Bright X-ray flares are routinely detected by the Swift satellite during the early afterglow of gamma-ray bursts, when the explosion ejecta drives a blast wave into the external medium. We suggest that the flares are produced as the reverse shock pro
pagates into the tail of the ejecta. The ejecta is expected to contain a few dense shells formed at an earlier stage of the explosion. We show an example of how such dense shells form and describe how the reverse shock interacts with them. A new reflected shock is generated in this interaction, which produces a short-lived X-ray flare. The model provides a natural explanation for the main observed features of the X-ray flares --- the fast rise, the steep power-law decline, and the characteristic peak duration Delta t /t= (0.1-0.3).
We test a model recently proposed for the persistent hard X-ray emission from magnetars. In the model, hard X-rays are produced by a decelerating electron-positron flow in the closed magnetosphere. The flow decelerates as it radiates its energy away
via resonant scattering of soft X-rays, then it reaches the top of the magnetic loop and annihilates there. We test the model against observations of three magnetars: 4U 0142+61, 1RXS J1708-4009, and 1E 1841-045. We find that the model successfully fits the observed phase-resolved spectra. We derive constraints on the angle between the rotational and magnetic axes of the neutron star, the object inclination to the line of sight, and the size of the active twisted region filled with the plasma flow. Using the fit of the hard X-ray component of the magnetar spectrum, we revisit the remaining soft X-ray component. We find that it can be explained by a modified two-temperature blackbody model. The hotter blackbody is consistent with a hot spot covering 1-10% of the neutron star surface. Such a hot spot is expected at the base of the magnetospheric e+- outflow, as some particles created in the e+- discharge flow back and bombard the stellar surface.
The early X-ray afterglow of gamma-ray bursts revealed by Swift carried many surprises. We focus in this paper on the plateau phase whose origin remains highly debated. We confront several newly discovered correlations between prompt and afterglow qu
antities (isotropic emitted energy in gamma-rays, luminosity and duration of the plateau) to several models proposed for the origin of plateaus in order to check if they can account for these observed correlations. We first show that the scenario of plateau formation by energy injection into the forward shock leads to an efficiency crisis for the prompt phase and therefore study two possible alternatives: the first one still takes place within the framework of the standard forward shock model but allows for a variation of the microphysics parameters to reduce the radiative efficiency at early times; in the second scenario the early afterglow results from a long-lived reverse shock. Its shape then depends on the distribution of energy as a function of Lorentz factor in the ejecta. In both cases, we first present simple analytical estimates of the plateau luminosity and duration and then compute detailed light curves. In the two considered scenarios we find that plateaus following the observed correlations can be obtained under the condition that specific additional ingredients are included. In the forward shock scenario, the preferred model supposes a wind external medium and a microphysics parameter epsilon_e that first varies as n^{- u} (n being the external density), with u~1 to get a flat plateau, before staying constant below a critical density n_0. To produce a plateau in the reverse shock scenario the ejecta must contain a tail of low Lorentz factor with a peak of energy deposition at Gamma >~ 10.
We study the observed correlations between the duration and luminosity of the early afterglow plateau and the isotropic gamma-ray energy release during the prompt phase. We discuss these correlations in the context of two scenarios for the origin of
the plateaus. In the first one the afterglow is made by the forward shock and the plateau results from variations of the microphysics parameters while in the second one the early afterglow is made by a long-lived reverse shock propagating in a low Lorentz factor tail of the ejecta.
The peak time of optical afterglow may be used as a proxy to constrain the Lorentz factor Gamma of the gamma-ray burst (GRB) ejecta. We revisit this method by including bursts with optical observations that started when the afterglow flux was already
decaying; these bursts can provide useful lower limits on Gamma. Combining all analyzed bursts in our sample, we find that the previously reported correlation between Gamma and the burst luminosity L_gamma does not hold. However, the data clearly shows a lower bound Gamma_min which increases with L_gamma. We suggest an explanation for this feature: explosions with large jet luminosities and Gamma < Gamma_min suffer strong adiabatic cooling before their radiation is released at the photosphere; they produce weak bursts, barely detectable with present instruments. To test this explanation we examine the effect of adiabatic cooling on the GRB location in the L_gamma - Gamma plane using a Monte Carlo simulation of the GRB population. Our results predict detectable on-axis orphan afterglows. We also derive upper limits on the density of the ambient medium that decelerates the explosion ejecta. We find that the density in many cases is smaller than expected for stellar winds from normal Wolf-Rayet progenitors. The burst progenitors may be peculiar massive stars with weaker winds or there might exist a mechanism that reduces the stellar wind a few years before the explosion.
GRB spectra appear non-thermal, but recent observations of a few bursts with Fermi GBM have confirmed previous indications from BATSE of the presence of an underlying thermal component. Photospheric emission is indeed expected when the relativistic o
utflow emerging from the central engine becomes transparent to its own radiation, with a quasi-blackbody spectrum in absence of additional sub-photospheric dissipation. However, its intensity strongly depends on the acceleration mechanism - thermal or magnetic - of the flow. We aim to compute the thermal and non-thermal emissions produced by an outflow with a variable Lorentz factor, where the power injected at the origin is partially thermal (fraction epsilon_th) and partially magnetic (fraction 1-epsilon_th). The thermal emission is produced at the photosphere, and the non-thermal emission in the optically thin regime. Apart from the value of epsilon_th, we want to test how the other model parameters affect the observed ratio of the thermal to non-thermal emission. If the non-thermal emission is made by internal shocks, we self-consistently obtained the light curves and spectra of the thermal and non-thermal components for any distribution of the Lorentz factor in the flow. If the non-thermal emission results from magnetic reconnection we were unable to produce a light curve and could only compare the respective non-thermal and thermal spectra. In the different considered cases, we varied the model parameters to see when the thermal component in the light curve and/or spectrum is likely to show up or, on the contrary, to be hidden. We finally compared our results to the proposed evidence for the presence of a thermal component in GRB spectra. Focussing on GRB 090902B and GRB 10072B, we showed how these observations can be used to constrain the nature and acceleration mechanism of GRB outflows.
