In this dissertation we aim to perform a multi-wavelength analysis of galaxy clusters. We discuss various methods for clustering in order to determine physical parameters of galaxy clusters required for this type of study. A selection of galaxy clusters was chosen from 4 papers, (Popesso et al. 2007b, Yoon et al. 2008, Loubser et al. 2008, Brownstein & Mo at 2006) and restricted by redshift and galactic latitude to reveal a sample of 40 galaxy clusters with 0.0 < z < 0.15. Data mining using Virtual Observatory (VO) and a literature survey provided some background information about each of the galaxy clusters in our sample with respect to optical, radio and X-ray data. Using the Kayes Mixture Model (KMM) and the Gaussian Mixing Model (GMM), we determine the most likely cluster member candidates for each source in our sample. We compare the results obtained to SIMBADs method of hierarchy. We show that the GMM provides a very robust method to determine member candidates but in order to ensure that the right candidates are chosen we apply a select choice of outlier tests to our sources. We determine a method based on a combination of GMM, the QQ Plot and the Rosner test that provides a robust and consistent method for determining galaxy cluster members. Comparison between calculated physical parameters; velocity dispersion, radius, mass and temperature, and values obtained from literature show that for the majority of our galaxy clusters agree within 3 range. Inconsistencies are thought to be due to dynamically active clusters that have substructure or are undergoing mergers, making galaxy member identi cation di cult. Six correlations between di erent physical parameters in the optical and X-ray wavelength were consistent with published results. Comparing the velocity dispersion with the X-ray temperature, we found a relation of T0:43 as compared to T0:5 obtained from Bird et al. (1995). X-ray luminosity temperature and X-ray luminosity velocity dispersion relations gave the results LX T2:44 and LX 2:40 which lie within the uncertainty of results given by Rozgacheva & Kuvshinova (2010). These results all suggest that our method for determining galaxy cluster members is e cient and application to higher redshift sources can be considered. Further studies on galaxy clusters with substructure must be performed in order to improve this method. In future work, the physical parameters obtained here will be further compared to X-ray and radio properties in order to determine a link between bent radio sources and the galaxy cluster environment. / MSc (Space Physics), North-West University, Potchefstroom Campus, 2013

In this dissertation we aim to perform a multi-wavelength analysis of galaxy clusters. We discuss various methods for clustering in order to determine physical parameters of galaxy clusters required for this type of study. A selection of galaxy clusters was chosen from 4 papers, (Popesso et al. 2007b, Yoon et al. 2008, Loubser et al. 2008, Brownstein & Mo at 2006) and restricted by redshift and galactic latitude to reveal a sample of 40 galaxy clusters with 0.0 < z < 0.15. Data mining using Virtual Observatory (VO) and a literature survey provided some background information about each of the galaxy clusters in our sample with respect to optical, radio and X-ray data. Using the Kayes Mixture Model (KMM) and the Gaussian Mixing Model (GMM), we determine the most likely cluster member candidates for each source in our sample. We compare the results obtained to SIMBADs method of hierarchy. We show that the GMM provides a very robust method to determine member candidates but in order to ensure that the right candidates are chosen we apply a select choice of outlier tests to our sources. We determine a method based on a combination of GMM, the QQ Plot and the Rosner test that provides a robust and consistent method for determining galaxy cluster members. Comparison between calculated physical parameters; velocity dispersion, radius, mass and temperature, and values obtained from literature show that for the majority of our galaxy clusters agree within 3 range. Inconsistencies are thought to be due to dynamically active clusters that have substructure or are undergoing mergers, making galaxy member identi cation di cult. Six correlations between di erent physical parameters in the optical and X-ray wavelength were consistent with published results. Comparing the velocity dispersion with the X-ray temperature, we found a relation of T0:43 as compared to T0:5 obtained from Bird et al. (1995). X-ray luminosity temperature and X-ray luminosity velocity dispersion relations gave the results LX T2:44 and LX 2:40 which lie within the uncertainty of results given by Rozgacheva & Kuvshinova (2010). These results all suggest that our method for determining galaxy cluster members is e cient and application to higher redshift sources can be considered. Further studies on galaxy clusters with substructure must be performed in order to improve this method. In future work, the physical parameters obtained here will be further compared to X-ray and radio properties in order to determine a link between bent radio sources and the galaxy cluster environment. / MSc (Space Physics), North-West University, Potchefstroom Campus, 2013

