Observations of extrasolar planets

Wednesday, November 9, 2011 1 Mercury

Wednesday, November 9, 2011 2 radar image from Magellan (vertical scale exaggerated 10 X) Venus

Wednesday, November 9, 2011 3 Mars

Wednesday, November 9, 2011 4 Jupiter

Wednesday, November 9, 2011 5 Saturn

Wednesday, November 9, 2011 6 Saturn

Wednesday, November 9, 2011 7 Uranus and Neptune

Wednesday, November 9, 2011 8 we need to look out about 10 parsecs or 2 million astronomical units to examine 100 stars for planets

Wednesday, November 9, 2011 9 why is finding planets hard?

• the Earth is visible in reflected light from the Sun • the total light reflected by Earth is about 10-9 of the light emitted by the Sun • Jupiter is bigger but more distant so only about 4X brighter than Earth These would be easy to see with modern telescopes at 10 parsecs if the star were not right beside them (imagine looking at a lighthouse 1000 km away and trying to detect a firefly flying 1 meter from the light) With current technology direct imaging requires one or more of: • planet with a large orbital radius (e.g. 100 AU vs. 30 AU for Neptune) • observations in the infrared (for a black body in thermal equilibrium at 100 AU around the Sun, emission peaks at 100 microns) • young planet (Jupiter’s internal luminosity falls at 1/age)

Wednesday, November 9, 2011 10 • star is at 7.7 parsecs • planet has 110 AU orbital radius, 900 yr orbital period • 100-300 Myr stellar age

Wednesday, November 9, 2011 11 HR 8799

• star is at 39 parsecs • planets have masses of 7-10 Jupiter masses and projected separations of 14-68 AU • 30-160 Myr age

Marois et al. (2008, 2010)

Wednesday, November 9, 2011 12 • current accuracy: velocity of 1 meter/sec (3 parts in 109) • cross-correlation uses all lines in spectrum • high S/N (bright stars, big telescopes) • iodine absorption cell • old stars (less rotation, less activity) • G stars • Jupiter: orbital period of 12 yr, reflex velocity of Sun 13 meter/sec • Earth: orbital period of 1 yr, reflex velocity of Sun 0.1 meter/ sec

Wednesday, November 9, 2011 13 Given the star mass M (known from spectral type), radial-velocity observations yield: • orbital period P • semi-major axis a • combination of planet mass m & inclination I, m sin I • orbit eccentricity e

Wednesday, November 9, 2011 14 • the star is similar to the Sun and 11 parsecs away • the planet orbits once every 15.8 days, has a mass 4 X that of Jupiter (if edge-on orbit), and is 0.11 AU 100 meters/second from the star

Wednesday, November 9, 2011 15 HD 82943

planet 1:

m sin I = 1.84MJ P = 435 d e = 0.18 ± 0.04

planet 2:

m sin I = 1.85MJ P = 219 d e = 0.38 ± 0.01 (Mayor et al. 2003)

Wednesday, November 9, 2011 16 Timing

• works for pulsars, pulsating stars, eclipsing binaries

• observed period is P=P0(1+v/c) so

Wednesday, November 9, 2011 17 Pulsar planets

• three planets discovered orbiting PSR B1257+12 by Wolszczan & Frail (1992)

• orbital parameters can be determined far more accurately than for radial-velocity measurements of nearby stars

• two planets near 3:2 resonance which enhances mutual perturbations, so these can be measured

• remarkably similar to the inner solar system

Wednesday, November 9, 2011 18 (a) no planets

(b) three planets

(c) three planets + mutual interactions

Konacki & Wolszczan (2003)

Wednesday, November 9, 2011 19 Konacki & Wolszczan (2003)

Wednesday, November 9, 2011 20 Transit of Venus 1769

• next June 6, 2012, 22:10 UTC at Tokyo

Jupiter’s satellite Io

Wednesday, November 9, 2011 21 Wednesday, November 9, 2011 22 Wednesday, November 9, 2011 22 Transit searches

Why are these so hard?

