Cosmology and Gravitation

Several mechanisms can lead to production of particles during inflation. I discuss how this phenomenon can induce a contribution to the primordial spectrum of gravitational waves with unusual properties: the tensors produced this way can violate parity; can have a large three-point function; can have a relatively large tensor-to-scalar ratio even if inflation occurs at low energies; finally, their spectrum can display a feature that can be directly detected by second-generation gravitational interferometers such as advanced LIGO.

The time variation of physical constants has been much discussed in the literature, motivated by claims of fine structure constant variations together with several theoretical ideas. Although it is well understood (by most, but not all!) cosmologists that one must consider only dimensionless constants, most discussions of the strength of gravity involve "G", which is of course dimensional. I discuss some applications of variations of "G" on cosmological observables, stressing the need to stay dimensionless. "Constants" might also vary in space. An idea which is perhaps less crazy is that cosmological parameters might vary across the observable Universe. I show how this leads to dipole modulation, which corresponds to a correlation between neighbouring multipoles in maps of the cosmic microwave sky. Searches for such signals could lead to constraints on the variation of the cosmological model on the largest accessible scales.

A recently discovered class of active galactic nuclei, TeV luminous blazars, constitute a small fraction of the power output of black holes. Nevertheless, there are suggestions that unlike the UV and X-ray luminosity of quasars, the very-high energy gamma-ray emission from the TeV blazars can be thermalized on cosmological scales with order unity efficiency, resulting in a potentially dramatic heating of the low-density intergalactic medium. The way in which this occurs, however, imparts a variety of peculiar properties to this novel heating source, resulting in a number of robust cosmological consequences. I will discuss the process by which TeV blazars heat the Universe, the strange properties that this heating has, and the variety of signatures that it has left behind, many of which have already been observed!

I will introduce the gravitational microlensing, its application to the compact dark matter detection and the extra-solar planet observations. EROS has been performed the microlensing observation in four directions of the Galactic plane, away from the Galactic center. I will report the observational results and the interpret the data within the Standard Galactic model. As a result we extract the best fit to the dust contribution in the Galactic disk, orientation of the Galactic bar and the abundance of the red giants compare to local stellar distribution.

I'll discuss how to systematically construct a (d+2)-dimensional solution of the vacuum Einstein equations that is dual to a (d+1)-dimensional fluid satisfying the incompressible Navier-Stokes equations with specific higher-derivative corrections. The solution takes the form of a non-relativistic gradient expansion that is in direct correspondence with the hydrodynamic expansion of the dual fluid. The dual fluid has nevertheless an underlying description in terms of relativistic hydrodynamics, with the unusual property of having a vanishing equilibrium energy density. Using the gravitational results, as well as an interesting and exact constraint on its stress tensor, we identify the transport coefficients of the dual fluid. A simple Lagrangian model is sufficient to realise its key properties.

Cosmic strings are predicted to arise in both inflationary and non-inflationary cosmological models. The signatures of such strings will stand out particularly well at higher redshifts. I will discuss how to look for these signatures in CMB redshift and polarization maps and in 21cm redshift surveys.

String theory should give a well-defined answer to the following question: What is the state of matter in the limit of infinite energy density? We use results obtained from the understanding of black hole entropy to conjecture this equation of state, noting that the maximum entropy state in string theory has vastly more entropy than the states used in traditional approaches to early Universe Cosmology. The evolution of the Universe with this equation of state can be obtained in closed form.

It is argued that the correct quantization of a scalar field theory in de Sitter spacetime involves a de Sitter invariant state which is not the Bunch-Davies vacuum. A novel but natural de Sitter invariant alternative exists and it is suggested that this and is the prefered state for scalar field theories. The argument is based on the exact solution of an interacting scalar field theory.

Though the observed CMB is at very low energy, it encodes ultra high-energy physics in spatial variations of the photon temperature and polarization fluctuations. This effect is believed to be dominated by the initial quantum state of the Universe. I will describe the first theoretical tools by which to construct such a state from fundamental physics. There are three specific observational effects this initial state will produce: a ringing signal in the power spectrum of quantum field fluctuations, an enfolded type of non-Gaussian fluctuations, and a calculable primordial gravitational wave background. We may soon be able to compare these predictions against experiment, allowing one to rule out classes of quantum gravity models. Now is the critical time to undertake such investigations, with a number of ongoing and planned experiments such as WMAP, Planck, and CMBPol poised to collect a wealth of precision data.

Recently, emergent phenomena have started to attract more attention. Instead of assuming a symmetric world, one begins with a chaotic one. In this talk, I will describe this picture, discuss the main constraints on emergence, and then present a few phenomenological procedures that can be implemented to study the emergent phenomena.

The model of local non-Gaussianity, parameterized by the constant non-linearity parameter fNL, is an extremely popular description of non-Gaussianity. However, a mild scale-dependence of fNL is natural. This scale dependence is a new observable, potentially detectable with the Planck satellite, which helps to further discriminate between models of inflation. It is sensitive to properties of the early universe which are not probed by the standard observables. In a complementary way, the trispectrum also contains important information about non-Gaussianity which the bispectrum does not capture. We explicitly calculate the scale dependence and trispectrum in several models including one with a very large infrared-loop contribution to the bispectrum and in various realizations of the curvaton scenario.