The Swift-XRT observations of the early X-ray afterglow of GRBs show that it usually begins with a steep decay phase. A possible origin of this early steep decay is the high latitude emission that subsists when the on-axis emission of the last dissip
ating regions in the relativistic outflow has been switched-off. We wish to establish which of various models of the prompt emission are compatible with this interpretation. We successively consider internal shocks, photospheric emission, and magnetic reconnection and obtain the typical decay timescale at the end of the prompt phase in each case. Only internal shocks naturally predict a decay timescale comparable to the burst duration, as required to explain XRT observations in terms of high latitude emission. The decay timescale of the high latitude emission is much too short in photospheric models and the observed decay must then correspond to an effective and generic behavior of the central engine. Reconnection models require some ad hoc assumptions to agree with the data, which will have to be validated when a better description of the reconnection process becomes available.
GRB 080503, detected by Swift, belongs to the class of bursts whose prompt phase consists of an initial short spike followed by a longer soft tail. It did not show any transition to a regular afterglow at the end of the prompt emission but exhibited
a surprising rebrightening after one day. We aim to explain this rebrightening with two different scenarios - refreshed shocks or a density clump in the circumburst medium - and two models for the origin of the afterglow, the standard one where it comes from the forward shock, and an alternative one where it results from a long-lived reverse shock. We computed afterglow light curves either using a single-zone approximation for the shocked region or a detailed multizone method that more accurately accounts for the compression of the material. We find that in several of the considered cases the detailed model must be used to obtain a reliable description of the shock dynamics. The density clump scenario is not favored. We confirm previous results that the presence of the clump has little effect on the forward shock emission, except if the microphysics parameters evolve when the shock enters the clump. Moreover, we find that the rebrightening from the reverse shock is also too weak when it is calculated with the multi-zone method. On the other hand, in the refreshed-shock scenario both the forward and reverse shock models provide satisfactory fits of the data under some additional conditions on the distribution of the Lorentz factor in the ejecta and the beaming angle of the relativistic outflow.
Recent detections of GeV photons in a few GRBs by Fermi-LAT have led to strong constraints on the bulk Lorentz factor in GRB outflows. To avoid a large gamma-gamma optical depth, minimum values of the Lorentz factor have been estimated to be as high
as 800-1200 in some bursts. Here we present a detailed calculation of the gamma-gamma optical depth taking into account both the geometry and the dynamics of the jet. In the framework of the internal shock model, we compute lightcurves in different energy bands and the corresponding spectrum and we show how the limits on the Lorentz factor can be significantly lowered compared to previous estimates. Our detailed model of the propagation of high energy photons in GRB outflows is also appropriate to study many other consequences of gamma-gamma annihilation in GRBs: (i) the gamma-gamma cutoff transition in a time-integrated spectrum is expected to be closer to a power-law steepening of the spectrum than to a sharp exponential decay; (ii) the temporal evolution of the gamma-gamma opacity during a burst favors a delay between the MeV and GeV light curves; (iii) for complex GRBs, the gamma-gamma opacity suppresses the shortest time-scale features in high energy light curves (above 100 MeV). Finally we also consider GRB scenarii where MeV and GeV photons are not produced at the same location, showing that the gamma-gamma opacity could be further lowered, reducing even more the constraint on the minimum Lorentz factor.
Recent detections of GeV photons in a few GRBs by Fermi-LAT imply huge bulk Lorentz factors to avoid a large gamma gamma optical depth at high energy. Estimates can be as high as Gamma ~ 1000 in the most extreme cases. This puts severe constraints on
models of the central engine and the jet acceleration in GRBs. These estimates are however obtained from a simplified single zone model. We present here a more realistic calculation which takes into account the time, space and direction dependent photon field existing in an outflow with several relativistically moving emitting zones. The formalism is general and can be applied to many models of the prompt GRB emission. We present results obtained for a numerical implementation in the framework of the internal shock model. We show that (i) the minimum Lorentz factor Gamma_min in bright LAT GRBs is reduced by a factor ~ 2-3 compared to previous estimates if the GeV and MeV emission are produced in the same region, and by an additional factor ~ 2-8 if the GeV emission is produced at larger radii. We provide an improved approximate formula for Gamma_min which is in good agreement with our numerical results and can be directly applied to LAT GRB data; (ii) a delayed GeV onset can be due to the time evolution of the opacity. As an illustration of these two first results, we present a synthetic GRB that reproduces most features of GRB 080916C with a mean Lorentz factor of ~ 340, an optically thin regime for gamma gamma opacity at 3GeV in bin b, a variability timescale of 0.5s in the MeV lightcurve and a delayed onset of ~ 5s of the GeV emission; (iii) the gamma gamma opacity can smooth the short timescale variability in the GeV lightcurve. This last result implies that the observed variability at high energy is not necessarily a good test to distinguish between an internal and an external origin for the GeV emission in GRBs. [abridged]
