In this thesis I present our recent work on gamma-ray searches for dark matter with the Fermi Large Area Telescope (Fermi-LAT). We have targeted dwarf spheroidal galaxies since they are very dark matter dominated systems, and we have developed a novel joint likelihood method to combine the observations of a set of targets. In the first iteration of the joint likelihood analysis, 10 dwarf spheroidal galaxies are targeted and 2 years of Fermi-LAT data is analyzed. The resulting upper limits on the dark matter annihilation cross-section range from about 10−26 cm3 s−1 for dark matter masses of 5 GeV to about 5 × 10−23 cm3 s−1 for dark matter masses of 1 TeV, depending on the annihilation channel. For the first time, dark matter models with a cross section above the canonical thermal relic cross section (∼ 3 × 10−26 cm3 s−1) are strongly disfavored by a gamma-ray experiment. In the second iteration we include 15 dwarf spheroidal galaxies in the combined analysis, employ 4 years of data and an improved calculation of the dark matter density. The obtained upper limits range from about 10−26 cm3 s−1 for dark matter masses of 2 GeV to about 10−21 cm3 s−1 for dark matter masses of 10 TeV, depending on the annihilation channel. I briefly describe some of the evidence for dark matter, the Fermi-LAT instrument and public data releases, dwarf spheroidal galaxies, likelihood analysis, and results from analyses of Fermi-LAT data. I also document some of the tests made to verify the method and to compare different analysis setups.

Although exotic objects like supermassive black holes (SMBHs) and dark matter halos do not emit or interact with light, we can still detect them across the vastness of space. By observing the gravitational dance of objects we can see, astronomers are able to infer the mass of the invisible objects they orbit. This has led to the discovery that nearly every massive galaxy hosts a SMBH at its center, and has confirmed that every galaxy is embedded in an extended halo of dark matter. However, the practice of inferring mass from the motions of bright kinematics tracers has many complications, not the least of which is that we seldom observe more than the line-of-sight component of the instantaneous velocity of a star. Consequently, astronomers must build dynamical models of the galaxies they wish to study. These models often rely on overly restrictive assumptions, or are crippled by degeneracies amongst their parameters and lack predictive power. In this thesis, I introduce a significant advancement into the field of dynamical modeling. My modeling technique is based on the powerful principle of orbit superposition, also known as Schwarzschild Modeling. This technique is robust to many of the degeneracies associated with dynamical modeling, and has enjoyed much success in measuring the SMBHs and dark matter halos of large elliptical or bulge-dominated galaxies. I use it in Chapter 2 to accurately measure the SMBH in the Sombrero Galaxy (NGC 4594) and to constrain its dark matter halo. Unfortunately, when measuring dark matter halos with Schwarzschild Modeling, the modeler is required to adopt a parameterization for the dark matter density profile. Often this is precisely the quantity one wishes to measure. To avoid this reliance on a priori parameterizations, I introduce a technique that calculates the profile non-parametrically. Armed with this powerful new technique, I set out to measure the distribution of dark matter in the halos of some of the smallest galaxies in the Universe. These dwarf spheroidal galaxies (dSphs) orbit the Milky Way as satellites, and their dark matter content has been studied extensively. However, the models used to probe their halos have been simplistic and required overly restrictive assumptions. As a result, robust conclusions about their dark matter content have remained elusive. Into this controversial and active area of study, I bring Non-Parametric Schwarzschild Modeling. The results I find offer the most robust and detailed measurements of the dark matter profiles in the dSphs to date. I begin my study with the first application of standard Schwarzschild Modeling to any dSph galaxy by using it in Chapter 3 on Fornax. This chapter details the process of re-tooling Schwarzschild Modeling for the purpose of measuring these small galaxies. In Chapter 4, I introduce the fully non-parametric technique, and apply it to Draco as proof of concept. Chapter 5 presents the main results of this thesis. Here I apply Non-Parametric Schwarzschild Modeling to Draco, Carina, Fornax, Sculptor, and Sextans. By relaxing the assumption of a parameterization for the dark matter profile, I find a variety of profile types in these five galaxies---some of which are consistent with past observations, others consistent with predictions from simulations, and still others were completely unanticipated. Finally, in Chapter 6 I describe the modeling of these galaxies in more detail. I demonstrate the accuracy of Non-Parametric Schwarzschild Modeling by recovering a known dark matter profile from artificial simulated data. I also expound upon the modeling results by presenting the detailed orbit structure of the five dSphs. Lastly, I compare my results to hydrodynamical simulations to explore the link between dark matter profile type and the baryon content of the dSphs. / text

No description available.

No description available.

No description available.

