Selection effects on GRB spectral-energy correlations

Abstract

Instrumental selection effects can act upon the estimates of the peak energy Ep, the fluence F and the peak flux P of GRBs. If this were the case, then the correlations involving the corresponding rest frame quantities (i.e. Ep, Eiso and the peak luminosity Liso) would be questioned. We estimated, as a function of Ep, the minimum peak flux necessary to trigger a GRB and the minimum fluence a burst must have to determine the value of Ep by considering different instruments (BATSE, Swift, BeppoSAX). We find that the latter dominates over the former. We then study the Ep-fluence (and flux) correlation in the observer plane. GRBs with redshift show well defined Ep-F and Ep-P correlations: in this planes the selection effects are present, but do not determine the found orrelations. This is not true for Swift GRBs with redshift, for which the spectral analysis threshold does affect their distribution in the observer planes. Extending the sample to GRBs without z, we still find a significant Ep-F correlation, although with a larger scatter than that defined by GRBs with redshift. We find that 6% are outliers of the Amati correlation. The Ep-P correlation of GRBs with or without redshift is the same and no outlier is found among bursts without redshift.