Investigating exoplanet systems using the Rossiter-McLaughlin effect

: through the application of least-squares deconvolution

Abstract

The Rossiter-McLaughlin (RM) effect describes the distortion of a stellar line profile and the apparent shift in the star’s radial velocity (RV) that occurs when a planet transits a rotating star. Classical RM looks at the RV shift whereas line profile tomography investigates the distortion of the line profile. In both methods, the accuracy of the stellar line profile is essential and would depend on the technique used to obtain it. Typically, Cross-correlation functions (CCFs) are used, however, their shape can be distorted by line blending and aliasing. Least-squares deconvolution (LSD) offers an alternative that directly fits the mean line profile of the spectrum. In this thesis, I introduce a novel method, ACID (Accurate Continuum fItting and Deconvolution), that builds on LSD techniques by simultaneously fitting the spectral continuum and line profile and performing LSD in effective optical depth. The application of ACID to HD189733b showed improved representation of the stellar line profile for slow-rotating stars. This thesis also outlines a line profile tomography model that was optimized throughout this work for use on a Graphics Processing Unit (GPU), performing over 12 times faster than the CPU model. Following this, an in-depth investigation of the RM effect of HD189733b was conducted using both CCFs and ACID profiles including classical RM, Reloaded RM and line profile tomography analysis. The chromatic RM effect is also investigated to verify the presence of a Rayleigh slope. Furthermore, ACID was applied to ExoGemS (Exoplanets with Gemini Spectroscopy) data, obtained with GRACES for WASP-76b, HAT-P-30b, WASP-140b and WASP-177b. In all applications, large variations were observed in the measured RVs which could result from the spectrograph not being stabilised. Similarly, HIP41378f, a possible ringed planet, was also investigated using ACID with NRES (Network of Robotic Echelle Spectrographs) data, however, this also highlighted issues with the data.

Date of Award	Jul 2024
Original language	English
Awarding Institution	Queen's University Belfast
Supervisor	Ernst de Mooij (Supervisor) & Christopher Watson (Supervisor)