Previous Session | Back to program of the week | Next Session |
Binary neutron star mergers are investigated using nonlinear 3+1 numerical relativity simulations and the analytical effective-one-body (EOB) model. The EOB model predicts quasi-universal relations between the mass-rescaled gravitational wave frequency and the binding energy at the moment of merger, and certain dimensionless binary tidal coupling constants depending on the stars Love numbers, compactnesses and the binary mass ratio. These relations are quasi-universal in the sense that, for a given value of the tidal coupling constant, they depend significantly neither on the equation of state nor on the mass ratio, though they do depend on stars spins. The spin dependence is approximately linear for small spins aligned with the orbital angular momentum. The quasi-universality is a property of the conservative dynamics, and emerges as the binary interaction becomes tidally dominated. This analytical prediction is qualitatively consistent with new, multi-orbit numerical relativity results for the relevant case of equal-mass irrotational binaries. Universal relations are thus expected to characterize neutron star mergers dynamics. In the context of gravitational wave astronomy, these universal relations may be used to constrain the neutron star equation of state using waveforms that model the merger accurately.
Extreme-mass-ratio inspirals (EMRIs) are one of the main sources of gravitational
radiation for space-based detectors like the eLISA concept proposed as the
strawman mission in “The Gravitational Universe”, the science theme selected
by the European Space Agency for its future L3 mission. The detection of EMRIs
rely on an accurate modelling of the gravitational waveforms produced. The most
accurate method proposed is the self-force approach, where the inspiral is described
by the action of a local force determined by the gravitational perturbations generated
by the small compact object in the background spacetime of the supermassive black hole.
In this talk I will present new developments that allow a simpler computation of the
self-force for the case of non-spinning black holes. These developments, based on the
particle-without-particle (PwP) technique, allow the computation of the self-force in
gauges that simplify this computation. We will also discuss the prospect of applying
this technique for spinning black holes.
Indirect (source-free) integration method for EMRIs: waveforms from geodesic generic orbits and self-force consistent radial fall
Self-consistent orbital evolution
After i) a brief introduction to the self-force, ii) our method of numerical integration for any orbit, we show our implementation of a self-consistent orbital evolution in Regge-Wheeler gauge for a plunge of an EMRI.
- S. Aoudia, A. Spallicci, 2011, Phys. Rev. D, 83, 064029.
- L. Blanchet, A. Spallicci, B. Whiting (eds.), 2011. Mass and motion in general relativity, Springer Series on Fundamental Theory of Physics.
- P. Ritter, A. Spallicci, S. Aoudia, S. Cordier, 2011, Class. Q. Grav., 28, 134012.
Compact binaries in hyperbolic orbits are plausible gravitational waves (GW) sources for both the ground and space based GW detectors. We develop an efficient prescription to compute post-Newtonian (PN) accurate ready-to-use GW polarization states for spinning compact binaries in hyperbolic orbits. This is achieved by invoking the 1.5 PN-accurate quasi-Keplerian parameterization to describe the binary dynamics that incorporates the dominant order spin-orbit interactions. We probe the effects of spins and gravitational radiation reaction during the hyperbolic passage. It turns out that both polarization states exhibit the memory effect for GWs from spinning compact binaries in hyperbolic orbits. In contrast, only cross polarization state exhibits the memory effects for GWs from non-spinning compact binaries. We compare amplitude corrected GW polarization states for non-spinning compact binaries in hyperbolic orbits arising from our approach with those obtained by invoking the generalized true anomaly parameterization.
A Hamiltonian Monte Carlo for Massive Black Hole Binaries
Markov chain algorithms have become a popular method for conducting Bayesian Inference of gravitational waves. The commonly used Hessian MCMC, while effective, can suffer from slow convergence due to the random walk nature of the algorithm. Here we present a Hamiltonian Monte Carlo algorithm, which improves the convergence by approximately the dimensionality of the problem. By equating the waveform parameters to state space variables, and introducing a set of canonical momenta, we eliminate the random walk nature by solving Hamilton's equations. The main breakthrough of the algorithm is, by analytically fitting the log-likelihood gradients, we eliminate the computational bottleneck that has previously plagued the algorithm, and actually runs faster than the equivalent Hessian MCMC.
Cédric Huwyler, Edward Porter, Philippe Jetzer
[preliminary abstract]
Motivated by the parameterized post-Einsteinian (ppE) scheme devised by Yunes and Pretorius, which introduces corrections to the post-Newtonian coefficients of the frequency domain gravitational waveform in order to emulate alternative theories of gravity, we compute analytical time domain waveforms that, after a numerical Fourier transform, aim to represent (phase correction only) ppE waveforms. We explore differences in the parameter estimation capabilities of such corrected time domain waveforms to the ppE scheme by employing a Markov Chain Monte Carlo (MCMC) code to assess the ability of the current eLISA configuration (as presented for the L3 mission) and a few other possible LISA-like configurations to detect departures from General Relativity.
In this talk we will see that by implementing a scheme that combines a reduced basis, empirical interpolation and reduced order quadrature , Bayesian inference on the $9$-dimensional parameter space of non-spinning binary neutron star inspirals can be sped up by a factor of $30$ for the early advanced detectors - which are expected to have a low-frequency sensitivity limit of ~40Hz. The speed-up will increase to 150 as the detectors improve this low-frequency limit to ~10Hz. Reduced order modelling techniques can provide one to two orders of magnitude acceleration of the data analysis infrastructure of a large experiment with no loss of accuracy. These techniques are broadly applicable to any scientific experiment where fast Bayesian analysis is needed.
Searches for the Stochastic Gravitational-Wave Background with Advanced LIGO and Advanced Virgo Detectors
With the possible detection of a gravitational-wave imprint on the cosmic microwave background by the BICEP2 experiment, we enter an exciting phase in the study of gravitational waves in which it appears possible to study cosmological models using the stochastic gravitational-wave background. Additionally, astrophysical sources (such as unresolved binary coalescences) are predicted to contribute to the stochastic background. In this talk, I give an overview of the science potential of searches for the stochastic background by advanced detectors such as LIGO and Virgo.