Day-side Fe I Emission, Day-Night Brightness Contrast and Phase Offset of the Exoplanet WASP-33b

TL;DR

This study detects Fe I emission from the day-side of exoplanet WASP-33b, revealing a thermal inversion and providing a new method to measure its brightness variation, day-night contrast, and phase offset using high-resolution spectroscopy.

Contribution

Introduces a novel approach to constrain exoplanet brightness variation and phase offset solely through high-resolution Doppler spectroscopy, confirming Fe I emission and thermal inversion in WASP-33b.

Findings

Fe I emission detected at >10.4σ significance

Night-side flux is less than 10% of day-side flux

Emission peak shifted westward of substellar point

Abstract

We report on Fe I in the day-side atmosphere of the ultra-hot Jupiter WASP-33b, providing evidence for a thermal inversion in the presence of an atomic species. We also introduce a new way to constrain the planet's brightness variation throughout its orbit, including its day-night contrast and peak phase offset, using high-resolution Doppler spectroscopy alone. We do so by analyzing high-resolution optical spectra of six arcs of the planet's phase curve, using ESPaDOnS on the Canada-France-Hawaii telescope and HDS on the Subaru telescope. By employing a likelihood mapping technique, we explore the marginalized distributions of parameterized atmospheric models, and detect Fe I emission at high significance ($>10.4\sigma$) in our combined data sets, located at $K_{\rm p}=222.1\pm0.4$ km/s and $v_{\rm sys}=-6.5\pm0.3$ km/s. Our values agree with previous reports. By accounting for WASP-33b's brightness variation, we find evidence that its night-side flux is $<10\%$ of the day-side flux and the emission peak is shifted westward of the substellar point, assuming the spectrum is dominated by Fe I. Our ESPaDOnS data, which cover phases before and after the secondary eclipse more evenly, weakly constrain the phase offset to $+22\pm12$ degrees. We caution that the derived volume-mixing-ratio depends on our choice of temperature-pressure profile, but note it does not significantly influence our constraints on day-night contrast or phase offset. Finally, we use simulations to illustrate how observations with increased phase coverage and higher signal-to-noise ratios can improve these constraints, showcasing the expanding capabilities of high-resolution Doppler spectroscopy.