Dark matter (DM), whose nature and interaction mechanisms are still an open issue, constitutes about 25 % of the Universe energy density. Weakly Interacting Massive Particles (WIMPs) are considered as strong candidates for particle DM and their search is conveniently carried out through the detection of gamma rays. The newly discovered ultra-faint dwarf spheroidal galaxies (dSphs), located in the vicinity of the Galaxy, exhibit high values of the mass to luminosity ratio, and are therefore considered as strongly dominated by DM. These objects are within reach of the Cherenkov Telescope Array (CTA), which is the future project for gamma-ray astronomy, with an better sensitivity (one order of magnitude) with respect to the current generation experiments. The main goal of the present work is the study of the sensitivity of CTA to WIMPs DM particles, by simulating the observation of the ultra-faint dwarf spherical galaxies Triangulum II, Reticulum II and Carina III, as well as of the classical dwarf galaxy Sculptor, for different annihilation channels, between 70 GeV and 100 TeV. The sensitivity curve in the WIMPs parameter space (velocity-averaged annihilation cross section < &sigma; &nu; > and DM mass mDM) was computed. We found that, within the sample of dwarf galaxies tested, Triangulum II is the most promising source, able to reach the thermal freeze-out values in the annihilation channel &tau;+&tau;- for only 50 hours of observation. Our result, the first estimation of the sensitivity for DM searches in ultra-faint dwarfs with CTA, is consistent with results from current generation experiments, showing better performance over an extended energy range. The limited sample of available stars in the targets induces uncertainties on the DM content. Future measurements, leading to a better understanding of the sources dynamic equilibrium, can improve this situation. Even so, the combination of the high DM content in the ultra-faint dwarf galaxies, together with the excellent expected performance of the future CTA, provides a promising result for indirect DM searches. / A matéria escura, cuja natureza e mecanismos de interação ainda estão em aberto, compõe 25% da densidade de energia do Universo. Weakly Interacting Massive Particles (WIMPs) apresentam-se como forte candidatas e sua busca é convenientemente conduzida através de raios gama. As recém descobertas galáxias esferoidais anãs ultra-fracas, situadas nos arredores da Galáxia, apresentam altos valores da razão entre massa e luminosidade, sendo portanto consideradas objetos fortemente dominados por matéria escura. Esses objetos estão ao alcance do Cherenkov Telescope Array (CTA), que é o futuro projeto da astronomia gama, com sensibilidade de uma order de grandeza melhor do que os experimentos atuais. O presente trabalho teve como objetivo estudar o potencial de detecção indireta de WIMPs através de raios gama com o futuro observatório CTA, observando as galáxias esferoidais anãs ultra-fracas Triangulum II, Retículum II e Carina III e a galáxia anã clássica Sculptor, para diferentes canais de aniquilação, entre 70 GeV e 100 TeV. A curva de sensibilidade no espaço de parâmetros livres de WIMPs (massa da partícula mDM e médida da seção de choque de aniquilação ponderada pela velocidade < &sigma; &nu; >) foi calculada. Nós encontramos que dentro da amostra de galáxias anãs testadas, Triangulum II é a fonte mais promissora, capaz de testar os valores térmicos no canal de aniquilação &tau;+&tau;- considerando apenas 50 horas de observação pelo CTA. Nosso resultado, a primeira estimativa da sensibilidade para busca de matéria escura em galáxias esferoidais anãs ultra-fracas com CTA, é consistente com resultados de experimentos da geração atual, e mostra um melhor desempenho em uma faixa de energia estendida. Os resultados são afetados pelas incertezas devido à pequena amostra de estrelas dos alvos escolhidos, que se reflete no conhecimento do conteúdo de matéria escura. Novas medidas podem ajudar a esclarecer essa situação. Ainda assim, a combinação de galáxias anãs ultra-fracas, aliada às melhorias do futuro CTA, apresenta-se como um passo muito promissor para buscas indiretas de matéria escura.

