Towards understanding stellar variability at the sub m/s level: isolating granulation signals in synthetic spectral lines

ABSTRACT Granulation in the photospheres of FGK-type stars induces variability in absorption lines, complicating exoplanet detection via radial velocities and characterization via transmission spectroscopy. We aim to quantify the impact of granulation on the radial velocity and bisector asymmetry of stellar absorption lines of varying strengths and at different limb angles. We use 3D radiation-hydrodynamic simulations from MURaM paired with MPS-ATLAS radiative transfer calculations to synthesize time series for four Fe i lines at different limb angles for a solar-type star. Our line profiles are synthesized at an extremely high resolution (R = 2000 000), exceeding what is possible observationally and allowing us to capture intricate line shape variations. We introduce a new method of classifying the stellar surface into three components and use this to parametrize the line profiles. Our parametrization method allows us to disentangle the contributions from p-modes and granulation, providing the unique opportunity to study the effects of granulation without contamination from p-mode effects. We validate our method by comparing radial velocity power spectra of our granulation time series to observations from the Laser-based Absolute Reference Spectrograph. We find that we are able to replicate the granulation component extracted from observations of the Fe i 617 nm line at the solar disc centre. We use our granulation-isolated results to show variations in convective blueshift and bisector asymmetry at different limb angles, finding good agreement with empirical results. We show that weaker lines have higher velocity contrast between granules and lanes, resulting in higher granulation-induced velocity fluctuations. Our parametrization provides a computationally efficient strategy to construct new line profiles, laying the groundwork for future improvements in mitigating stellar noise in exoplanet studies.