On the association of gamma-ray bursts with supernovae

The recent discovery of a supernova (SN 1998bw) seemingly associated with GRB 980425 adds a new twist to the decades-old debate over the origin of gamma-ray bursts. To investigate the possibility that some (or all) bursts are associated with supernovae, we performed a systematic search for temporal/angular correlations using catalogs of BATSE and BATSE/Ulysses burst locations. We find no associations with any of the precise BATSE/Ulysses locations, which allows us to conclude that the fraction of high-fluence gamma-ray bursts from known supernovae is small (less than 0.2%). For the more numerous weaker bursts, the corresponding limiting fraction of 1.5% is less constraining due to the imprecise locations of these events. This limit (1.5% similar or equal to 18 bursts) allows that a large fraction of the recent supernovae used as a comparison data set (18 supernovae similar or equal to 20%) could have associated gamma-ray bursts. Thus, although we find no significant evidence to support a burst/supernova association, the possibility cannot be excluded for weak bursts.