[Hybrid Zoom/in vivo]

The problem of the missing mass of the Universe is one of the most puzzling conundrums of modern physics. Assuming that it stems from the existence of yet unknown dark matter (DM) particles, one can probe the cosmos in the hope of detecting unambiguous signatures thereof. The properties of cold DM (CDM) particle candidates lead to the prediction that gravitational structures form on scales much smaller than typical galaxies, below the resolution scope of current cosmological simulations. This clustering translates into a large population of subhalos in galaxies, hence in the Milky Way. Only analytical models can encompass and describe their full distribution. My work is centred upon building a theoretically consistent and dynamically constrained subhalo population model. This relies mostly on three ingredients: the minimal subhalo mass, the cosmological mass function, and dynamical effects. I start from a simplified model of thermally produced dark matter (with scalar, pseudo-scalar, vector and axial-vector interactions with standard model particles). I classify models in terms of small-scale cut-off on the matter power spectrum, which is directly related to the minimal halo mass, making explicit the role of velocity dependence in the interaction processes. Secondly, I improve on the determination of the cosmological subhalo mass function by deriving it from the excursion set theory and a merger tree, while it was previously calibrated on simulations. Thirdly, I incorporate new dynamical effects by analytically studying the impact of stellar encounters susceptible of occurring in galactic discs. Besides, I illustrate an application of this model for indirect detection experiments looking for traces of DM annihilation in the Milky Way. In light of the 1525 unassociated point sources discovered by the Fermi-LAT collaboration, I give prospects for the detection of point-like subhalos as gamma-ray sources with Fermi-LAT-like and CTA-like instruments. I also shortly discuss the impact of additional effects, like the Sommerfeld effect.

PhD thesis committee:
Pr. Shin'ichiro Ando (GRAPPA-Amsterdam), Pr. Céline Boehm (U. Sydney), Dr. Marco Cirelli (LPTHE), Pr. Jorge Peñarrubia (U. Edinburgh), Pr. Pierre Salati (USMB & LAPTh), Dr. Pasquale Serpico (LAPTh), Pr. Tracy Slatyer (MIT-Boston).

In the coalescence of black holes, the ringdown phase consists of the relaxation of the highly perturbed, newly formed black hole to its equilibrium state. Quasi-normal modes (QNM) are associated with the complex frequencies which characterise the dynamics of this relaxation phenomenon (oscillation + damping). Observations of the ringdown could be extremely effective to constrain (or reveal) potential deviations from General Relativity (GR) if the QNM spectrum is found to agree (or disagree) with predictions of GR. This is also an ideal playground to test alternative theories of gravitation. However, in these alternative theories of gravity, not only the problem of computing QNM is generally (much) more involved than in GR but also it is not clear whether one could technically proceed as in GR. For these reasons, with a few exceptions, there has been few results on black holes perturbation theory and QNM in modified theories of gravity. In this seminar, I will present a novel (and potentially powerful) approach we have recently developped in a series of paper to study black holes perturbation theory and to compute QNM in a very large class of theories of modified gravity. I will illustrate the method with physically relevant examples for which the usual techniques are not available and I will present new interesting physical effects concerning QNM in scalar-tensor theories. This new approach paves the way for a systematic treatment of QNM in theories of modified gravity and might be an important step towards a deeper understanding of gravitation in the strong field regime.

In this talk, I will first describe our approach to test gravity with the Dark Energy Survey weak lensing and clustering measurements. Weak gravitational lensing is indeed a powerful probe of the laws of gravity on cosmological scales and its measurements by galaxy surveys such as Euclid or LSST will enable an even deeper search of modified gravity. In the second part of this talk, I will present a new approach I proposed to classify modified gravity theories using unsupervised learning in order to efficiently test gravity with these future surveys. I will show preliminary results and conclude on the outlooks for the future surveys of the 2020 decade.