• probability that a given planet will transit is small, ~

rstar/a (only 0.5% at a = 1 AU)

• transit duration is short, ~(rstar/a)P/π

• transit depth is small, <1% *

• confusion from grazing eclipsing binary stars

• star spots, stellar pulsations, stellar flares

• incomplete sampling (daytime, weather, observing schedules, etc.)*

* much easier in space

Wednesday, November 9, 2011 23 0.5%

Wednesday, November 9, 2011 24 Kepler (NASA)

• launch March 6 2009

• stare 24/7 for five years at a single patch of sky

• monitor 200,000 stars for transits

• ppm precision

Wednesday, November 9, 2011 25 transit: planet moves in front of the star; U-shape because of limb darkening in star occultation: planet moves behind the star; square shape

Wednesday, November 9, 2011 26 transit: planet moves in front of the star; U-shape because of limb darkening in star occultation: planet moves behind the star; square shape

Wednesday, November 9, 2011 26 van Kerkwijk et al. (2010) • orbital period P = 5.2 days • two curious features: • sinusoidal brightness variations at fundamental and first harmonic • transit (U shape) is shallower than occultation (square well)

Wednesday, November 9, 2011 27 van Kerkwijk et al. (2010)

• orbital period P = 5.2 days • two curious features: • sinusoidal brightness variations at fundamental and first harmonic • transit (U shape) is shallower than occultation (square well) • both can be explained if the companion is a white dwarf rather than a planet: • occultation is deeper because the white dwarf is hotter than the primary (T=13,000 K vs. 9,400 K) • first harmonic due to tidal distortion of the primary by the white dwarf • fundamental due to Doppler boosting • white dwarf has mass 0.22±0.03 M⊙; radius 0.043±0.004 R⊙

Wednesday, November 9, 2011 28 • P/2 variations due to tidal distortion of the primary star by the white dwarf • P variations due to Doppler boosting orbital velocity to 1 km/s Struve (1952)

Wednesday, November 9, 2011 29 Gravitational lensing

• a particle traveling at high speed v past a mass M with impact parameter b suffers angular deflection α =2GM/v2b • in general relativity, deflection of a photon is obtained by replacing v by c and multiplying by 2:

• three effects: position shift, image splitting, image magnification

Wednesday, November 9, 2011 30 Gravitational source track lensing

lensing mass the gravitational field Einstein ring from the lensing star:

- splits image into two - magnifies one image and demagnifies the other - if source, lens and observer are exactly in line the image appears as an Einstein ring

Wednesday, November 9, 2011 31 Gravitational microlensing

Consider a source star near the center of the Galaxy, lensed by an

intervening star at half that distance. Then θE=0.001 arcsec ~ 4 AU.

• image splitting or shift is impossible to see

• image magnification is easy to see

• time required to transit Einstein ring ~DLθE/v~0.2 yr, for v~100 km/s

• substantial magnification if and only if impact parameter less than Einstein radius

• chance that any given star is microlensed is only ~10-6

Wednesday, November 9, 2011 32 Mao et al. (2002)

Wednesday, November 9, 2011 33 Gravitational microlensing of planets

• Einstein radius scales as M1/2 so cross-section and expected duration scale as M1/2~ 0.03 for Jupiter, i.e. duration ~ 1 day for Jupiter, ~1 hour for Earth

• image magnification is the same

• Einstein ring radius ~ typical planet orbital radius

Wednesday, November 9, 2011 34 Beaulieu et al. (2006): 5.5 (+5.5/-2.7) MEarth, 2.6

(+1.5/-0.6) AU orbit, 0.22(+0.21/-0.11) MSun,

DL=6.6±1.1 kpc

Wednesday, November 9, 2011 35 xxx

factor of ten

Gaudi et al. (2008) days Wednesday, November 9, 2011 36 Gaudi et al. (2008)

Wednesday, November 9, 2011 37 star path

1% of Einstein radius or 15 µas

source size

Wednesday, November 9, 2011 38 • two planets, b and c • can detect orbital motion of Earth and planet c • mb = 0.71 ± 0.08 MJupiter, mc = 0.27 ± 0.03 MJupiter • assuming coplanar, circular orbits ab = 2.3 ± 0.2 AU, ac = 4.6 ± 0.5 AU • distance 1.49 ± 0.13 kpc • M*=0.50 ± 0.05 M⊙

Gaudi et al. (2008)

