The central engine of GRBs is widely believed to be a nascent stellar black hole surrounded by a hyper-accreting disk. With the current observational data it is, however, not easy to pin down the physical parameters, for example, the mass and the spin of a central black hole, the accretion rate (or alternatively the accretion-disk mass), and the initial radius of the outflow where it is launched (R0). The main reason is that the central engine hides deeply behind the electromagnetic-radiation surface and it is hard to break the degeneracies between the parameters with very limited observational constraints. For some long bursts, a reliable estimate of R0 with the identified thermal spectrum component in the prompt spectrum is found. However, for short events the lack of a reliable identification of such a component renders a reasonable estimate difficult.

A recent work of two astronomers of PMO(ApJ 739:47 2011 http://iopscience.iop.org/0004-637X/739/1/47/) sheds some light on this problem. In the specific scenario of a binary–neutronstar merger, Dr. FAN Yizhong and Dr. WEI Daming try to parameterize the central engine of short GRBs and successfully estimate the disk mass for a sample of 10 short GRBs. Their semi-analytical estimation of the disk mass of ~0.01-0.1 M_Sun for about half of the GRBs in their sample is in agreement with that found in the numerical simulation of the merger of binary–neutron-star. The authors also outline how to constrain R0 without the identification of an ideal thermal spectrum component. Applying the method to GRB 090510, the R0 estimated by the authors is again consistent with the compact-object merger model.

If some short GRBs are really of compact-object merger origin, they should be good candidates for the future detection of gravitational wave. The binary-neutron-star merger model and the estimations in the work introduced here will be directly tested with the future short-burst-associated gravitational wave data.