The ESPRESSO transmission spectrum of HD$\,$189733$\,$b : Extracting the planetary sodium and lithium signatures amid stellar contamination

TL;DR

This study uses high-resolution ESPRESSO spectra to detect and analyze sodium and lithium in HD189733b's atmosphere, emphasizing the importance of correcting for stellar contamination effects like POLDs and stellar activity.

Contribution

It demonstrates the impact of planetary occulted line distortions on transmission spectra and provides refined measurements of sodium and lithium signatures after accounting for stellar effects.

Findings

Confirmed sodium detection in the planetary atmosphere.

Detected lithium signature at 6.4 sigma significance.

Highlighted the importance of correcting stellar contamination effects.

Abstract

While transmission spectroscopy has allowed us to detect many atomic and molecular species in exoplanet atmospheres, the improvement in resolution and signal-to-noise ratio enabled us to become sensitive to planet-occulted line distortions (POLDs) in the spectrum that are induced by center-to-limb variations and the Rossiter-McLaughlin effect. POLDs can bias the interpretation of the transmission spectrum, and it is difficult to correct for them with stellar models. We analyzed two ESPRESSO transits (R $\sim$ 140$\,$000) of the archetypal hot Jupiter HD$\,$189733$\,$b. The transmission spectrum of this aligned system is heavily affected by POLDs, stellar activity, and instrumental effects. It is therefore a challenging study case of how to account for these effects when the planetary signal is retrieved from chemical species through transmission spectroscopy. We confirm the previous detections of the sodium doublet signature in the upper atmosphere of HD$\,$189733$\,$b. When we accounted for POLDs and isolated the planetary signal from uncorrected stellar residuals, we found a shallower (0.432 $\pm$ 0.027 %) and more strongly blueshifted (-7.97 $\pm$ 0.28 km/s) signal. We attempted to reinterpret the other high-resolution sodium studies of this system in light of our results. We suggest that the POLDs and stellar activity are insufficiently corrected for in all analyses, including ours. We also detected a planetary lithium signature of 0.102 $\pm$ 0.016 % (6.4$\sigma$) at a blueshift of -2.4 $\pm$ 1.8 km/s.