Metallicity of Gamma-Ray Burst Progenitors: Connection between Star Formation and Gamma-Ray Burst Production

Long Gamma-Ray Bursts (GRBs) are the brightest astronomical transient events which provide us clues to study high redshift universe. It is broadly accepted that at least some of the long GRBs originate from dying massive stars. Therefore GRBs can be used as a tracer of star formation which are detectable to very high redshift (z > 10). However, studies using stellar evolution models suggest that GRBs do not simply trace star formation, but preferentially occur in low-metallicity environment. The observational evidence is still controversial. We need to understand the low-metallicity preference of long GRBs before using them as a tracer of star formation. Using cosmological smoothed particle hydrodynamic simulations, we compute the UV luminosity distribution of GRB host galaxies for two different cases: (i) GRBs simply trace star formation, and (ii) GRBs preferentially occur in low-metallicity environment. We compare the simulation with observations, and discuss the low-metallicity preference of GRBs. We reproduce the observed luminosity probability distribution function of ORB host galaxies when we assume that GRBs originate from stars with metallicities Z less than or similar to 0.1Z(circle dot), supporting the suggestion from the theoretical studies.