Statistical Analysis of Stellar Flares from the First Three Years of TESS Observations

In this paper, we study stellar light curves from the TESS satellite (Transiting Exoplanet Survey Satellite) for the presence of stellar flares. The main aim is to detect stellar flares using two-minutes cadence data and to perform statistical analysis. To find and analyze stellar flares we prepared automatic software WARPFINDER. We implemented three methods described in this paper: trend, difference, and profile fitting. Automated search for flares was accompanied by visual inspection. Using our software we analyzed two-minute cadence light curves of 330,000 stars located in the first 39 sectors of TESS observations. As a result, we detected over 25,000 stars showing flare activity with the total number of more than 140,000 flares. This means that about 7.7% of all the analyzed objects are flaring stars. The estimated flare energies range between $10^{31}$ and $10^{36}$ erg. We prepared a preliminary preview of the statistical distribution of parameters such as a flare duration, amplitudes and energy, and compared it with previous results. The relationship between stellar activity and its spectral type, temperature and mass was also statistically analyzed. Based on the scaling laws, we estimated the average values of the magnetic field strength and length of the flare loops. In our work, we used both single (about 60%), and double (about 40%) flare profiles to fit the observational data. The components of the double profile are supposed to be related to the direct heating of the photosphere by non-thermal electrons and back warming processes.