De nombreuses observations astrophysiques indiquent l'existence de grandes quantités de masse manquante dans l'Univers, et ce de l'échelle galactique à l'échelle cosmologique. Découvrir la nature de cette masse invisible constitue le problème de la matière noire, qui apparaît comme l'un des enjeux majeurs de la physique moderne. Cette thèse s'inscrit dans le contexte de la détection indirecte de matière noire. Cette dernière serait composée de nouvelles particules élémentaires, dont les produits d'annihilation pourraient être observés dans le rayonnement cosmique. Nous étudions dans ce travail deux des canaux de recherche les plus prometteurs : les photons gamma et les anti-noyaux.Les objets astrophysiques permettant de placer les meilleures contraintes actuelles en rayons gamma sont les galaxies naines sphéroïdes (dSphs) de la Voie Lactée. La première partie de notre travail a été consacrée à contraindre les facteurs J d'annihilation de ces objets, qui quantifient l'amplitude des flux gamma attendus. Nous avons pour cela mis au point une configuration optimisée d'analyse de Jeans, pour reconstruire les profils de densité de matière noire et leurs incertitudes à l'aide des données cinématiques stellaires. Notre configuration a été obtenue à l'aide de tests systématiques sur de très nombreuses dSphs simulées, et nous l'avons appliquée à vingt-trois dSphs de la Voie Lactée. La seconde partie de notre travail a consisté à mener une recherche de noyaux d'anti-hélium dans les données collectées par l'expérience AMS-02 sur la Station Spatiale Internationale. Nous avons pour cela mis au point une classification par arbres de décision boostés, et notre analyse préliminaire a permis d'obtenir les meilleures contraintes actuelles sur les rapports anti-hélium sur hélium. / Many astrophysical observations suggest the existence of large amounts of missing mass in the Universe, from the galactic to the cosmological scale. Discovering the nature of this invisible mass forms the dark matter problem, which appears as one of the major challenges of modern physics. This thesis is established in the context of indirect detection of dark matter. The latter could consist of new elementary particles, whose annihilation products may be observed in cosmic rays. We study in this work two of the most promising research channels!: gamma-rays and anti-nuclei.The best constraints on dark matter properties from gamma-ray observations come from the dwarf spheroidal galaxies (dSphs) of the Milky Way. The first part of our work was devoted to computing the annihilation J-factors of these objects, which quantify the magnitude of the expected gamma-ray flux. We have developed an optimized Jeans analysis setup in order to reconstruct the dark matter density profiles of these objects and their associated uncertainties, using stellar kinematic data. Our optimized setup was obtained using systematic tests on numerous simulated dSphs, and we applied it to twenty-three dSphs of the Milky Way. The second part of our work was dedicated to the search for anti-helium nuclei in the cosmic ray data collected by the AMS-02 experiment on the International Space Station. We have developed a classification method using boosted decision trees, and our preliminary analysis has led to the best constraints to date on the anti-helium to helium ratio.

L’Univers est dominé par une composante invisible appelée Matière Noire (MN), de nature inconnue mais dont les effets gravitationnels sur la matière visible sont clairement observés. Il a été proposé que la MN soit constituée de particules massives et interagissant faiblement avec la matière, permettant ainsi de concilier théorie, observations et simulations. L’annihilation de ces particules dans les régions où la MN est fortement concentrée pourrait produire des rayons γ de très haute énergie dont les signatures spectrales peuvent être détectées par le réseau de télescopes H.E.S.S. Un excès à ~3σ est observé dans la direction de la Galaxie Naine du Sagittaire, avec la méthode standard d’analyse ON-OFF. Plus de données sont nécessaires pour conclure quant à son origine. La seconde partie du travail concerne la recherche de raies spectrales en γ dans la région du Centre Galactique. Une méthode de Maximum de Vraisemblance Complète a été développée, étalonnée et appliquée à une fraction d’un ensemble de 20h de données prises en 2014. Aucun excès de γ n’étant observé, des limites sur la section efficace d’annihilation de la MN sont produites pour des masses de 100 GeV à 2 TeV, la sensibilité de H.E.S.S. à basse énergie étant obtenue par l’ajout d’un 5ème télescope depuis 2012. Ces limites complètent efficacement les précédents résultats de Fermi-LAT et H.E.S.S. D’autre part l’analyse finale devrait permettre d’exclure un potentiel signal à 130 GeV observé dans les données de Fermi-LAT en 2012 et ce avec plus de 95% CL, et de proposer les limites les plus solides à ce jour sur les modèles d’émission de raies spectrales en γ dans le domaine d’énergie couvert par H.E.S.S. / The Universe is full of gravitational evidence of a dominant invisible Dark Matter (DM) component at the Galactic and cosmological scales. Although its nature is still one of the major puzzles of the 21st Century, Weakly Interacting Massive Particles (WIMPs) are an excellent scenario for matching theoretical predictions with observations and simulations. In particular, their self-annihilations would give rise to characteristic spectral signatures in γ-rays, detectable at Very High Energies (VHE) with the H.E.S.S. telescope array in regions such as the Galactic Centre (GC) and Dwarf Spheroidal Galaxies (dSphs). The standard ON-OFF analysis method is applied in the observation of the Sagittarius dSph where a ~3σ hotspot is observed above 300 GeV, although more statistics is required to conclude on its potential DM origin. The second part of the work is focused on the search for monochromatic γ-ray line signatures in the GC region. A Full Likelihood method has been developed, calibrated with Monte-Carlo simulations and applied to a sub-sample of a 20h dataset acquired in 2014. No excess signal is found, thus leading to limits on the DM annihilation cross-section down to a 100 GeV mass range, the sensitivity at the lowest energies being achieved by the 5th H.E.S.S. telescope added in 2012. These limits efficiently fill the gap in mass between results from Fermi-LAT and the first phase of H.E.S.S. On the other side the analysis of the complete dataset is expected to exclude the 130 GeV line-like feature recently reported in the Fermi-LAT data, with more than 95% CL, and to provide the most constraining DM limits so far on γ-ray line emission in the VHE range.