X‐Ray Hardness Evolution in GRB Afterglows and Flares: Late‐Time GRB Activity withoutN_HVariations

Abstract

We show that the X-ray and γ-ray spectra of Swift GRBs and their afterglows are consistent with the emission characteristic of an expanding, relativistic fireball. The classical afterglow due to the impact of the fireball on the external medium is often not observed until one to several hours after the GRB. Focusing on GRBs 061121, 060614, and 060124, but generalizing to the full (>50 Msec XRT exposure) Swift sample up to and including GRB 061210, we show that the early emission in >90% of early afterglows has a characteristic νF ν spectral energy E peak which likely evolves from the γ-rays through the soft X-ray band on timescales of 10 2 − 10 4 s after the GRB. The observed spectra are strongly curved when plotted with logarithmic axes and have often been incorrectly fitted in other studies with a time-varying soft X-ray absorption. The spectral evolution inferred from fitting instead models used to fit GRBs demonstrates a common evolution-a powerlaw hardness intensity correlation and hard to soft evolution-for GRBs and the early X-ray afterglows and X-ray flares. Combined with studies of short timescale variability, our findings indicate a central engine active for longer than previously suspected. The GRB spectra are observed to become very soft at late times due to an intrinsic spectral evolution and due to the surprising faintness of some afterglows. We discuss models for the early X-ray emission.

The SEDs and host galaxies of the dustiest GRB afterglows

“…In a standard analysis, the introduced curvature caused by the spectral break is then interpreted as a larger dust column . The N H,X measurements are also prone to errors: spectral variation intrinsic to the afterglow can lead to incorrect estimates of the soft X-ray absorption (e.g., Butler & Kocevski 2007). …”

GRB 080517: a local, low-luminosity gamma-ray burst in a dusty galaxy at z = 0.09

“…GRB 080517 appears to have much more in common with these pre-Swift events, with a T90 = 65±27s and a hard photon index of Γ ∼ 1.5. While Ep is difficult to directly constrain with the limited BAT bandpass, the Bayesian method of Butler & Kocevski (2007) suggests that Ep > 55 keV, making GRB 080517 a significant outlier in the EpEiso relation, with a similar location to GRB 980425 and GRB 031203 (see figure 11). Its recovery, some 6 years after the initial detection implies that other, similar, low luminosity events may be present within the Swift catalog, since a significant number of bursts have not been followed in depth due to observational constraints.…”

TheSwift Burst Analyser

“…Butler & Kocevski (2007a) note that absorption derived from early-time XRT data, when strong spectral evolution is present, can be misleading. Therefore, if there are at least 200 X-ray photons detected at T > 4000 s post-trigger, a new spectrum is built using only the data from this time (using the software presented in Evans et al 2009).…”