2020MNRAS.494.3766O

Detections of stellar coronal mass ejections (CMEs) are still rare. Observations of strong Balmer line asymmetries during flare events have been interpreted as being caused by CMEs. Here, we aim to estimate the maximum possible Balmer line fluxes expected from CMEs to infer their detectability in spectroscopic observations. Moreover, we use these results together with a model of intrinsic CME rates to infer the potentially observable CME rates for stars of different spectral types under various observing conditions, as well as the minimum required observing time to detect stellar CMEs in Balmer lines. We find that generally CME detection is favoured for mid- to late-type M dwarfs, as they require the lowest signal-to-noise ratio for CME detection, and the fraction of observable-to-intrinsic CMEs is largest. They may require, however, longer observing times than stars of earlier spectral types at the same activity level, as their predicted intrinsic CME rates are lower. CME detections are generally favoured for stars close to the saturation regime, because they are expected to have the highest intrinsic rates; the predicted minimum observing time to detect CMEs on just moderately active stars is already >100 h. By comparison with spectroscopic data sets including detections as well as non-detections of CMEs, we find that our modelled maximum observable CME rates are generally consistent with these observations on adopting parameters within the ranges determined by observations of solar and stellar prominences.