Cloudy Atmospheres on Directly Imaged Exoplanets: The Need for Accurate Particulate Representation in Photopolarimetric Simulations

Abstract

Missions like the upcoming Roman Space Telescope and its follow-on missions, Habitable Exoplanet Observatory (HabEx) and the Large UV/Optical/IR Surveyor (LUVOIR), will provide direct imaging observations of stellar light reflected by exoplanets with successively closer orbits. The synergistic use of ground-based polarimeters like Gemini Planet Imager and Very Large Telescope/Spectro-Polarimetric High-contrast Exoplanet Research instrument (SPHERE) would allow us to characterize cloudy exoplanet atmospheres using spectropolarimetric direct imaging. We present an extension of our semianalytic 3D radiative transfer modeling framework for brown dwarfs to include stellar light reflected by exoplanets with cloudy atmospheres. Using Mie theory to compute scattering by cloud and haze consisting of spherical particles, we show that the currently widespread use of approximations like the scalar Two-Term Henyey–Greenstein or the vector Henyey–Greenstein Rayleigh (HGR) composite result in a blurring of the phase-dependent features of exoplanet lightcurves, causing a 10%–39% loss of sensitivity to atmospheric parameters in an average measurement for signal-to-noise ratios (S/Ns) between 5 and 500. The HGR approximation creates the misleading impression that clouds are as polarizing as Rayleigh scatterers, regardless of their droplet size. This not only causes significant errors in the scientific interpretation of polarimetric measurements, but also results in a negligible sensitivity of HGR simulations to polarization measurements at the S/Ns considered, whereas Mie simulations show a 10%–30% gain in parametric sensitivity through the addition of polarimetry.