Determination of rotation periods in solar-like stars with irregular sampling: the Gaia case

TL;DR

This study evaluates the ability of Gaia's irregular sampling to accurately determine rotation periods of solar-like stars, revealing a strong dependence on ecliptic latitude and rotation period, with detection rates varying significantly.

Contribution

It introduces a comprehensive simulation framework to assess Gaia's effectiveness in measuring stellar rotation periods considering irregular sampling and stellar activity.

Findings

Detection rate peaks at 70% around ecliptic latitude 45° for P ~ 1 day

Detection rates drop below 10% at high ecliptic latitudes for short periods

Longer periods (>5 days) have very low detection success, around 5% on average

Abstract

We present a study on the determination of rotation periods (P) of solar-like stars from the photometric irregular time-sampling of the ESA Gaia mission, currently scheduled for launch in 2013, taking into account its dependence on ecliptic coordinates. We examine the case of solar-twins as well as thousands of synthetic time-series of solar-like stars rotating faster than the Sun. In the case of solar twins we assume that the Gaia unfiltered photometric passband G will mimic the variability of the total solar irradiance (TSI) as measured by the VIRGO experiment. For stars rotating faster than the Sun, light-curves are simulated using synthetic spectra for the quiet atmosphere, the spots, and the faculae combined by applying semi-empirical relationships relating the level of photospheric magnetic activity to the stellar rotation and the Gaia instrumental response. The capabilities of the Deeming, Lomb-Scargle, and Phase Dispersion Minimisation methods in recovering the correct rotation periods are tested and compared. The false alarm probability (FAP) is computed using Monte Carlo simulations and compared with analytical formulae. The Gaia scanning law makes the rate of correct detection of rotation periods strongly dependent on the ecliptic latitude (beta). We find that for P ~ 1 d, the rate of correct detection increases with ecliptic latitude from 20-30 per cent at beta ~ 0{\deg} to a peak of 70 per cent at beta=45{\deg}, then it abruptly falls below 10 per cent at beta > 45{\deg}. For P > 5 d, the rate of correct detection is quite low and for solar twins is only 5 per cent on average.