Two populations are better than one: Short gamma-ray bursts from SGR giant flares and NS-NS mergers

TL;DR

This paper investigates the origins of short gamma-ray bursts by modeling two potential progenitor populations—magnetar giant flares and neutron star mergers—and finds that a dual-population model best explains observed data.

Contribution

It introduces a dual-population model with separate luminosity functions for SGR giant flares and NS-NS mergers, fitting observed sGRB distributions better than single-population models.

Findings

A dual-population model reproduces sGRB distributions and counts.

Best-fit luminosity functions align with observed luminosities.

Only combined populations match the local and overall sGRB data.

Abstract

With a peak luminosity of ~10^47 erg/s, the December 27th 2004 giant flare from SGR1806-20 would have been visible by BATSE (the Burst and Transient Source Experiment) out to ~50 Mpc. It is thus plausible that some fraction of the short duration Gamma-Ray Bursts (sGRBs) in the BATSE catalogue were due to extragalactic magnetar giant flares. According to the most widely accepted current models, the remaining BATSE sGRBs were most likely produced by compact object (neutron star-neutron star or neutron star-black hole) mergers with intrinsically higher luminosities. Previously, by examining correlations on the sky between BATSE sGRBs and galaxies within 155 Mpc, we placed limits on the proportion of nearby sGRBs. Here, we examine the redshift distribution of sGRBs produced by assuming both one and two populations of progenitor with separate Luminosity Functions (LFs). Using the local Galactic SGR giant flare rate and theoretical NS-NS merger rates evolved according to well-known Star Formation Rate parameterisations, we constrain the predicted distributions by BATSE sGRB overall number counts. We show that only a dual population consisting of both SGR giant flares and NS-NS mergers can reproduce the likely local distribution of sGRBs as well as the overall number counts. In addition, the best fit LF parameters of both sub-populations are in good agreement with observed luminosities.