Les Noyaux Actifs de Galaxies (NAG) sont parmi les sources les plus énergétiques dans l'Univers. Certains possèdent un jet relativiste dont l'émission est purement non-thermique. Lorsque le jet est aligné avec notre ligne de visée, ces objets appelés "blazars" voient leur émission amplifiée par effet Doppler relativiste. Depuis l'avènement de l'astrophysique gamma aux très hautes énergies (THE; E>100 GeV), les télescopes Cherenkov comme H.E.S.S. ont observé près d'une trentaine de NAG, principalement des blazars, depuis le sol. Ils tirent parti du rayonnement Cherenkov issu des gerbes de particules secondaires engendrées par l'arrivée d'un photon gamma dans l'atmosphère terrestre, pour remonter aux propriétés du photon incident et ainsi étudier ces sources extragalactiques. Nous avons étudié l'émission gamma THE rapidement variable du blazar PKS 2155-304, dont deux éruptions majeures ont été détectées en Juillet 2006, grâce à un modèle synchrotron self-Compton (SSC) dynamique. Cette émission variable présente des propriétés extrêmes excluant certains scénarios d'émission standard des blazars. Nous avons aussi développé un modèle d'émission SSC de jets relativistes non alignés avec la ligne de visée, afin d'interpréter la découverte récente par H.E.S.S. de deux radio galaxies, M 87 et Cen A, qui prouve l'émergence d'une nouvelle famille d'émetteurs cosmiques au TeV. Nous concluons avec une étude systématique menée sur l'ensemble des NAG actuellement connus au TeV, à l'aide d'un modèle SSC stationnaire. Nous présentons des diagnostics de prédictions de densités de flux dans ces objets, pouvant être confrontés aux observations futures du Cherenkov Telescope Array (CTA).

This thesis presents studies of cosmic populations of extragalactic radio sources. The problems selected for this thesis are 1) the derivation of constraints on the emergence of new sub-mJy populations at flux density below about 1mJy (at1.4 GHz) paying careful attention to including sources with low surface brightness and counting sources rather than components 2) development of a new method to estimate the asymmetry in the large scale galaxy environment with respect to the axes of extended radio sources and use this to examine for evidence of impact of the environment on the morphology of radio sources. The studies presented herein have been carried out using the Australia Telescope Low Brightness Survey (ATLBS), which is a sensitive radio survey at 1.4 GHz, imaging 8.42 square degrees of the sky along with accompanying optical observations of the same region. There are multiple populations of extragalactic radio sources in the cosmos. These consist of populations of powerful radio-loud quasars and radio galaxies to populations of weaker radio sources such as star-forming galaxies. These populations of radio sources show evidence of evolution with cosmic epoch. Because the radio galaxy phenomenon and the AGNs at the centers of their host galaxies may influence the evolution of the galaxy via feedbacks, examinations of these source populations over cosmic time are a necessary complementary study to understanding the process of galaxy formation and in general, cosmology. Below we give brief introduction to the problems studied in this thesis. Sub-mJy Radio Source Counts Radio source counts, which have historically been a key probe of cosmology, now represent a measure of cosmological evolution in radio source populations. Currently, the estimation of source counts at sub-mJy flux density as well as the nature and evolution of these sources is undetermined. At flux densities ≤1.0 mJy a ‘flattening’ of normalized differential source counts has been widely reported in literature( Windhorst et al.(1985),Hopkins et al.(2003),Huynh et al.(2005) and references therein). The flattening is observed as an apparent change of slope for the normalized differential source counts from ∼0.7 at5.0 −100.0 mJy to about 0.4 in the 0.25 −5.0 mJy range. Attempts to understand the nature of the sub-mJy population have arrived at discordant results and identify the sub-mJy sources with different populations: starburst galaxies(Condon(1989), Benn et al. (1993), Huynhet al. (2005)), early type galaxies (Gruppioniet al., 1999),low(radio) luminosity activegalactic nuclei(AGNs)(Huynh et al.,2008) or a mixture of these. Due to unavailability of spectroscopically complete samples of hosts of sub-mJy sources, the exact nature of the sub-mJy radio source population is currently uncertain. However, the presence of a population which emerges at sub-mJy flux density and is different from the AGN-dominated radio source population is not in doubt. The studies in the literature are inconsistent in identifying the precise location of the emergence of flattening in counts at sub-mJy flux density. Several studies show that the source counts are consistent with a continuation in the slope of the differential counts below mJy flux density (Prandoniet al.