Wednesday, November 9, 2011 39 why this is hard:

Astrometry • typical motions << 0.001 arcsec ~ 10-8 radians

• not many nearby stars

• confusion from outer planets: maximum radial velocity is (m/M)(GM/a)1/2 but maximum wobble is (m/M)a

space missions:

• GAIA (ESA) • launch 2013 • every Jupiter analog within the Sun’s motion 50 pc as seen from • 104 - 5 × 104 planets 10 pc •also transits via photometry

Wednesday, November 9, 2011 40 the current track record:

• imaging: 26 • radial velocity: 644 • transits: 185 • gravitational lensing: 13 • timing: 12 • astrometry: 0

see http://www.exoplanet.eu, http://exoplanets.org/

figure from Seager (2011)

Wednesday, November 9, 2011 41 Current record-holders (RV surveys)

• smallest semi-major axis a = 0.0143 AU = 3.06 RSun

• largest semi-major axis a=11.6 AU (Jupiter = 5.2 AU)

• biggest eccentricity e = 0.94

• smallest eccentricity e = 0

• smallest mass 0.0061 MJupiter = 1.9 MEarth

Wednesday, November 9, 2011 42 solar radius 0.00465 AU Mercury Venus Earth Mars Jupiter

Wednesday, November 9, 2011 43 solar radius 0.00465 AU Mercury Venus Earth Mars Jupiter

GJ 1214 b, a=0.0143 AU

Wednesday, November 9, 2011 43 undercounted because of selection effects

Wednesday, November 9, 2011 44 Wednesday, November 9, 2011 45 “hot Jupiters”

Wednesday, November 9, 2011 45 Grether & Lineweaver (2006)

Wednesday, November 9, 2011 46 deuterium hydrogen burning limit burning limit

Grether & Lineweaver (2006)

Wednesday, November 9, 2011 46 tidal circularization

Wednesday, November 9, 2011 47 binary stars eccentricity distribution of massive planets is similar to that of binary stars

massive planets (M > 4 Jupiter masses)

planets (M < 4 Jupiter masses)

Ribas & Miralda-Escudé (2007)

Wednesday, November 9, 2011 48 high-mass stars are more likely to host planets

Johnson (2007)

Wednesday, November 9, 2011 49 (from J. Winn)

Wednesday, November 9, 2011 50 obliquity = angle between spin angular momentum of star and orbital angular momentum of planet Rossiter-McLaughlin measures angle between projection of spin angular momentum of star and orbital angular momentum of planet on the sky plane

(from J. Winn)

Wednesday, November 9, 2011 50 Wednesday, November 9, 2011 51 HAT P-7b λ = 183±9° HAT P-30b Winn et al. (2009) HAT P-14b λ = 74 ± 9° HAT-P-30b 7 Winn et al. (2009) λ = 189 ± 5° Winn et al. (2011)

40

]

1 20 − 4.0 0

RV [m s −20 3.5 ] 1 −40 −

[km s i ] 10 1

− 3.0 Wednesday, November 9, 2011 52 sin 0 v C [m s −

O −10 2.5

−2 0 2 50 60 70 80 90 100 Time from midtransit [hr] ! [deg]

Fig. 4.— Rossiter-McLaughlin effect for HAT-P-30 Left.—Apparent radial velocity variation on the night of 2011 Feb 21, spanning atransit.ThetoppanelshowstheobservedRVs.Thebottompanel shows the residuals between the data and the best-fitting model. Right.—Joint constraints on λ and v sin i!.Thecontoursrepresent68.3%,95.4%,and99.73%confidencelimits. The marginalized posterior probability distributions are shown on the sides of the contour plot.

HATNet operations have been funded by NASA grants Hungarian Scientific Research Foundation (OTKA) for NNG04GN74G, NNX08AF23G and SAO IR&D grants. support through grant K-81373. This research has GT acknowledges partial support from NASA grant made use of Keck telescope time granted through NASA NNX09AF59G. We acknowledge partial support also (N167Hr). Based in part on data collected at Subaru from the Kepler Mission under NASA Cooperative Telescope, which is operated by the National Astronom- Agreement NCC2-1390 (D.W.L., PI). G.K. thanks the ical Observatory of Japan.

