Some Observational Consequences of GRB Shock Models

In the internal shock scenario for GRBs we expect some fraction of the energy of the burst to be carried by slow moving shells that were ejected at late times. These slow shells collide with faster moving outer shells when the outer shells have slowed down as a result of sweeping up material from the ISM. This gives rise to a forward shock that moves into the outer shell producing a bump in the afterglow light curve of amplitude roughly proportional to the ratio of the energy in the inner and the outer shells. In addition, a reverse shock propagates in the inner shell and produces emission at a characteristic frequency that is typically much smaller than the peak of the emission from the outer shell by a factor of $\sim 25 \gamma_{0c}^2 (E_2/E_1)^{1.1}$, and the observed flux at this frequency from the reverse shock is larger compared to the flux from the outer shell by a factor of $\sim 8 (\gamma_{0c} E_2/E_1)^{5/3} $; where $\gamma_{0c}$ is the bulk Lorentz factor of the outer shell at the time of collision, and $E_1 & E_2$ are the total energy in the outer and the inner shells respectively. The Lorentz factor is related to the observer time as $\sim 5 (t/day)^{3/8}$. The shell collision could produce initial temporal variability in the early afterglow signal. The lack of significant deviation from a power-law decline of the optical afterglow from half a dozen bursts suggests that $E_2/E_1$ is small. Future multi-wavelength observations should be able to either detect bumps in the light curve corresponding to both the forward and the reverse shocks or further constrain the late time release of energy in ejecta with small Lorentz factor, which is expected generically in the internal shock models for the gamma-ray bursts.