2022A&A...657L...7X

The total number of observed fast radio burst (FRB) events is rising rapidly thanks to the improvement of existing radio telescopes and the delivery of new facilities. In particular, the Five-hundred-meter Aperture Spherical radio Telescope Collaboration recently reported more than one thousand bursts in a short observing period of 47 days. The striking bimodal distribution in their work motivated us to revisit the definition of FRBs. In this work, we ascribe the bimodal distribution to two physical kinds of radio bursts that may exhibit different radiation mechanisms. We propose using brightness temperature to separate two subtypes. For FRB 20121102A, the critical brightness temperature is T_B, cri_ ≃ 10³³ K. Bursts with T_B ≥ T_B, cri_ are denoted as "classical" FRBs and we find a tight pulse width-fluence relation (T ∝ F_ν^0.306) for them. On the contrary, the other bursts are considered as "atypical" bursts that may have originated from a different type of physical process. We suggest that for each FRB event, a similar dividing line should exist but that the T_B, cri_ is not necessarily the same in such cases. Its exact value depends on the FRB radiation mechanism and the properties of the source.