REFERENCES Anderson, D. R., et al. 2010, ApJ, 709, 159 Johnson, J. A., Winn, J. N., Albrecht, S., Howard, A. W., Marcy, Bakos, G. A.,´ Noyes, R. W., Kov´acs, G., Stanek, K. Z., Sasselov, G. W., & Gazak, J. Z. 2009, PASP, 121, 1104 D. D., & Domsa, I. 2004, PASP, 116, 266 Jenkins, J. M., et al. 2010, ApJ, 724, 1108 Bakos, G. A.,´ et al. 2007, ApJ, 670, 826 Kov´acs, G., Zucker, S., & Mazeh, T. 2002, A&A, 391, 369 ´ Bakos, G. A.,´ et al. 2010, ApJ, 710, 1724 Kov´acs, G., Bakos, G. A., & Noyes, R. W. 2005, MNRAS, 356, Borucki, W. J., et al. 2011, ApJ, 728, 117 557 Buchhave, L. A., et al. 2010, ApJ, submitted, arXiv:1005.2009 Lin, D. N. C., & Papaloizou, J. 1986, ApJ, 309, 846 Butler, R. P. et al. 1996, PASP, 108, 500 Latham, D. W., et al. 2009, ApJ, 704, 1107 Carpenter, J. M. 2001, AJ, 121, 2851 Ohta, Y., Taruya, A., & Suto, Y. 2005, ApJ, 622, 1118 Charbonneau, D., Brown, T. M., Latham, D. W., & Mayor, M. P´al, A., & Bakos, G. A.´ 2006, PASP, 118, 1474 2000, ApJ, 529, L45 P´al, A. 2009b, arXiv:0906.3486, PhD thesis Claret, A. 2004, A&A, 428, 1001 Mandel, K., & Agol, E. 2002, ApJ, 580, L171 Droege, T. F., Richmond, M. W., & Sallman, M. 2006, PASP, Marcy, G. W., & Butler, R. P. 1992, PASP, 104, 270 118, 1666 McLaughlin, D. B. 1924, ApJ, 60, 22 Fabrycky, D. C., & Winn, J. N. 2009, ApJ, 696, 1230 Morton, T. D., & Johnson, J. A. 2011, ApJ, 729, 138 Ph.D. thesis, University of Szeged, Hungary Moutou, C., et al. 2009, A&A, 498, L5 Gaudi, B. S. & Winn, J. N. 2007, ApJ, 655, 550 Noguchi, K., et al. 2002, PASJ, 54, 855 Hansen, B. M. S., & Barman, T. 2007, ApJ, 671, 861 Queloz, D., Eggenberger, A., Mayor, M., Perrier, C., Beuzit, Henry, G. W., Marcy, G. W., Butler, R. P., & Vogt, S. S. 2000, J. L., Naef, D., Sivan, J. P., & Udry, S. 2000, A&A, 359, L13 ApJ, 529, L41 Queloz, D., et al. 2010, A&A, 517, L1 H´ebrard, G., et al. 2008, A&A, 488, 763 Rossiter, R. A. 1924, ApJ, 60, 15 H´ebrard, G., et al. 2010, A&A, 516, A95 Sato, B., Kambe, E., Takeda, Y., Izumiura, H., & Ando, H. 2002, Hirano, T., Narita, N., Shporer, A., Sato, B., Aoki, W., & PASJ, 54, 873 Tamura, M. 2011, PASJ, in press [arXiv:1009.5677] Sato, B., et al. 2005, ApJ, 633, 465 Howard, A. W., et al. 2010, ApJ, 721, 1467 Schlaufman, K. C. 2010, ApJ, 719, 602 Isaacson, H., & Fischer, D. 2010, ApJ, 725, 875 Skrutskie, M. F., et al. 2006, AJ, 131, 1163 Johnson, J. A., et al. 2008, ApJ, 686, 649 Torres, G. et al. 2007, ApJ, 666, 121 Triaud, A. H. M. J., et al. 2010, A&A, 524, A25 Wednesday, November 9, 2011 53 detection of CH4,

H2O, Na, CO, CO2

Seager & Deming (2010)

Wednesday, November 9, 2011 54 Wednesday, November 9, 2011 55