The environments of short-duration gamma-ray bursts and implications for their progenitors

The study of short-duration gamma-ray bursts (GRBs) experienced a complete revolution in recent years thanks to the discovery of the first afterglows and host galaxies starting in May 2005. These observations demonstrated that short GRBs are cosmological in origin, reside in both star forming and elliptical galaxies, are not associated with supernovae, and span a wide isotropic-equivalent energy range of similar to 10(48)- 10(52) erg. However, a fundamental question remains unanswered: What are the progenitors of short GRBs? The most popular theoretical model invokes the coalescence of compact object binaries with neutron star and/or black hole constituents. However, additional possibilities exist, including magnetars formed through prompt channels (massive star core-collapse) and delayed channels (binary white dwarf mergers, white dwarf accretion-induced collapse), or accretion-induced collapse of neutron stars. In this review I summarize our current knowledge of the galactic and sub-galactic environments of short GRBs, and use these observations to draw inferences about the progenitor population. The most crucial results are: (i) some short GRBs explode in dead elliptical galaxies; (ii) the majority of short GRBs occur in star forming galaxies; (iii) the star forming hosts of short GRBs are distinct from those of long GRBs, and instead appear to be drawn from the general field galaxy population; (iv) the physical offsets of short GRBs relative to their host galaxy centers are significantly larger than for long GRBs; (v) there is tentative evidence for large offsets from short GRBs with optical afterglows and no coincident hosts; (vi) the observed offset distribution is in good agreement with predictions for NS-NS binary mergers; and (vii) short GRBs trace under-luminous locations within their hosts, but appear to be more closely correlated with the rest-frame optical light (old stars) than the UV light (young massive stars). Taken together, these observations suggest that short GRB progenitors belong to an old stellar population with a wide age distribution, and generally track stellar mass. These results are fully consistent with NS-NS binary mergers and rule out a dominant population of prompt magnetars. However, a partial contribution from delayed magnetar formation or accretion-induced collapse is also consistent with the data. (C) 2010 Elsevier B.V. All rights reserved.