The Gravitational wave stochastic background: modeling and detection of non gaussian regimes
Prof. Tania Regimbau (regimbau@oca.eu)

Einstein theory of gravitation predicts the existence of gravitational waves propagating through space-time at the speed of light, like ripples on the surface of water.
These waves can be generated when masses move or interact, and could be observed if the masses are large enough, for instance when two dense stars or black holes coalesce. The superposition of all the sources of gravitational waves since the Big Bang produces a stochastic background, which could be detected by cross-correlating two (or more) detectors, such as the interferometers Virgo near Pisa in Italy or LIGO in the US, or LISA in space.
We can distinguish between two contributions to the stochastic background: a background from cosmological origin, memory of the early stages of the Universe, just a fraction of second after the Big Bang, and a background of astrophysical origin, memory of the evolution of the galaxies and star formation.
The current detection methods assume that the stochastic background is Gaussian. The purpose of this thesis is to model the gravitational stochastic background, study its statistical properties and develop a detection method for non Gaussian stochastic backgrounds. Such backgrounds could have been produced for instance by cusps of cosmic strings, or by astrophysical populations such as compact binary coalescences or core collapses to black holes or neutron stars. This method could also be extended to the case when the stochastic backgrounds are not isotropic.

Cosmology with compact binaries coalescences: prospect for next generation GW detectors
Prof. Tania Regimbau (regimbau@oca.eu)

Gravitational Waves ground based experiments, after a decade of detector installation and commissioning, have reached or surpassed their design sensitivities, opening a new window into the Universe, and allowing coincident searches with electromagnetic (gamma, X, radio) or neutrino detectors.
The first generation of ground-based interferometers has already put interesting astrophysical constraints on the ellipticity of the Crab or the gravitational wave stochastic background. With the second generation that will see the light in the next few years, we expect to see at least close compact binary coalescences, while third generation detectors such as the planned Einstein Telescope, or the space antenna LISA, would bring GW astronomy to the next level, when it would be possible to detect address on a wide variety of sources, and address a large range of problems in astrophysics, fundamental physics and cosmology.
In particular, coalescing compact binaries, that could be detected up to very large redshifts (z~2-3 for double neutron stars), for some of them in coincidence with short GRB, and whose waveform is well modeled up to the last stable orbit, are ideal standard candles (or standard sirene) to study dark energy. Regimbau et al. has developed a simulator for Advanced LIGO/Virgo and the Einstein Telescope Mock Data Challenge, that permits to generate cosmological populations of sources up to redshifts of z~10. One of the goal of this Phd is to make this tool more realistic, using for instance results from the binary evolution code starTrack.
The next step would be to use these simulations to model the population detectable by Advanced detectors or Einstein Telescope, eventually in coincidence with GRB observations, and study the constraints one could but on the mass of the graviton, cosmological parameters or dark energy equation of state, and compare them to the constraints derived from supernovae or galaxy clusters observations.