Statistical Analysis of Long GRBs’ Prompt Emission and X-Ray Flares: Multivariate Clustering and Correlations

Abstract

The extensive observations done by the X-ray Telescope on board Neil Gehrels Swift Observatory have revealed the presence of late-time flares concurrent with the decaying afterglow emission. However, the origins of these flares are elusive. In this work, we make use of the large database of Swift observations (2005–2020) of long gamma-ray bursts (GRBs) to conduct a systematic statistical study between the prompt gamma-ray emission and X-ray flares by characterizing their temporal and spectral properties of duration, quiescent period, peak flux, fluence, minimum variability timescale, and spectral power-law index. The multidimensional database of parameters thereby generated was investigated by principal component analysis, which revealed there is no evident correlation between the different parameters of the prompt emission and X-ray flares. Furthermore, the correlation studies revealed that while there is a trend of positive correlation between the minimum variability timescale of the flare and its duration, and of strong negative correlation with its peak flux, there are no such correlations observed in the prompt emission. Similarly, we find a positive correlation between the quiescent period and the flare duration, and a negative correlation with the flare peak flux, while no such correlations are observed for the prompt emission of the GRBs. Finally, among the X-ray flares, we find two dominant classes, whose variations are driven by the minimum variability timescale, peak flux, and fluence of the flares. A catalog of these different parameters characterizing the prompt and flare emissions is presented.