Exoplanet Interiors

The field of exoplanets has made tremendous advances since the first discovery of a planet orbiting a distant sun-like star in 1995. More than 800 exoplanets have been detected to date, and observational techniques have improved sufficiently to allow for the detection planets smaller than Neptune (4 Earth-radii, 17 Earth-masses). Sub-Neptune planets are intriguing because they are abundant in exoplanetary systems, but unrepresented within the Solar System. Although we cannot (yet) send a rover or satellite to observe these distant planets directly, there is still a great deal of information we can gather about these bodies. The mass of a planet can be estimated from the radial velocity wobble its gravitational pull induces in its host star. The radius can also be measured if the planet transits, crossing in front of the disk of its star and decreasing the apparent brightness of the star as observed from Earth. Planets detected through both radial velocity and transit observations are especially valuable, since with both mass and radius measurements we can gauge the planet's average density and learn about its interior composition.

The main objective of my research is to interpret the measured masses and radii of sub-Neptune exoplanets, employing models to constrain the planets' bulk compositions, formation histories, and habitability.
Composition Degeneracies: 
I have developed a computer model for the internal structure of low-mass exoplanets (Rogers & Seager 2010a). I consider differentiated planets consisting of up to four layers: an iron core, silicate mantle, ice layer, and gas layer. For a given planet mass and radius, the computer model outputs all possible ways that the planet's mass can be distributed among the four layers. Although it is impossible to pinpoint the unique true composition of a planet from radial velocity and transit observations alone, I can nonetheless use my models to quantify the span of plausible compositions. I have applied my model to constrain bulk compositions of low-mass transiting planets CoRoT-7b, GJ 436b, HAT-P-11b, and GJ 1214b. 

In December 2009, when the MEarth survey announced the discovery of GJ 1214b, the coldest transiting super-Earth yet known, I was first to interpret the measured planet mass and radius (Rogers & Seager 2010b). GJ 1214b is intriguing because, at 6.5 Earth masses and 2.7 Earth radii, its average density is low enough that it almost certainly has a gas layer. I considered three possible sources for the gas layer on GJ 1214b: direct accretion of H/He gas from the protoplanetary nebula, sublimation of ices, and outgassing of volatiles from a rocky interior. Depending on which gas layer source dominates, GJ 1214b could have very different interior properties: it could be a miniature Neptune with an interior of ice and rock surrounded by a primordial H/He envelope containing 0.01%-5% of the planet mass, or a water planet (composed of at least 47% H2O by mass) shrouded in a layer of vapor from sublimated ices, or a terrestrial super-Earth harboring a hydrogen-rich outgassed atmosphere. In all three scenarios, I found that GJ 1214b would be too hot to have liquid water under most conditions. GJ 1214b would need to be releasing no more than 5x10-4 W/m2 from its interior (less than one hundredth of the Earth’s interior heat flux) for its envelope to pass through the pressure-temperature regime of liquid water. My case study helped to motivate follow-up observations of GJ 1214b with the Spitzer Space Telescope, Hubble Space Telescope, and ground based observatories to discriminate between the composition scenarios I presented.

Low-Density Exo-Neptunes:
The Kepler Space Telescope's discovery of hundreds of Neptune-size (2 - 6 Earth-radii) planet candidates was an unexpected surprise for planet formation theories. Assuming that many of these candidates are true planets, what are they, how did they form, and why are they so numerous? In collaboration with Peter Bodenheimer from UCSC and Jack Lissauer from NASA Ames, I explored both core-nucleated accretion and outgassing as two separate formation pathways for Neptune-size planets with voluminous atmospheres of light gases (Rogers et al. 2011). We demonstrated, with the most recent version of the UCSC-Ames planet accretion code, that planets 3-8 Earth-masses with substantial H/He envelopes can plausibly form by core-nucleated accretion beyond the snow line and migrate to equilibrium temperatures above 500 K given reasonable protoplanetary disk surface densities and disk dissipation timescales. We also derived a limiting low-density mass-radius relation for rocky planets with outgassed hydrogen envelopes, demonstrating that outgassed terrestrial planets will not account for the Kepler planet candidates larger than about 3 Earth-radii.

