Characteristics of Two-episode Emission Patterns in {\em Fermi} Long Gamma-Ray Bursts

TL;DR

This study systematically analyzes 101 Fermi long gamma-ray bursts, revealing two-episode emission patterns with distinct spectral and temporal properties, and suggests a common physical origin for these emission components.

Contribution

It provides the first comprehensive spectral and temporal analysis of two-episode emission patterns in Fermi long GRBs, highlighting their similarities and potential shared origin.

Findings

Two-episode emission components are common in long GRBs.

No significant correlation between the durations of the two episodes.

Spectral properties of the two episodes are similar, indicating a shared physical origin.

Abstract

Two-episode emission components separated by quiescent gaps in the prompt emission of gamma-ray bursts (GRBs) have been observed in the {\em Swift} era, but there is a lack of spectral information due to the narrow energy band of the {\em Swift}/Burst Alert Telescope. In this paper, a systematic analysis of the spectral and temporal properties of the prompt emission of 101 {\em Fermi}/Gamma-ray Burst Monitor detected long GRBs show the existence of two-episode emission components in the light curves, with quiescent times of up to hundreds of seconds. We focus on investigating the differences of those two emission episodes. We find that the light curves of the two emission components exhibit different behavior, e.g., a soft emission component that either precedes or follows the main prompt emission or that the intensity of the two emission episodes are comparable with each other. No statistically significant correlation in the duration of the two emission episodes can be claimed. We define a new parameter $\varepsilon$ as the ratio of the peak flux of the first and second emission episodes and find that a higher $\varepsilon$ corresponds to a larger fluence. The preferred spectral model in our analysis is a cutoff power-law model for most GRBs. The distribution of $E_p$ for episodes I and II range from tens of keV to 1000 keV with a lognormal fit and there are no significant differences between them. Moreover, we do not find significant relationships between $\varepsilon$ and $E_p$ for the two emission episodes. Those results suggest that these two-episode emission components likely share the same physical origin.