2024MNRAS.530..457C

M dwarfs are often highly flare productive, and changes in the rate of flaring can potentially affect the radiation and energetic particles an exoplanet is exposed to. Rate variability has been found to be rare on G-type stars, and there has been limited study of the variability of flaring on M dwarfs. Using data from the TESS mission up to Observation Sector 66, we analyse 658 stars ranging from K7 to M5 to test for superflare rate variability by calculating an optimal piecewise constant-rate model for flare occurrence using a Bayesian algorithm and identify 66 stars with rate changes, a larger proportion than our previous study into G-type stars (7 out of 270 stars). We find that rate variability does not depend on the stellar parameters for our sample. We investigate how the flare-frequency distribution for these stars varies and whether this variation affects exoplanet habitability. No stars in our sample provide sufficient UV flux to support prebiotic chemistry. We investigate the phase distributions on the rate variable stars, finding 11 cases of potential phase dependence in flaring across seven stars. We also compare the Bayesian method to another measure of activity, the fraction of flux emitted through flares, and do not observe a consistent relationship between this flaring activity metric and the optimal Bayesian model. We discuss the possible reasons for the discrepancy and identify a problem with the flaring activity metric due to the power-law distribution of the flare events.