Most Kepler transiting planet candidates orbit stars that are too faint for radial velocity measurements of the planet mass. Using theoretical models of planet formation, structure and survival, my collaborators and I constrained the minimum plausible planet mass as a function of planet radius and equilibrium temperature (Rogers et al. 2011). Our main conclusion was that Neptune-size (2 - 6 Earth-radii) planets at equilibrium temperatures Teq=500-1000 K can potentially have masses less than 4 Earth-masses (many times smaller than Neptune's mass of 17 MEarth). The possibility for low-mass low-density planets has important implications for the interpretation of the Kepler planet radius distribution, and for the radial velocity precision required to confirm planet candidates from transit surveys such as Kepler, HAT, WASP, and MEarth. 

Liquid Water Oceans on Super-Earths:
With models of planet interior structure and climate, I have comprehensively explored and constrained the scenarios in which a sub-Neptune planet could harbor a liquid water ocean on its surface. Whereas the classical studies of the habitable zone for Earth-like planets have assumed H2O-CO2 dominated atmospheres, planet energy budgets dominated by incoming starlight, and gas layers with negligible contributions to the planet radius, these assumptions can break-down for sub-Neptune planets. In my more general treatment of the habitable zone, I consider the possibility of hydrogen-rich atmosphere compositions, take into account the intrinsic planet luminosity (from radioactive heating or release of left-over heat from accretion), and compute the radial extent of the gas envelope. By combining my planet interior structure model with a one-dimensional radiative-convective semi-grey climate model, I have constrained the combinations of observable planet properties (mass, radius, orbital period, stellar spectral type) that are conducive to liquid water oceans. An important result from this work is the first upper limit on the transit radius of habitable planets (as a function of planet mass and equilibrium temperature). None of the currently known planets with both mass and radius measurements could harbor liquid water oceans, but as planet searches push toward smaller planets at greater orbital distances my doctoral thesis results will be a useful tool to identify the most promising candidate habitable planets for follow-up observations and detailed climate modeling.

Planet Microlensing

Microlensing is an approach to discover exoplanets that relies upon chance alignments between two stars along the line of sight to Earth. The gravitational field of the foreground star acts as a lens, magnifying the light from the background star. Anomalies in the magnification light curve can reveal planets orbiting the lens star. I have collaborated with Paul Schechter to calculate cross-sections for various categories of planetary microlensing events. Here is a pedagogical microlensing poster I made for the IAU 276 Symposium.

Interacting Binary Stars

I became interested in transiting white dwarfs after Kepler discovered two very unusual transiting objects in its first 43 days of science photometry. These curious transiting companions, named KOI-81 and KOI-74, are planet-sized (with radii similar to Jupiter), while also being hotter than their stellar hosts (with effective temperatures in excess of 10,000K). KOI-81 and KOI-74 are likely white dwarfs – the degenerate cores of dead stars that have lost their outer layers. With Saul Rappaport, I am performing a stellar population synthesis calculation to compute occurrence rates for white dwarfs transiting main sequence stars, and to predict the statistical distributions of masses, periods, and transit depths for these binary star systems. Comparing the outcomes of this calculation to the eventual full Kepler transit sample will help to confirm whether KOI-81 and KOI-74 are indeed white dwarfs, and may lead to new insights into the physics governing interacting stellar binaries.

Meteoroid Ablation

Observations of the 1998 Leonid meteor storm yielded exceptional meteors that have helped to push the boundaries of meteoroid ablation theory. While conventional meteor ablation altitudes range between 80−125 km, meteors having high beginning heights up to 200 km were detected. Whereas other observational studies have found meteors with transverse spread on the order of meters, very wide meteors exhibiting comet-like diffuse structures on the order of a kilometer were detected among the 1998 Leonids. Meteoroid ablation models have typically assumed that the destruction of a meteoroid in the Earth’s atmosphere is primarily a thermally driven process: the meteoroid evaporates following intense heating during atmospheric flight. Light is emitted when excited atomic and molecular states produced by collisions between ablated meteor atoms and atmospheric constituents decay. However, thermal ablation cannot account for the exceptional meteors observed during the 1998 Leonid meteor storm.

With Dr. Robert Hawkes, at Mount Allison University in New Brunswick Canada, I explored physical sputtering as a possible meteoroid ablation mechanism. I demonstrated that non-thermal sputtering could explain the extreme starting heights (Rogers et al. 2005, Hill et al. 2004), but not the extreme transverse widths (Rogers, L.A. 2006, Honours Thesis), of meteors observed during the 1998 Leonid meteor storm. I also quantified the observational biases affecting the detection of high-velocity meteors of interstellar origin (Rogers et al. 2004, Hill et al. 2005).