2002A&A...385..377Q

A comprehensive temporal analysis has been performed on the 319 brightest GRBs with T₉₀>2s from the BATSE current catalog. The GRBs were denoised using wavelets and subjected to an automatic pulse selection algorithm as an objective way of identifying pulses and quantifying the effects of neighbouring pulses. The number of statistically significant pulses selected from the sample was greater than 3000. The rise times, fall times, full-widths at half-maximum (FWHM), pulse amplitudes and pulse areas were measured and the frequency distributions are presented here. All are consistent with lognormal distributions provided the pulses are well separated. The distribution of time intervals between pulses is not random but compatible with a lognormal distribution when allowance was made for the 64 ms time resolution and a small excess (5%) of long duration intervals that is often referred to as a Pareto-Levy tail. The time intervals between pulses are most important because they may be an almost direct measure of the activity in the central engine. Lognormal distributions of time intervals also occur in pulsars and SGR sources and therefore provide indirect evidence that the time intervals between pulses in GRBs are also generated by rotation powered systems with super-strong magnetic fields. A range of correlations are presented on pulse and burst properties. The rise and fall times, FWHM and area of the pulses are highly correlated with each other. The pulse amplitudes are anticorrelated with the FWHM. The time intervals between pulses and pulse amplitudes of neighbouring pulses are correlated with each other. It was also found that the number of pulses, N, in GRBs is strongly correlated with the fluence and duration and that can explain the well known correlation between duration and fluence. The GRBs were sorted into three categories based on N i.e. 3≤N≤12, 13≤N≤24 and N≥25. The properties of pulses before and after the strongest pulse were compared for three categories of bursts. No major differences were found between the distributions of the pulse properties before and after the strongest pulse in the GRB. However there is a strong trend for pulses to have slower rise times and faster fall times in the first half of the burst and this pattern is strongest for category 3≤N≤12. This analysis revealed that the GRBs with large numbers of pulses have narrower and faster pulses and also larger fluences, longer durations and higher hardness ratios than the GRBs with smaller numbers of pulses. These results may be explained by either homogeneous or inhomogeneous jet models of GRBs. The GRBs with larger number of pulses are closer to the axis if Γ varies with the opening angle of the jet and the imprint of the jet is preserved in the pulse structure of the burst. The distribution of the number of pulses per GRB broadly reflects the beaming by the jet.