Transport processes occurring in the radiative interior of solar-type stars and low-mass stars are evidenced by the surface variation of light elements, in particular lithium (Li) and beryllium (Be), as well as by the evolution of the stellar rotation rate. While there are many independent transport models for angular momentum and chemical species, there is a lack of self-consistent theories that permit stellar evolution models to simultaneously match the present-day observations. Low-mass stars (F- to K- spectral type) of Population I are characterised by an important Li-depletion during their evolution from the pre-main sequence to the age of the Sun, as well as the so-called Li-dip (and Be-dip) for F-type stars and observed for a few open clusters like the Hyades. Rotation-induced mixing has been seen as an insufficient process to fully explained the observed Li depletion. The aim of the PhD thesis is to explore how additional transport processes can improve the agreement between evolutionary models and observational constraints. Using the stellar evolution code STAREVOL, we explored several transport processes from the literature for chemicals and for angular momentum (penetrative convection, tachocline mixing, additional turbulence, additional viscosity). The need for additional transport processes in stellar evolution models for both chemicals and angular momentum in addition to atomic diffusion, meridional circulation, and turbulent shear is confirmed. We identified the rotational dependence of the penetrative convection as a key process. Two additional and distinct parametric turbulent mixing processes (one for angular momentum and one for chemicals) are required to simultaneously explain the observed surface Li depletion and the solar internal rotation profile. This is also necessary for a larger range of masses from 0.8 to 1.2 Msol in order to reproduce the Li depletion trend with decreasing effective temperature on the cold side of the dip. Finally, the Li (and Be) dip, for stars of mass 1.3-1.5 Msol, displays a different trend, and we have shown that a readjustment of the prescription and parameters, notably for shear turbulence and magnetic braking, are needed to reproduce it.

In this talk I will discuss the main motivations to explore theories of gravity beyond General Relativity. In particular, I will focus on the popular class of scalar-tensor theories, discussing their formulation and how to parametrise them efficiently on a cosmological background. This framework is optimal to describe large class of models at once, which makes it ideal to test deviations from the standard cosmological model. I will discuss some of the main results in the literature, as well as the tools available to compare cosmological data with theoretical predictions. I shall focus mostly on hi_class, an Einstein-Boltzmann solver optimised for beyond-Horndeski models. Finally, I will speculate about the future of alternative theories of gravity and show what we can realistically expect from the next generation of surveys.

The ΛCDM concordance model is very successful at describing our Universe with high accuracy and only a few parameters. Despite its successes, a few persistent tensions remains to this day: most notably, the best-fit ΛCDM model, as derived from the Planck CMB data, largely overpredicts the abundance of Sunyaev–Zel’dovich clusters when using their standard mass calibration. Whether this is the sign of an incorrect calibration or the need for new physics remains a matter of debate. In this talk, I will present the results from our exploration of extensions of the standard model -- massive neutrinos and a simple modified gravity model -- and their ability to release the aforementioned tension, pitted against a potential bias in the mass calibration of clusters. Under certain assumptions, we find that our conclusions may call for new physics in the gravitational sector.

The beta-function formalism was introduced to identify universality among the wide zoology of inflationary models. It relies on a formal analogy between the equation describing the evolution of a scalar field in an expanding background and the renormalization group equation in quantum field theory. In this framework, near-scale invariance experienced by the Universe during inflation is interpreted as a departure of the newly defined cosmological beta-function from a fixed point corresponding to an exact de Sitter spacetime. A single parametrization of the beta-function, close to the fixed point, defines a universality class of models that share a single asymptotic behavior. In this talk, I will illustrate the general ideas of the approach by considering the simplest realization of inflation, then introduce the analytical gains of the method on more sophisticated scenarios, such as warm inflation, and I will conclude with phenomenological considerations in view of the next generation of cosmological experiments.

Modern theories that aim at improving the theory of Einsten often induce effects that can be searched for at all scales. Indeed, many of them operate a change of the whole core theory (standard model of particle + general relativity) at the same time. In this seminar, I will show various phenomenological consequences of some of these theories in order to illustrate this. I shall notably show how a variation of the fundamental coupling constant follows from basic properties of the low-energy effective action of string theories, and what it may induce for the dynamics of the universe, as well as for various observable in the universe (such as the distance luminosity for instance), but also for the ratio between the time given by different atomic clocks, or for the universality of free fall. I shall show that one can also build interesting dark matter candidates from similar considerations, and test their presence with various probes. Along the way I shall also present some consequences alternative theories induce on gravitational waves, and how one can notably make use of gravitational wave detectors in order to constrain those alternatives.

Compact astrophysical objects, like black holes and neutron stars, are opening a new and enlightening window on gravity. The study of such objects in strong curvature regimes now allows us to test alternative gravity theories, and notably the existence of other fundamental fields. I will discuss no-hair theorems, and how one can still build solutions that deviate from general relativity. I will investigate the robustness of these solutions, through their stability or the existence of a well-posed Cauchy problem. I will also discuss the associated observational imprints, focusing in particular on the interplay with dark energy.