2016ApJ...823...14T

Spectral retrieval has proven to be a powerful tool for constraining the physical properties and atmospheric compositions of extrasolar planet atmospheres based on observed spectra, primarily for transiting objects but also for directly imaged planets and brown dwarfs. Despite its strengths, this approach has been applied to only about a dozen targets. Determining the abundances of the main carbon- and oxygen-bearing compounds in a planetary atmosphere can lead to the C/O ratio of the object, which is crucial for understanding its formation and migration history. We present a retrieval analysis of the published near-infrared spectrum of κ  Andromedae b, a directly imaged substellar companion to a young B9 star. We fit the emission spectrum model utilizing a Markov Chain Monte Carlo algorithm. We estimate the abundance of water vapor, and its uncertainty, in the atmosphere of the object. In addition, we place an upper limit on the abundance of CH₄. We qualitatively compare our results with studies that have applied model retrieval on multiband photometry and emission spectroscopy of hot Jupiters (extrasolar giant planets with orbital periods of several days) and the directly imaged giant planet HR 8799b.