Modelling the effect of 3D temperature and chemistry on the cross-correlation signal of transiting ultra-hot Jupiters: a study of five chemical species on WASP-76b

Ultra-hot Jupiters are perfect targets for transmission spectroscopy. However, their atmospheres feature strong spatial variations in temperature, chemistry, dynamics, cloud coverage, and scale height. This makes transit observations at high spectral resolution challenging to interpret. In this work, we model the cross-correlation signal of five chemical species - Fe, CO, H₂O, OH, and TiO - on WASP-76b, a benchmark ultra-hot Jupiter. We compute phase-dependent high-resolution transmission spectra of three-dimensional (3D) SPARC/MITgcm models. The spectra are obtained with gCMCRT, a 3D Monte-Carlo radiative-transfer code. We find that, on top of atmospheric dynamics, the phase-dependent Doppler shift of the absorption lines in the planetary rest frame is shaped by the combined effect of planetary rotation and the unique 3D spatial distribution of chemical species. For species probing the dayside (e.g. refractories or molecules like CO and OH), the two effects act in tandem, leading to increasing blueshifts with orbital phase. For species that are depleted on the dayside (e.g. H₂O and TiO), the two effects act in an opposite manner, and could lead to increasing redshifts during the transit. This behaviour yields species-dependent offsets from a planet's expected K_p value that can be much larger than planetary wind speeds. The offsets are usually negative for refractory species. We provide an analytical formula to estimate the size of a planet's K_p offsets, which can serve as a prior for atmospheric retrievals. We conclude that observing the phase-resolved absorption signal of multiple species is key to constraining the 3D thermochemical structure and dynamics of ultra-hot Jupiters.