Transcript Temperature, df/f

Opportunities in the Study of
Extrasolar Planets
Peter R. McCullough, STScI
Astrophysics Enabled by the Return to the Moon
Nov 30, 2006
Outline (also the summary)
1) Transiting planets are great science.
2) Somewhere in space is a good site for a 0.6-m telescope to
monitor known transiting systems.
3) The Moon is in space.
4) Polarization may be more practical than spectra for physical
characterization of exo-earths.
5) A 10-m diameter telescope with imaging polarization
capability and more modest wavefront quality requirements
than TPF-C can detect oceans if they exist on terrestrial
exoplanets and nearly map continental boundaries.
6) Return to the Moon will inspire creative ways to overcome
challenges and tap new opportunities.
Historical Precedent
Captain Cook’s International Expedition
Transit
Hoped
How
of Venus, 1769 in Tahiti
to measure the physical scale of the solar system.
much did it cost?
Endeavor
Site Infrastructure
lasts for
Generations
XO’s Site #1 is
Haleakala’s Site #1,
the SAO BakerNunn
Building.
1957
1958
www.ifa.hawaii.edu/users/steiger/post_cook.htm
XO-1 telescope parade:
Exoplanets are relevant to these Satellites:







Hubble
Spitzer
MOST
COROT
Kepler
JWST
TPF *
*Flight
Opportunity
for TPF is
TBD.
HD 209458 with HST/STIS Spectrophotometry
Egress
Ingress
Eclipse
Brown et al 2001
XO targets Bright Stars that allow …


Absorption Spectra of Planetary Atmosphere
Precise photometry and timing of transits





Oblateness (rotation rate) of planet
Rings, Satellites of planet
Perturbations from TP: d(TOA) ~ many seconds
Limb darkening and star spots
Secondary Eclipse



Temperature, df/f ~ ( Tp/Ts ) ( Rp/Rs )2 ~ 0.4% in IR
Eccentricity better than RV method
Albedo, df/f ~ 0.02% x optical albedo
An Extrasolar Planetary Atmosphere
Charbonneau et. al. 2002, ApJ, 568, 377


Spectrophotometric observations of four planetary
transits taken in sodium doublet at 589.3nm.
DI (sodium/cont.):

