According to dynamo theory, stars with convective envelopes efficiently generate surface magnetic fields, which manifest as magnetic activity in the form of starspots, faculae, and/or flares, when their rotation period is shorter than their convective turnover time. Most red giants, having undergone significant spin down while expanding, have slow rotation and no spots. However, based on a sample of about 4500 single and binary red giants observed by the NASA Kepler mission, it has been observed that about 8% of them display spots, about 15% of which belong to close binary systems.

We shed light on a puzzling fact: for rotation periods less than 80 days, a red giant that belongs to a close binary system displays a photometric modulation about an order of magnitude larger than that of a single red giant with a similar rotational period and similar physical properties. We investigate whether binarity leads to larger magnetic fields when tides lock systems, or if a different spot distribution on single versus close binary stars can explain this fact. For this, we use the aforementioned sample of about 4500 red giants to measure several chromospheric activity indicators probing different heights in the chromosphere: the near-ultraviolet (NUV) excess by using photometric data from GALEX, as well as chromospheric indices from the Hα, MgI, CaII H&K and infared CaII spectral lines by using spectroscopic data from LAMOST.

We show that red giants in a close-binary configuration with spin-orbit resonance display significantly larger chromospheric activity levels than single red giants or red giants in binary systems that do not have any special tidal configuration. This result suggests that tidal locking leads to larger magnetic fields, a possible candidate mechanism being the elliptical instability. This allows us to provide criteria to classify active red giants (single or binary), based on their rotation period and magnetic activity indices. In addition, we strikingly observe that the depths of the MgI and Hα lines are clearly correlated with the amplitude of solar-like oscillations for a given surface gravity log(g), independently of the detection of photometric rotational modulation caused by spots on the photosphere. This result opens up future possibilites of estimating the value of magnetic fields at the surface of red giant stars, whether quiet or active, by combining spectroscopy and asteroseismology with three-dimensional atmospheric models including radiative transfer.