Using Swift observations of prompt and afterglow emission to classify GRBs

Abstract

We present an analysis of early BAT and XRT data for 107 gamma--ray bursts (GRBs) observed by the Swift satellite. We use these data to examine the behaviour of the X-ray light curve and propose a classification scheme for GRBs based on this behaviour. As found for previous smaller samples, the earliest X-ray light curve can be well described by an exponential which relaxes into a power law, often with flares superimposed. The later emission is well fit using a similar functional form and we find that these two functions provide a good description of the entire X-ray light curve. For the prompt emission, the transition time between the exponential and the power law gives a well-defined timescale, T_p, for the burst duration. We use T_p, the spectral index of the prompt emission, beta_p, and the prompt power law decay index, alpha_p to define four classes of burst: short, slow, fast and soft. Bursts with slowly declining emission have spectral and temporal properties similar to the short bursts despite having longer durations. Some of these GRBs may therefore arise from similar progenitors including several types of binary system. Short bursts tend to decline more gradually than longer duration bursts and hence emit a significant fraction of their total energy at times greater than T_p. This may be due to differences in the environment or the progenitor for long, fast bursts.