = −
for
for
and
Found this in April. Extended in June.This Function Extinguishes Poisson’s Spot to High Precision
W. Cash – University of Colorado 15New Worlds Imager
Suppression of Edge Diffraction Can Be Understood
Using Fresnel Zones and Geometry
CThe occulter is a true binary optic–Transmission is unity or nil
CEdge diffraction from solid disk is suppressed by cancellation –The power in the even zones cancels the
power in the odd zonesØNeed enough zones to give good deep cancellation
• Sets the length of the petals
–Petal shape is exponential ~exp(-((r-r1)/r2)2n)Ør2 is scale of petal shapeØn is an index of petal shapeØr1 is the diameter of the central circle
r1
r2
W. Cash – University of Colorado 16New Worlds Imager
CThe Residual Intensity in the Shadow is
CBy Babinet’s Principle where EA is field over Aperture
CSo We Must Show
C d is distance to starshade, s is radius of hole, k is 2π/λ
CTo one part in
Doing the Math(Cash, 2005)
2s sI E=
1s AE E= −
22 2 11
2
1
2 2cos cos2 2
0 0 0
12
nrr ik ik s ik ik s
rd d d d
r
ke e d d e e e d d
d
ρπ πρ ρ θ ρ ρ θ
ρ ρ θ ρ ρ θπ
−∞ − − − + =∫ ∫ ∫ ∫
510C −≈
W. Cash – University of Colorado 17New Worlds Imager
Contrast Ratio
CPreceding integral shows the contrast ratio is
–
– n is an integer parameter, currently n=4
CTo keep R small r1~r2– this is the reason the occulter has that symmetric look
( )22
2 21 2
2 !2
n
n n
n dR
r rλπ
=
W. Cash – University of Colorado 18New Worlds Imager
Off Axis PerformanceCThe off axis performance shows a rapid rise to unit transmission
for the radii greater than the inner edge of the habitable zoneShadow Profile
Radius (meters)
0 5 10 15 20 25 30 35
Lo
g10
Co
ntr
ast
-18
-16
-14
-12
-10
-8
-6
-4
-2
0
n=2
n=4
n=6
W. Cash – University of Colorado 19New Worlds Imager
Modified Rendering
W. Cash – University of Colorado 20New Worlds Imager
“Standard” Observatory Views the Starshade
~0.1” resolution is needed (just to separate planets)
High efficiency, low noise spectrograph (e.g. COS)
W. Cash – University of Colorado 21New Worlds Imager
Count rate estimation
CAssuming visible solar flux and a half-earth viewed at 10 pc,
CCan achieve 5 counts per second with 80% efficient 10 meter telescope
Telescope Time required for S/N=10 detection
1 meter 33.3 minutes 2 meter 8.3 minutes 4 meter 2.1 minutes 8 meter 31 seconds
C ∝FSrE
2DT2
εγ dS2
W. Cash – University of Colorado 22New Worlds Imager
Another Issue:Scattered Light
Sun
Target
CSunlight Scatters Off StarshadeCCan be Controlled in Multiple Ways–Look at right angles to sunØImposes restrictions on revisit times
–Operate in shadowØEarth’s umbraØWith additional shade
• Likely hard at L2• Easier in heliocentric orbit
W. Cash – University of Colorado 23New Worlds Imager
Starshade Tolerances
CPositionØLateral Several MetersØDistance Many Kilometers
CAngleØRotational NoneØPitch/Yaw Many Degrees
CShapeØTruncation 1mmØScale 10%ØBlob 3cm2 or greater
CHolesØSingle Hole 3cm2
ØPinholes 3cm2 total
W. Cash – University of Colorado 24New Worlds Imager
Fly the Telescope into the Shadow
W. Cash – University of Colorado 25New Worlds Imager
Dropping It In
W. Cash – University of Colorado 26New Worlds Imager
Typical Observing Timeline
CAlignment 3 days Travel– Other astrophysics
CDeep Photometry 1 day Find PlanetsCPreliminary Spectroscopy 1 day Classify PlanetsCDetailed Studies 3 days Search for
– Deep Spectroscopy Water– Extended Photometry Surface Features– Life ?!
