Numerical Relativity for Gravitational-Wave detection

The first direct detection of gravitational waves (GW) is expected in the next few years, from a network of ground- based detectors currently taking data. One of the most promising sources for the first detection is the merger of two black holes (BHs). Identifying the signature of a merger waveform in noisy detector data requires matched filtering against a template bank of theoretical waveforms, and the only way to calculate the waves predicted by Einstein`s full general theory of relativity is to solve the equations on a computer. That is will be done is this project. Numerical simulations of BH binaries became possible in 2005, and since then a wealth of results have been obtained relevant to mathematical relativity, BH physics, and galactic astrophysics. But only a handful of simulations have been performed for the purpose of GW detection. I have been involved in many of them, and for this project I plan, with the aid of a PhD student and postdoc, to extend this study to produce a complete-enough set of numerical waveforms to allow a full mapping of all possible con?gurations of BH binaries following quasi- circular inspiral. The main goal of this work is to construct a phenomenological analytic model that dramatically increases the probability of detection (over current waveform approximation methods), and to make possible accurate estimation of the physical parameters of the GW source. I have already completed important work that lays the foundations for this task, which has continued since I started a Lise-Meitner Fellowship at the University of Vienna in September of this year. With the aid of a PhD student and postdoc, I will be better able to fully realize the goal of a complete analytic waveform family that is ready to use in time for the commissioning of the second generation of detectors in the next few years.

We simulate one of the most violent processes in the universe since the Big Bang: collisions of black holes (BHs). The ultimate goal is the direct observation of black-hole coalescences through the gravitational-wave signals that they emit. Gravitational waves (GW) are ripples in space and time, predicted by Einstein`s general theory of relativity. They are generated by accelerating masses and are far weaker than electromagnetic waves that have told us everything we have learnt about the cosmos since ancient times. Detection of gravitational waves would push open a new window on the universe, and tell us about its "dark side". The first direct detection of gravitational waves is expected in the next few years, from a network of ground-based detectors, including the US Laser Interferometer Gravitational-Wave Observatory (LIGO) and the European GEO600 and Virgo detectors. Identifying the signature of a merger waveform in noisy detector data requires comparison against a bank of theoretical template waveforms, and the only way to calculate the waves predicted by Einstein`s full general theory of relativity is to solve the equations on supercomputers. In this project we have computed GW signals from BH binary systems with different masses and spins that will allow us to produce a better phenomenological analytic model of the GW waveform over the parameter space. This model will dramatically increase the probability of detection (over current waveform approximation methods), and make possible accurate estimation of the physical parameters of the GW source. BH binaries that are not perturbed by other massive objects start with a relatively eccentric orbit from their birth through supernova explosions. Over billions of years their orbit slowly becomes more and more circular and they come closer to each other under the emission of GWs. At the very close separations (10s of Schwarzschild radii) where we start our numerical simulations the orbit is almost like a circle, but they still inspiral and emit GWs more and more strongly until they merge into a single BH. We have created a new method to calculate what the initial velocities of the BHs must be for them to inspiral in an almost circular way and have estimated how much error in these velocities is tolerable without spoiling the GW signal with unwanted oscillations from eccentricity.