**2. Why Multi-Messenger Astrophysics? What We Want to Learn from Compact Binary Coalescence**

Binary systems including two compact objects (black holes, BH, and neutron stars, NS) are unique laboratories for studying physics that is inaccessible in terrestrial laboratories; in particular strong-field general relativity, relativistic acceleration mechanisms, and matter under extreme conditions (density, temperature, magnetic field). It was postulated for many years that the NS-NS and BH-NS coalescence leads to the production of a short GRB, a bright flash of gamma-rays lasting less than a few seconds [3] (see also the reviews) [4,5]. The reason for this can be outlines as follows. The outcome of the coalescence is a BH (or a metastable NS), including most of the mass of the binary system, and an accretion disk including the leftovers (on the order of a few percent of the solar mass). Accretion of this matter onto the newly formed compact object can release 1052–10<sup>53</sup> ergs of gravitational energy, so even if a small fraction of this large amount can be converted to electromagnetic radiation a GRB can be generated. The duration of the burst is determined by the lifetime of the disk, which is expected to be a fraction of a second at these masses. Accreting compact objects can produce powerful jets, which can transport the energy from the launching site near the inner engine to the photosphere, usually at 107–10<sup>8</sup> Schwarzschild radii *R*S. The jet launching site is thought to be near the newly formed BH, a potential laboratory for GR in the strong-field and the dynamics of matter under extreme conditions, or even new physics. In other words, are the BH and NS we observe in GW the same as described by GR? GWs are produced during the binary coalescence and settling of their remnants; therefore, they should precede the launching of the jet (the short GRB), while providing precise information on the physical condition of the system just before engine ignition. The physics at play outside the photosphere includes particle acceleration in internal shocks to produce the GRB prompt emission and external shocks to produce the GRB afterglow; both can be probed by electromagnetic observations. Detailed GRMHD simulations express this in a more quantitative framework ([6] and references therein). However, only observations can quantitatively confirm this scenario, connecting the physics near the event horizon of the newly formed BH to the physics of jet formation, collimation, and propagation. For example, the variability of the observed gamma-ray light curve may reflect the energy injection at the base of the jet [7,8]. The following key questions can be addressed by the multi-messenger approach to CBC:

• **What happens during the merger of compact objects?** How frequent is the coincidence with short GRBs; how frequent is the formation of powerful relativistic jets? Non-detections are, in principle, as important as detections. This question is addressed through simultaneous observations and studies of the GW event and the high-energy emission associated with jet production.


The first three questions require the detection of GW signals from a statistical sample of NS-NS and NS-BH systems and their sensitive high-energy coverage over the full sky. The fourth and fifth questions require the accurate determination of the position of the source of GWs and multiwavelength follow-up observations, from radio waves to gamma-rays, of the electromagnetic counterpart.