CReturn After Months Measure Orbits– New Planets from Glare
W. Cash – University of Colorado 27New Worlds Imager
The First Image of Solar System
JupiterSaturn
Uranus
Neptune
Zodiacal Light
Galaxies
10 arcseconds
W. Cash – University of Colorado 28New Worlds Imager
Great Science with Small TelescopesCLower limit on telescope size set by need to acquire adequate
signal and resolve planets from one another– 1 m diameter telescope needed to see 30M object in minutesØResolution of 0.1 arcsec
– 2 m diameter gives count rate 0.2 sec-1 for Earth at 10 pc at half illumination
50,000 seconds 400,000 seconds
Jupiter
Earth
Mars
Saturn
Uranus
Neptune
W. Cash – University of Colorado 29New Worlds Imager
Zodiacal light
CPlanet detectability depends on system inclination and telescope resolution– Face-on 0.3 AU2 patch of zodi equal to Earth’s brightness
CZodiacal light can wash out planets at low inclinations
60°
30°10°
W. Cash – University of Colorado 30New Worlds Imager
Zodiacal Light – 0.05” IWA
Pole-on Edge-on
W. Cash – University of Colorado 31New Worlds Imager
Spectroscopy
CR > 100 spectroscopy will distinguish terrestrial atmospheres from Jovian with modeling
O2
H2O
CH4
NH3
S. Seager
W. Cash – University of Colorado 32New Worlds Imager
Photometry
Calculated Photometry ofCloudless Earth as itRotates
It Should Be Possible to Detect Oceans and Continents!
W. Cash – University of Colorado 33New Worlds Imager
Alternate Operations Concepts
CGround based telescopeØRelay mirror at GEOØSouth Pole
CSpace based telescopeØAs JWST instrumentØDedicated telescope and mission
W. Cash – University of Colorado 34New Worlds Imager
Occulter and Detector Craft Functions
CPropulsion CStation keeping CAlignment establishment and maintenance
–Measurement and reporting of relative location
CData transfersCPointing requirements dependent on tolerancing of
occulter–Pointing error results in an error in the occulter shape by projection
CWhat is the role of the ground in directing the two SC?–Cost trade?
W. Cash – University of Colorado 35New Worlds Imager
Formation Flying Simulation
CLargest problem is solar radiation pressure–Pinhole craft’s cross sectional area: 7150 m²–Craft will be thrown out of libration point orbit
after several daysTotal stationkeeping ? V [m/s]
20.79.8200,000 km
20.310.220,000 km
L5L2
Number of burns during exposure
38106700200,000 km
3740670020,000 km
L5L2
L4
L5
L1 L2
W. Cash – University of Colorado 36New Worlds Imager
Formation Flying Simulation
CStationkeeping ?V estimated in STK/Astrogator– Detector craft assumed active; pinhole craft assumed passive– Control box of 10 cm half-width defined– Active S/C thrusts when box boundaries reached– Gravity of Earth, Sun, Moon included, plus solar radiation pressure– Separations of 20,000 km and 200,000 km considered at Earth-Sun L2 and L5
DetectorPinhole
h
Detector craftcenter of mass
Control box[inertially fixed with respect to optics craft]
W. Cash – University of Colorado 37New Worlds Imager
EELV 5 meter heavy
Generic L2 Bus
Up tp 150 m New Worlds Observer Will Fit in an ELV Heavy
W. Cash – University of Colorado 38New Worlds Imager
Simple low cost solar array style deployment
New Worlds Deploys Like Solar Arrays
CSimple, robust, proven deployment scheme
W. Cash – University of Colorado 39New Worlds Imager
Earth Viewed at Improving Resolution
100 km300 km3000 km 1000 km
TRUE PLANET IMAGING
W. Cash – University of Colorado 40New Worlds Imager
Solar System Survey at 300km Resolution
W. Cash – University of Colorado 41New Worlds Imager
collector craft1500km
starshades
collector craft
combiner
50,0
00km
1500km
NWI Concept
W. Cash – University of Colorado 42New Worlds Imager
Holding the Array
planet collimator
field star collimator
primary collector
to combiner
planet collimator
field star collimator
primary collector
to combiner
planet beams
field star beams
Delay Lines – Mixers - Detectors
planet beams
field star beams
Delay Lines – Mixers - Detectors
W. Cash – University of Colorado 43New Worlds Imager
Planet Q-Ball
One Imperfection
Fringes as TelescopesSeparate
Imperfection visiblewith adequate signal
Information is there:We will study the realistic limits of two element interferometers
W. Cash – University of Colorado 44New Worlds Imager
Resolution Limitation Set By Signal
CAt 10 km resolution the interferometer is photon-limitedCNeed Much Bigger Telescopes – Too Expensive
W. Cash – University of Colorado 45New Worlds Imager
The Phase II Study
CTwo Year Study Began on September 1
CObserver Mode Well UnderstoodØComplete Architecture Study Completed in First YearØLaboratory Demonstration of Diffraction Suppression
CImager Mode More DifficultØWill Study Requirements in DetailØWill Look for Ways to Make the Mission More Affordable
W. Cash – University of Colorado 46New Worlds Imager
Conclusion
By 2011
By 2018
O2
H2O