Spectral Characterization


Dozens of planetary mass objects have been resolved from their host stars at distances > 5 AU (Pepe, Ehrenreich, & Meyer, 2014). Spectra of these objects enable us to estimate temperatures, surface gravity, and composition of the objects in order to further characterize their physical properties, as well as test specific models of planet formation. Gas giant planets in our solar system are enriched in heavy elements and thought to have formed by core accretion. Gas giants formed through gravitational instability might be expected to have composition comparable to that of their host stars. Comparing observed constraints on key diagnostics such as the C/O ratio from resolved planetary mass companions to those measured for the central star help us assess models for formation and evolution.

Modeling spectra

Sophisticated forward models of the spectra of brown dwarfs and gas giant planets have provided fundamental insights into the structure and evolution of extremely cool astrophysical compact objects (temperatures < 2500 Kelvins). However high signal to noise spectra now obtainable with current instrumentation on 6-10 meter ground-based telescopes, as well as the Hubble Space Telescope (and soon the James Webb Space Telescope) call for a new approach. Spectral retrieval models, based on simple atmospheric models (e.g. plane-parallel 1D pressure-temperature profiles) enable millions of models to be calculated quickly, for comparison to observational data. Using techniques such as MCMC to efficiently sample the sample enables us to infer probability distribution functions for all key parameters including effective temperature, surface gravity, and column densities of key molecular species such as H2O, CO, CO2, HNC, and NH3. Constraining these major carriers of volatile elements C, N, and O in the gas phase are vital to assess the C/O ratio and test models of planet formation.

What are we looking for?

Specifically, we are searching for correlations in temperature and gravity, as a function of planet mass (relative to host star), orbital separation, and molecular abundances.

We are involved in several research programs in this area including:

1. Characterization of known planetary companions with ground-based photometry from 1-10 microns using adaptive optics assisted imaging (in collaboration with S. Quanz, ETH Z├╝rich).

2. Obtaining spectra of widely separated planetary companions from the Hubble Space Telescope in the 1.4 micron spectral region with WFC3,  preparing guaranteed time observation programs with the James Webb Space Telescope (NIRCam and NIRISS), and planning for Cycle 1 proposals with JWST.

3. Obtaining spectra of planetary companions with extreme AO systems including SPHERE, GPI, and MagAO (e.g. Zurlo et al. 2016).

Figure 10 from Zurlo et al. 2016, A&A, showing spectra from both SPHERE and GPI of planet HR8799d, comparing with isolated brown dwarf spectra.

4. Developing spectral retrieval models in order to constrain companion properties (e.g. Todorov et al. 2016; Howe et al. 2017).

Figure 2 from Todorov et al. 2016 comparing the observed spectrum of Kappa And b (black points from Hinkley et al. 2013) with spectrum retrieval (orange lines).