2.32  (0.57) x 10-4
Habitable worlds

Satellites of warm Jupiters
 Could be more prevalent
habitable exoworlds
than exoplanets (!?)
 Detectable by its
gravitational influence
on its host planet:
transits arrive early and
late.
 At some wavelengths,
planet is much darker
than its satellite(s).
 There are ~10 transiting
Jupiters of stars V<12
and periods ~1 year;
they will be found
before lunar sorties.
70 Vir, available from Extrasolar Visions Inc.
Lunar Deployable
Telescope
Drake Deming (PI), Peter McCullough, and David Charbonneau
LDT’s terrestrial
radius mass diagram
Drake Deming (PI), Peter McCullough, and David Charbonneau
LDT’s terrestrial
planet transit
Drake Deming (PI), Peter McCullough, and David Charbonneau
LDT as concept
Drake Deming (PI), Peter McCullough, and David Charbonneau
•By 2019 …
•Kepler will have completed its mission, detecting Earth-like planets orbiting solar-type stars
from 100 to 300 pc distant.
•Ground- and space-based transit surveys, augmented by precise radial velocities, will have
detected 100 hot Jupiters transiting Sun-like stars and some terrestrial planets transiting nearby
K- and M-dwarfs.
•Spitzer measurements of thermal emission from exoplanets will have carried over to JWST,
which will have measured thermal emission spectra of hot Jupiters and hot Neptunes. JWST
will be pushing toward measuring the thermal spectrum of a close-in Earth-like planet around a
lower main sequence star, using the secondary eclipse technique.
•The most important requirement for transiting planet studies is a site in space, free of the
terrestrial atmosphere, and with the capability to observe a given transiting planet system with
photon-limited photometric precision for long durations (multiple transits).
•LDT will be deployable by two astronauts and operate autonomously. It will …
•Observe planets transiting bright, nearby, stars.
•Stare at a given star for weeks to months.
•Discover additional terrestrial planets of known transiting systems based upon:
•Transits of the terrestrial planets: the premise (unproven) is the inclination is favorable.
•Timing giant-planet transits.
•Measure accurate sizes for terrestrial planets orbiting the nearest stars and measure their
visible reflected light (at secondary eclipse).
Drake Deming (PI), Peter McCullough, and David Charbonneau
Much Bigger
Light Buckets…
Transiting Planet Spectroscopy (1 of 2)
•
Proven technique: V=8, Jovian planet HD 209458b orbiting
sun-like star, observed with HST STIS, Na etc detected.
•
Earth-Sun at 10 pc: Absorption-lines due to transiting planet
atmosphere feasible with 10-m steerable light bucket in
vacuum plus spectrograph. (Gilliland, p.c.)
•
PSF FWHM = 1 arcmin is ok
Transiting Planet Spectroscopy (2 of 2)
•
Emission spectrum of planet disappearing behind star.
planet atmosphere.
•
Proven technique: V=8, Jovian planet HD 209458b, etc
provides spectral energy distributions of planet in Spitzer
bands.
•
JWST will get true spectra of hot Jupiters by this technique
– but probably not terrestrial exoplanets.
Imaging Stars with Transits
Does this help?
Size of error bar ~ 1/(A Dn t)1/2
Water… needs more
than a bucket
McCullough (2006 astroph  2007 ApJ)
A few others independently working on this
concept also: Ford, Hough, Williams, Stam,
Schmid
Unpolarized Light
Satellite
images
Simulations
Polarized Light
S-pol
P-pol
10-m or 20-m precision-surface (but
not TPF-C) steerable telescope
required.
Total flux
2x flux difference
What’s great about this?
Speckle
pattern (think JWST-like PSF) can be designed to be
nearly identical in the two polarizations, so the glare of the star
can be suppressed not only by coronagraphy but also by
subtracting one polarized image from the other. The
unpolarized star cancels out; the polarized planet doesn’t.
Spectra imply dividing light into 100+ bins; linear polarization
implies two bins.

Rayleigh scattering is very blue; glint from oceans is
achromatic. (so four bins: 2 polarization; 2 wavelengths)


Glint is very localized (~15 degrees of “longitude”).
The glint’s flux difference in the two polarizations is 0.15
photons per second for a 10-m telescope observing Earth-Sun
system at 10 pc. Long integrations (days) can pick out the
polarized light from an oceanic planet in the glare of the star.
One hour integration gives Poisson S/N ~ 20 if star light can be
suppressed entirely by a superb technology.

No atmospheric absorption
Earth-like clear atmospheric absorption
Projected like we’d observe.
Light curve for Earth
Ocean planet with clear atmosphere;
Rayleigh suppressed (long wavelength)
With clouds too
Half-Baked
Better than “not even wrong”…
About costs…
Net
worth of US households is 50 T$ in 2000.
The
aggregate value of corporate equities directly held was 9
T$ in 2000 and was 4 T$ in 2003, so it declined by ~2 T$ per
year for three consecutive years. That's 1 B$ per hour of each
and every working day for three years.
The
median market capitalization of a corporation in the DJIA
is 108 B$ (as of Oct 31, 2006).
Pfizer,
a pharmaceutical company founded in 1849, in 2005
had annual revenue of 51 B$ and spent 7 B$ on R&D. NASA
by comparison was awarded 16 B$ in 2005 by the US
Congress.
Something more immediate than global
warming…
Europe, different rendering…
Various thoughts…sustainability
Is
colonization of the Moon a metaphor for solving Earth's
geopolitical problems? A quasi-sustainable presence on the
Moon wouldn't use fossil fuels; it would use solar or nuclear
with a considerable emphasis on conservation.
Making
scientific equipment on the moon is desirable
strategically even if its possible to bring it there from Earth.
A
flywheel with 2-m radius and 2-m height filled with lunar
regolith has a mass of 40 metric tons, and if spun to 1000 rpm,
has a kinetic energy of 110 kwh, which corresponds to 330
Watts for the 300-hour duration of the lunar night.
On
the Moon, a need for UPS may exist for these reasons:
to
buffer solar power to the night
to
provide for the variable power demand of human
habitation
to
mitigate risk
Various thoughts…bandwidth

JWST is bandwidth limited from L2.

