Finding habitable earths around white dwarfs with a ...tfa.cfht.hawaii.edu/presentations/Agol_tfa_2Mar2011.pdf · Finding habitable earths around white dwarfs with a robotic telescope

Finding habitable earths around white dwarfs with a robotic

telescope transit survey

Eric Agol

Associate Professor

Department of Astronomy

University of Washington (UW)

Feb 16, 2011 1

Evolution of a Sun-Like Star

1 AU

radius of Earth

Minitial=1.2 M⊙

Mfinal =0.6 M⊙

Data from Jimenez et al. (2004), Renedo et al. (2010)

Main sequence

Red giant

White dwarf

Earth flux distance ∝L1/2

2nd generation planets?

Formation of short period planets

• Reform planets in WDHZ: (1) disk formation? (2)

migration via instability + tidal circularization? (3)

companion evaporation?

• Hints: (1) pulsar planets; (2) polluted WD; (3) dust

disks around WD.

• For now, assume these planets form with a frequency:

η⨁= fraction of WD w/ 0.1-10 M⨁ planets & a < 0.02 AU

• Transit probability is ≈(Rp+RWD)/a ≈ 1%, but transit

depth can be up to 100%

• Requires ≈200η⨁-1 white dwarfs to be surveyed for 32

hr each to detect 1 planet: requires robotic telescopes

Habitable planet transits

White dwarf transit survey

• Assume global network of 1 meter telescopes, e.g. Las Cumbres Observatory Global Telescope Network or

• Follow each white dwarf for ≈32 hours to cover 3 Gyr continuously habitable zone (≈0.02 AU)

• If ≈few minute transit not detected, move to next white dwarf

WD detected planet distribution

• Mass decreases as dn/dM ∝ M-4/3

• Search out to 100 pc (20,000 WD): 73 yr on sky (!) • Peak is near planets with radius and temperature of Earth • Detect ≈1-10 planets @ >6σ:

η⨁ ≈1-10%

Few massive

Low probability

Low S/N

Few hot WD

Survey of WD CHZ with LSST:

• LSST will identify ≈107 White dwarfs w/ RPM

• Probability in transit: ≈(Rp+RWD)/(πa) ≈1/300

• With ≈103 epochs, ≈3 points in transit ➭ ephemeris

• Binary WD contamination (>103) requires follow-up: secondary eclipse, Doppler beaming, lensing, Roemer delay, eclipse shape

• LSST can detect 1-10 HZ Earths if η⨁≈0.05-0.5%

Properties of WDHZ

• Planets should be tidally locked:

– permanent day/night;

– rotation period ≈1 day

• Star will appear similar in size & color to Sun

• Longest duration in WDHZ is ≈8 Gyr

• Energy source is thermal + gravitational + crystallization, not nuclear (‘dead’ star)

Conclusions

• White dwarfs have a potentially ‘habitable’ zone from ≈.005-.02 AU lasting few Gyr

• If planets could (re)-form close to white dwarfs, easy to detect via transit (p ≈ 1%)

• Ground-based robotic surveys could find these planets, and next generation ground/space telescopes could characterize them; LSST may reach small η⨁ ≈ 0.05%

• Would have some properties similar to Earth

Extra slides

Liquid water is essential for life (as we know it)

• Clever biochemists have suggested that non-carbon-based, non-water-dependent life could possibly exist

• Nonetheless, the best place to begin the search for life is on planets like the Earth

• This means that we should look within the conventional habitable zone around nearby stars

• This does not necessarily mean these must be Sun-like stars

Finding the boundaries of the habitable zone

• In the Kasting et al. (Icarus, 1993) model, planets are assumed to develop dense atmospheres near either boundary of the habitable zone – Dense H2O atmosphere near the inner edge

(runaway greenhouse) – Dense CO2 atmosphere near the outer edge (from

the carbonate-silicate cycle feedback)

• Stars must have steady or slowly varying luminosities for planet to spend a long duration in the habitable zone

Liebert et al.

