2 GRBs studies with AGILE

GRBs can emit a substantial fraction of their energy above 100 MeV both during their prompt emission phase and during the afterglow. About 10 GRBs have been detected by EGRET's spark chamber during $\sim 7$years of operations ([Dingus 1995]; [Schneid et al. 1996]). This number appears to be limited by the EGRET FOV and sensitivity and not by the GRB emission mechanism. The GRB detection rate by AGILE is expected to be a factor of $\lower 0.6ex\hbox{$\stackrel{\gt}{\sim}{}$}5$ larger than that of EGRET, i.e., $\ge 5 - 10$ events/year. We note that AGILE's flux sensitivity for GRB detection is comparable to that of EGRET that typically detected GRBs off-axis. Figure 2 shows the effective areas of AGILE and EGRET.

A short deadtime for gamma-ray detection ($\sim 1$ ms, i.e., a factor of $\sim 100$ times smaller than that of EGRET's spark chamber) is obtained by AGILE's fast electronic readout and a cyclic buffer. GRB pulses during the typically short risetimes ($\lower 0.6ex\hbox{$\stackrel{<}{\sim}{}$}0.5$ s) can be efficiently studied for the first time. Furthermore, AGILE might discover an hypothetical and theoretically expected class of events, i.e., high-energy very short GRBs ($\sim 0.1-1$ s) that could not be efficiently detected above 100 MeV with previous instrumentation.

Long events that we could call "gamma-ray afterglows'' (lasting $\sim 100-1000$ s) can also be detected. The remarkable discovery by EGRET of gamma-ray emission up to $\sim 20$ GeV for GRB 940217 ([Hurley et al. 1994]) lasting $\sim 5,400$ s shows that gamma-rays can be emitted during a timescale much longer than that observed in the hard X-ray range. Obviously, this fact suggests the existence of a quasi-continuous acceleration process shifting the peak of the spectral power per energy decade at photon energies substantially larger than those typical of the prompt pulse emission. This late-time particle acceleration challenges theoretical models for the early GRB afterglow emission. AGILE can contribute in obtaining more information on these enigmatic events. Currently, 2 out of 10 events detected by EGRET, GRB 930131 ([Sommer et al. 1994]) and GRB 940217 ([Hurley et al. 1994]) clearly show durations (in the gamma-ray band above $\sim 30$ MeV) longer by a factor of $\sim 10$ or more than the durations established in the 50-300 keV by BATSE ([Paciesas et al. 1998]).

AGILE is also expected to be quite efficient in detecting photons above 1 GeV because of limited backscattering from a mini-calorimeter of small radiation length. For relatively bright events, GRB error boxes derived by gamma-ray data alone are expected to have radii smaller by a factor of $\sim 2$ than those of EGRET.

GRB broad-band spectral data from $\sim 1$ MeV up to $\sim 10$ GeV can be obtained by the baseline instrument by combining information of the CsI mini-calorimeter and silicon tracker. Additional information at lower energies can be obtained by the response of silicon tracker planes to the passage of hard X-rays differentially absorbed or Compton scattered throughout the tracker.

The Super-AGILE ultra-light coded mask system is planned to provide additional spectral information in the $\sim 10-40$ keV range. A relatively unambiguous "trigger'' for a fast search in tracker data of GRB-related events can be obtained by ratemeter information from the additional silicon plane with readout electronics optimized in that energy range.

Super-AGILE will be able to locate GRBs within a few arcminutes and will systematically study the interplay between hard X-ray and gamma-ray emissions. A rapid alert and communication of quicklook analysis results is planned in case of GRB detection by Super-AGILE.