Composition of Sub-Saturns

One of the most powerful aspects of exoplanet research is the ability to study types of planets not represented in our own Solar System. Our Sun hosts no planets between Uranus and Saturn size (4.0–9.4 Earth-radii). The few well-characterized sub-Saturns span a wide range of planet densities. For example, HD-149026b and Kepler-79d are both about 8 times the size of the Earth. However, HD-149026b is 20 times denser than Kepler-79d, which having a mean density of 0.07 g/cc, is as dense as Styrofoam. What aspects of planet formation physics account for this diversity of planet composition? To find out, I’m leading an observational campaign to detect sub-Saturns using K2 and to measure their masses with Keck/HIRES. Here, you can read about K2-24, a star that hosts two low-density sub-Saturns caught in the 2:1 mean-motion resonance (Petigura et al. 2016).

Artists conception of a protoplanetary disk in the process of forming planets. Studying the properties of Sub-Saturn planets informs our theories of how planets form and acquire gas from their protosolar nebulae. Credit: NASA/FUSE/Lynette Cook

Small Planets, Small Stars

Planets around small stars are prime targets for further characterization. A planet transiting a small star blocks a greater fraction of the star’s light, which makes them more amenable to transmission spectroscopy from HST and soon JWST. For example, an Earth-size planet transiting a 4000 K M0 dwarf, produces a dimming that is 4 times deeper than the same planet transiting a 5700 K G2 star like the Sun.

The prime Kepler mission (2009–2013) was designed to measure the frequency of Earth-size planets around Sun-like stars. As a result, Kepler surveyed only about 4000 M-dwarfs. However, the ongoing K2 mission (2014–present) will observe 10 times more M dwarfs than Kepler, casting a wider net for these exciting planets. The plot below shows an early K2 discovery made by my team. K2-3 is a M0 star half the size of the sun hosting three transiting planets all 2 Earth-radii or smaller. Planet d, the outermost planet, receives 50% more light from it’s host star than the Earth. At only 45 pc away, K2-3d is the nearest Earth-size planet in the habitable zone.

Transits of K2-3 b, c, and d, planets between 1.5 and 2.0 times the size of the Earth, transiting a cool star about half the size of the sun (Crossfield, Petigura, et al. 2015)

Transits of K2-21 b and c. Like the K2-3, this star is a cool M-dwarf hosting multiple transiting planets 2 Earth-radii and smaller. These planets have a peculiar period ratio: P(outer) / P(inner) = 1.6624. These planets are extremely close to the 5:3 mean-motion resonance, and will likely exhibit strong transit-timing variations (Petigura et al. 2015).

Prevalence of Earth-size Planets

One of the prime goals of the Kepler mission was to measure the frequency of Earth-like planets around Sun-like stars. Not only does this measurement require detecting a large sample of planets, but it also requires carefully understanding the number of planets missed in the survey. In order to measure the occurrence of planets from the Kepler mission, I searched for planets in the Kepler photometry using a custom photometric pipeline (TERRA), and measured survey completeness by injecting synthetic planets into the real Kepler a photometry, and measuring the recovery rate. Key results from this work include:

  1. Nearly every star (74%) has at least one planet, just within 1 AU.
  2. Small planets are common. Earth-size planets out number Jovians by 16 to 1.
  3. With modest extrapolation, one can estimate the fraction of Sun-like stars hosting an Earth-size (1–2 Earth-radii) planet in the Habitable Zone (defined as the range of orbits where a planet can plausibly support liquid water oceans). We find that 22% of Sun-like stars host an Earth-size planet that receives the same light energy from its host star as Earth, to a factor of four. The boundaries of the habitable zone are extremely uncertain, but point is that “eta-earth” is tens of percent. Not one in a million.

This work was awarded the Cozzarelli Prize by the National Academy of Sciences as 2013’s top paper in the physical sciences.

Fraction of stars with planets having periods between 5–100 days as a function of planet size. Planets (especially small planets) are common outcomes of star formation (Petigura, Howard, Marcy 2013).