(2007)

Disk White dwarf luminosity function

Simulation of survey

• Mario Juric et al.’s catalog of white dwarfs in LSST (thanks Rob)

• Impose r < 24.5 cutoff; require at least 3 epochs observed with >7 σ detection of transit

• LSST can detect CHZ Earths if 0.05-0.5% of WD

LSST WD/planet properties

Intrinsic

Detected

White dwarf temperature Planet mass

Planet semi-major axis

Life Track of a Sun-Like Star

White dwarfs in distant globulars

Typically 1/10,000 Luminosity of Sun

White dwarf cooling

• White dwarfs cool by emitting neutrinos or photons

• Interiors are highly conductive: nearly isothermal

• As they cool, surface temperature decreases, so cooling rate slows:

Hubble data

L 4R2T 4 104L

M

M

8Gyr

7 /5

The age of the disk of our Galaxy is about 5-10 Gyr, so this formula predicts the faintest white dwarfs have L 10-4 L

for 8 Gyr

White dwarf

luminosity

Figure from Renedo et al.

(2010)

Continuously habitable zone

White dwarf mass distribution

White dwarf mass-radius relation

Provencal et al. (1998)

Rad

ius

in u

nit

s o

f Su

n

Mass in units of Sun

Extrasolar planet discovery & characterization: why?

1. Comparative planet formation

2. Planetary physics: • EOS at high pressure;

• Atmospheric physics

3. Uniqueness of Earth & signposts for life: Earth-sized & temperature planets

η⨁= fraction of stars w/ 0.1-10 M⨁ planets & T♂ < T < T♀

White dwarfs

• White dwarfs are the remaining cores of dead stars, but size of earth

• Electron degeneracy pressure supports them against gravity

• White dwarfs, once they cool, crystallize; carbon white dwarfs are ‘cosmic diamonds’: 1034 carats

Sirius B

Largest diamond on earth: Star of Africa

530 carats

White dwarfs

• White dwarfs are the remaining cores of dead stars, but size of earth

• Electron degeneracy pressure supports them against gravity

• White dwarfs, once they cool, crystallize; carbon white dwarfs are ‘cosmic diamonds’: 1034 carats

Sirius B

Largest diamond on earth: Star of Africa

530 carats

Gravitational interaction

Electromagnetic interaction

• Transit • Secondary eclipse • Imaging

• Radial velocity • Astrometry • Microlensing

Gmu =-8pTmu

Gravitational interaction

Electromagnetic interaction

• Transit • Secondary eclipse • Imaging

• Radial velocity • Astrometry • Microlensing

Gmu =-8pTmu

Winn (2009) 29

White dwarf temperature distribution

• Then need to survey more stars...

• More stars = fainter stars

• Time prohibitively large... and requires larger (=more expensive) telescope

• At some point becomes more efficient to survey many stars at once, with a single, wide-field telescope... Pan STARRS or LSST

What if planet frequency is small?

Robotic telescope survey

Is LSST* a Terrestrial Planet Finder?

≟

*Large Synoptic Survey Telescope

White Dwarf Habitable Zone (WDHZ)

• White dwarfs ≈Earth ≈ 1% of the Sun’s radius

• Most common white dwarfs have temperature of the Sun and 1/10,000 the luminosity

• So habitable zone is ≈0.01 AU (Kasting et al. 1993)

• Transit probability is ≈(Rp+RWD)/a ≈ 1%, but transit depth can be up to 100%

• Habitable Earth-size planets could be detected from the ground! But, need to form after red giant phase...

Agol (2011, ApJL, submitted)

Kasting et al., Icarus (1993)

Nuclear burning habitable zone

Luminosity of a Sun-Like Star

Main sequence

Red giant

White dwarf

M=1.2 M⊙

Data from Jimenez et al. (2004), Renedo et al. (2010)

2nd generation planets?