Revisiting the Radius Gap with Gaia DR2

Transcription

1 Revisiting the Radius Gap with Gaia DR2 The California-Kepler Survey. VII. Precise Planet Radii Leveraging Gaia DR2 Reveal the Stellar Mass Dependence of the Planet Radius Gap BJ Fulton & Erik Petigura arxiv: (in review)

2 The California- Kepler Survey Keck/HIRES high-resolution spectra of 1305 stars hosting 2025 planet candidates Precision spectroscopy: Teff, logg, Fe/H, mass, radius, vsini Stellar radius precision: 39% > 10% Petigura et al. (2017)

3 Huber et al. (2014); Mullally et al. (2015) Johnson, Petigura et al. (2017)

4 The Radius Gap Fulton et al. (2017)

5 The Radius Gap Super- Earths Sub- Neptunes Fulton et al. (2017)

6 XUV photons 20% H/He ~3% H/He ~3 Me Core Oewn & Wu (2017)

7 XUV photons ~0.3% H/He 0% H/He ~3 Me Core ~3 Me Core Oewn & Wu (2017)

8 Fulton et al. (2017) Observations Photoevaporation Major Implications Constrains core mass distribution Earth-density cores (water-poor) Large scale migration after 100 Myr is uncommon Planet Size [Earth radii] Model Normalised Planet Density Owen & Wu (2017)

9 Flux Dependency high-mass stars Photoevaproation Desert Low-mass stars Radius Gap low completeness BJ Fulton Fulton, et al. (2017) ExoSoCal 2018

10 Gaia DR2

11 Spectroscopy + Parallax Table 1.ErrorBudget Fulton & Petigura (2018) Updated stellar and planetary parameters available upon request prior to publication Parameter Median Uncert. T e m K A K 60 K 0.02 mag mag µ 0.01 mag BC 0.03 mag R? 2.2% R p /R? 4.1% R p 4.9%

12 Radius Distribution

13 Radius Distribution CKS Only CKS + Gaia Fulton et al. (2017) Fulton & Petigura (2018)

14 Stellar Mass Dependence Photoevaporation desert extends to lower fluxes for more massive stars Gap and planets are larger around more massive hosts Populations are split more cleanly when split up by mass Period distributions are indistinguishable Fulton & Petigura (2018)

15 Stellar Mass Dependence 8 Owen, J. E. & Murray-Clay R. All Periods Planets orbiting more massive stars are, on average, larger and hotter Caveat: stellar mass is correlated with both metallicity and age Mass and Scalings for Super-Earths Radius [R ] 1.0 Periods < 2.5 days Owen & Murray-Clay (2018) Periods > 25 days 6 Radius [R ] Wu (2018) Normalised Planet Density Radius [R ] Metallicity Figure 8. Same a Figure 4, but only for planets whose host stars have masses in the range M and metallicities > 0.2. BJ Fulton ratios in the size histogram Fulton & Petigura (2018) Fig. 5. Planet sizes have no correlation with host star metal- work are safe if one is really interested in how planet formalicity. There are 1141 planets in GKS that overlap with the CKS Figure 9. Same as Figure 6, except th ets whose host stars have masses in t and metallicities > 0.2. and low mass stars (split at the med 1.02 M ). Small number statistics robust statements, but there is no evidence of a clear trend with ste cautious optimism to presume that bigger (more massive) solid cores a stars is robust. We need to be more cautiou small planets at long periods are mo metallicity stars. There is weak, bu evidence, that both sub-samples c in metallicity, but wider range in in stellar mass, but a wider range trends that could be responsible for servations that small planets are m riods around CKS stars with lower here we do not have the sample siz the driver; one would suspect that Additionally, one might wor metallicity stars are more massive the CKS sample, small planets at la observe around these objects. Petigu suring absolute occurrence rates, co days, small planets are more ExoSoCal 2018

