## 7 Summary

Update
It is hoped that as gravitational collapse simulations become more sophisticated, the historically
widely varying estimates of the magnitude of GW emission from collapse may start to converge.
Steady progress in this field has been made in the last decade. Some researchers have begun to
use progenitor models produced with stellar evolution codes, which thus have more realistic
angular momentum profiles, as starting points for collapse simulations [86]. This reduces the
need for collapse studies that include large surveys of the angular momentum parameter space.
Other progress made in the numerical study of collapse includes the use of realistic equations of
state [86], advanced neutrino transport and interaction schemes [130, 150, 197, 240], and the
performance of 3D Newtonian [198, 35, 91, 92] and improved methods for general relativistic
simulations [222, 61].
There is still much work to be done toward the goal of self-consistent, 3D general relativistic
collapse simulations. The most rapidly rotating progenitors (with possibly the strongest GW
signals) may be produced only in specific binary systems [95, 84, 265]. If the GW signal is
dominated by convection, then it is critical that scientists actually understand the true supernova
mechanism. Accurate progenitor modelling and collapse simulations must include the effects of
magnetic fields, as they can significantly alter the amount of angular momentum and differential
rotation present in collapsing stars. Many of the more advanced studies, which include proper
microphysics treatment and/or general relativistic effects, have been limited to axisymmetry.
Full 3D simulations are necessary to compute the characteristics of the GW emission from
non-axisymmetric collapse phenomena. Furthermore, simulations that follow both the collapse
and the evolution of the collapsed remnant are necessary to consistently predict GW emission.
One benefit of long duration simulations is that they will facilitate the investigation of the
effects of the envelope on any instabilities that develop in the collapsing core or remnant. Of
course, lengthy 3D simulations are computationally intensive. This burden may be reduced by
the use of advanced numerical techniques, including adaptive mesh refinement and parallel
algorithms.

The current numerical simulations of gravitational collapse indicate that interferometric observatories
could detect GWs emitted by some collapse phenomena. LIGO-I may be able to detect GWs from secular
bar-mode instabilities in core collapse SNe [146] and magnetized tori surrounding black hole collapse
remnants [251]. LIGO-II could observe GWs from dynamical bar-mode instabilities in AIC [155] and core
collapse SNe [86]. LISA should be able to detect the collapse (and any bar-mode instabilities that develop
during the collapse) of SMSs [16] and the ringdown of black hole remnants of collapsed Population III
stars [86] and SMSs [208]. These observations will provide unique information about gravitational collapse,
the supernova and gamma-ray burst explosion mechanisms, and their associated progenitors and
remnants.