学术文献库

In photometry, the short-timescale stellar variability ("flicker"), such as that caused by granulation and oscillations, can reach amplitudes comparable to the transit depth of Earth-sized planets and is correlated over the typical transit timescales. It can introduce systematic errors on the inferred planetary parameters when a small number of transits are observed. The objective of this paper is to characterize the statistical properties of this noise and quantify its impact on the inferred transit parameters. We used the extensive solar observations obtained with SoHO/VIRGO to characterize flicker noise. We simulated realistic transits across the solar disk using SDO/HMI data and used these to obtain transit light curves, which we used to estimate the errors made on the transit parameters. We make these light curves publicly available. To extend the study to a wider parameter range, we derived the properties of flicker noise using Kepler observations and studied their dependence on stellar parameters. Finally, we predicted the limiting stellar apparent magnitude for which the properties of the flicker noise can be extracted using high-precision CHEOPS and PLATO observations. Stellar granulation is a stochastic colored noise, and is stationary with respect to the stellar magnetic cycle. Both the flicker correlation timescales and amplitudes increase with the stellar mass and radius. If these correlations are not taken into account when fitting for the parameters of transiting exoplanets, this can bias the inferred parameters. In particular, we find errors of up to 10$\%$ on $R_p/R_s$ for an Earth-sized planet orbiting a Sun-like star. For F and G stars, flicker will significantly affect the inferred parameters of transits observed with CHEOPS and PLATO. Dedicated modeling strategies need to be developed to accurately characterize both the star and the transiting exoplanets.