LSST = 10 Terabyte/night.
That data rate may be impractical to transmit to Earth from
sensors on Moon or L2, but can be carried home on a disk if
there’s a regular data delivery service:

One kg equals 1 Terabyte in 2006. By 2020, 14 doublings
later, that’ll be 8000 Terabytes per kg, so 400 Petabytes will
weigh as much as a human.

Alternatively, transmitting from lunar surface to lunar orbit
for storage and subsequent transport to Earth may be
desirable.

Various thoughts…
Hoyle’s ideas on panspermia could be tested by trying to
detect life in lunar regolith, suitably nurtured, much like Viking
did on Mars but much more seriously. PCR or equivalent
technique(s) might permit detection at very trace levels.

A satellite can be sustained with some propulsion inside L2
so it is in Earth’s shade all the time. Nuclear power required.

Things too dangerous (or not permitted) on or near Earth
could perhaps be done on the Moon. One example,
Clementine. Another example: perturbing a small asteroid to
orbit (or hit) the moon instead of the Earth (in case it misses).

Nuclear explosions could be used for excavations of and
transmuting elements of the regolith. Also for pinging the moon
for tomography.

The pristine fossil water ice in the lunar polar craters could
be burned like fossil fuels are on earth.

Various thoughts… security and economics
Those examples (I hope) make you think that not all things
that are possible are desirable or ethical.

The return to the Moon is supposed to enhance our national
security. Bored children tend to fight; busy and intrigued
children fight less. Perhaps the same could be true for nations
and their peoples.


Economic enhancement?
When asked how much the US should spend and on which
scientific disciplines, someone (I forget who) replied, “[exactly
as much as it takes to be number one in each and every one.]”

Seward’s Folly (buying Alaska) and multi-national Antartica
are possible answers to why the US will go to the Moon.

Yankee ingenuity fostered at grad-student level if few-100 kg
payloads are delivered regularly to space. (If not, the best
brains will do biology, computer science, etc.)

Various thoughts…


Human+robot surgery will develop tools we can use.
Parabola or Sphere?
Rotating Liquid Mirrors produce parabolas naturally.
Two surfaces rubbed together produce a sphere naturally. So
does a bubble via surface tension.
Arecibo and Hobby Eberly Telescope are both spheres.
Spheres have advantages over Parabolas:
1) multiple focal planes can look at very different directions on
the sky using a single sphere.
Parabola looks up.
F
Parabola or Sphere?
C
F
F
F
Parabola or Sphere?
Rotating Liquid Mirrors produce parabolas naturally.
Two surfaces rubbed together produce a sphere naturally. So
does a bubble via surface tension.
Arecibo and Hobby Eberly Telescope are both spheres.
Spheres can have advantages over Parabolas:
1) multiple focal planes can look at very different directions on
the sky using a single sphere.
2) segmented optics are identical for a sphere.
3) replica optics fabricated on the moon, subsequently ion
polished (or equivalent) and coated in place courtesy of the
lunar vacuum, and actively controlled to maintain figure
could be an alternative to rotating liquid mirrors.
Summary
1) Transiting planets are great science.
2) Somewhere in space is a good site for a 0.6-m telescope to
monitor known transiting systems.
3) The Moon is in space.
4) Polarization may be more practical than spectra for physical
characterization of exo-earths.
5) A 10-m diameter telescope with imaging polarization
capability and more modest wavefront quality requirements
than TPF-C can detect oceans if they exist on terrestrial
exoplanets and nearly map continental boundaries.
6) Return to the Moon will inspire creative ways to overcome
challenges and tap new opportunities.