Limits on the population of repeating fast radio bursts from the ASKAP/CRAFT lat50 survey

Abstract

A method is presented to limit the volumetric density of repeating fast radio bursts based on the number (or lack) of repeating bursts identified in a survey. The method incorporates the instantaneous sensitivity of the instrument, its beam pattern, and the dwell time per pointing, as well as the energy and timing distribution of repeat bursts. Applied to the Australian Square Kilometre Array Pathfinder's (ASKAP's) Commensal Real-time ASKAP Fast Transients (CRAFT) `lat50' survey, the presence of an FRB similar to FRB 121102 is excluded within a volume of $1.9 \cdot 10^6$ Mpc$^3$ at 95% confidence level (C.L.). Assuming a burst energy cut-off at $10^{42}$ erg, the 95% C.L. upper limit on the population density of repeating FRBs in the current epoch is $27$ Gpc$^{-3}$, assuming isotropic (unbeamed) emission. This number is much lower than expected from even rare scenarios such as magnetar formation in gamma-ray bursts. Furthermore, the maximally allowed population under-predicts the observed number of single bursts in the survey. Comparisons with the observed dispersion measure distribution favours a larger population of less rapidly repeating objects, or the existence of a second population of non-repeating FRBs. In any scenario, FRB 121102 must be an atypical object.