16 Updated stellar and planetary radii using Gaia DR2 CKS-Gaia Table 1. Error Budget Parameter Median Uncert. Te 60 K mk 0.02 mag AK mag µ 0.01 mag BC 0.03 mag R? 2.2% Rp /R? 4.1% Rp 4.9% PULATION Detected Planets dii we derived planet radii ulated in Mullally et al. incident stellar flux Sinc These Rp and Sinc mea3. Figure 4 shows the P -Rp and Sinc -Rp planes. rrow gap separating two R. While the gap is of 1944 planets, we cauetected planets does not tion of planets, due to sen Section 4.2. ution of Planets BJ Fulton currence, the number of Summary Little change to 1D 5 radius distribution More massive stars = larger and hotter 3. Stellar effective temperature. We restricted our planets planet sample to stars with T = K, the CKS temperatures are reliable. Gapwhere widens and moves to larger radii for more 4. Stellar dilution (Gaia). Dilution from nearby stars stars can also alter the apparent massive planetary radii. For e each target, we queried all Gaia sources within 8 arcsec (2 Kepler pixels) and computed the sum of their G-band fluxes. The ratio between this cumulative flux and the target flux r8 approximates the Kp-band dilution for each transiting planet. We required that r8 < Stellar dilution (imaging). Furlan et al. (2017) compiled high-resolution imaging observations performed by several groups. When a nearby star is detected, Furlan et al. (2017) computed a radius correction factor (RCF), which accounts for dilution assuming the planet transits the bright- ExoSoCal 2018

17 Backup Slides

18 Spectroscopy + Parallax Stefan-Boltzmann Law Teff from CKS spectra bolometric correction single mag extinction Distance mod. (parallax)

19 The Gap is Not Empty Simple toy model: Count number of planets in several boxes Simulate distributions of planets Figure 6. Left: Two-dimensional distribution of planet size and orbital period. The median uncertainty is plotted in the upp left. Right: same as left but with insolation flux on the horizontal axis. In both plots, the two peaks in the population observed by F17 are clearly visible, but with greater fidelity. Compare simulations to real detections Fulton & Petigura (2018) ExoSoCal 2018 Figure 7. Toy model demonstrating that the two populations of planets have intrinsic widths. Left: Real planet detections w BJ Fulton

20 Photoevaporation Owen & Wu (2017)

21 Stellar Mass Dependence 8 Owen, J. E. & Murray-Clay R. 6 All Periods Planets orbiting more massive stars are, on average, larger and hotter Caveat: stellar mass is correlated with both metallicity and age Radius [R ] Periods < 2.5 days Owen & Murray-Clay (2018) Normalised Planet Density Fulton et al. Radius [R ] Figure 9. Same as Figure 6, except th ets whose host stars have masses in t and metallicities > 0.2. and low mass stars (split at the med 1.02 M ). Small number statistics robust statements, but there is no evidence of a clear trend with ste Periods > 25 days 6 cautious optimism to presume that Wu (2018) bigger (more massive) solid cores a stars is robust. We need to be more cautiou small planets at long periods are mo metallicity stars. There is weak, bu evidence, that both sub-samples c in metallicity, but wider range in in stellar mass, but a wider range trends that could be responsible for servations that small planets are m riods around CKS stars with lower here we do not have the sample siz Metallicity the driver; one would suspect that Additionally, one might wor metallicity stars are more massive Figure 8. Same a Figure 4, but only for planets whose host stars the CKS sample, small planets at la have masses in the range M and metallicities > 0.2. observe around these objects. Petigu Fulton & Petigura (2018) suring absolute occurrence rates, co Figure 9. Mean planet properties as a function of mean stellar mass for super-earths and days, sub-neptunes (left and right Fig. 5. Planet sizes have no correlation with host star metal- work are safe if one is really interested in how planet forma small planets are more 5 Radius [R ] Mass and Scalings for Super-Earths BJ Fulton ratios in the size histogram licity. There are 1141 planets in GKS that overlap with the CKS ExoSoCal 2018

22 0.3/0.7 Fe/MgSiO3 +0.2% H/He +2% H/He +10% H/He BJ Fulton Fulton, Petigura, et al. (submitted) ExoSoCal 2018

23 The Radius Gap BJ Fulton Fulton, Petigura, et al. (2017) ExoSoCal 2018

24 The Radius Gap BJ Fulton Fulton, Petigura, et al. (2017) ExoSoCal 2018

25 The Radius Gap USPs (P < 1 d) BJ Fulton Fulton, Petigura, et al. (2017) ExoSoCal 2018

26 The Radius Gap BJ Fulton Fulton, Petigura, et al. (2017) ExoSoCal 2018