(2001) and Subrahmanyan et al.(2010)). The scatter in the sub-mJy counts from different studies may be because of the relatively small areas covered by deep surveys(in many cases, a single pointing of an interferometric array) which may have relatively large errors arising from large scale clustering in the spatial distribution of cosmic radio sources(however the study by Condon (2007) concludes that the scatter in the source counts stems from variations in corrections and sensitivity in different studies)In contrast, wide-field surveys may not reach the depth to probe sub-mJy counts. Another reason is the correction applicable to the observed source counts necessary to estimate the true source counts; these are especially pertinent at low flux densities. To resolve these is-sues, a survey which combines the attributes of wide spatial coverage as well as excellent sensitivity and a procedure which accounts for the biases in estimation of the sub-mJy source counts is needed. In conclusion, accurate measurements of the source counts at sub-mJy flux densities are needed to correctly estimate the cosmic evolution of radio sources. Environments of Extended Radio Sources Another issue of importance in the study of extragalactic radio sources is their interaction with their environments. The gas environments in which radio sources reside and evolve ought to have an influence on the morphology of the radio sources. This has been shown in many case studies where the radio structures have been compared with the X-ray gas environments (Blanton et al. (2011); Boehringer et al.(1993)). Studies of the optical environments of radio galaxies have also been carried out previously (Longair & Seldner, 1979; Yee & Green, 1984; Hill & Lilly, 1991; Zirbel, 1997). The motivation behind these studies has been to examine differences between different classes of radio sources, the evolution of environments with cosmic epoch as well as the possibility of identifying clusters/groups of galaxies using radio sources as a tracer(Wing &Blanton, 2011). Many previous studies have found that the environments of FRI/FRII sources are different and are dependent on the cosmic epoch. FRI sources, typically, are found in rich environments. FRII sources in the local universe are generally hosted by field galaxies, but at higher red shifts are found in richer environments(Hatch et al.,2011;Best et al.,2003;Overzier et al.,2008). However, there have been fewer studies that relate the richness of the environments and morphological asymmetries of radio galaxies. Earlier investigations by Subrahmanyan et al. (2008) and Safouris et al.(2009) are noteworthy in this regard where the radio structures of two giant radio galaxies were examined in the context of the large-scale galaxy distributions in their vicinity(also see Chen et al.(2012) and references therein). The study was also used to infer properties of the ambient thermal gas medium in which the structures evolved. Clear correlations between structural asymmetries and associated extended emission-line gas were also found for radio galaxies that have relatively smaller sizes of a few hundred kpc(McCarthy et al., 1991). Thesis Work To progress the field in the problems highlighted above, the following work has been done in this thesis. Radio Imaging of ATLBS Survey To characterize the cosmic evolution of radio sources and their properties, observations and imaging of faint radio sources is essential. The Australia Telescope Low Brightness Survey(ATLBS), which has been used in the studies presented in this thesis, has been designed specifically to image diffuse radio emission to relatively high red shifts(z ∼1−1.5). Therefore obtaining good surface brightness sensitivity was a prime objective in planning the radio observations and in imaging the data obtained from these observations. This requires a nearly complete synthesized aperture and observations of a representative patch of the extra galactic sky. These requirements have been fulfilled in ATLBS survey, which has excellent uv coverage, especially at short spacings, and images a region off the galactic plane that is devoid of strong radio sources. The observations were carried out for two adjacent fields, designated as A and B with their centers at RA:00h 35m 00s,DEC:−67◦00 00 and RA:00h 59m 17s,DEC:−67◦00 00 ,in the 20 cm band, with a center frequency of1388MHz,infullpolarization mode. The radio data was imaged by using techniques such as multi-frequency deconvolution and self-calibration to make two mosaics of region A and B which are free of artefacts. These high-resolution radio images(with beamFWHM of 6 “)of the ATLBS survey regions cover 8.42 square degrees sky area with rms noise 72 µJy beam−1 and are of exceptional quality in that there are no imaging errors or artifacts above the thermal noise over the entire field of view. The images have excellent surface brightness sensitivity and hence provide good representation of extended emission components associated with radio sources. Optical Imaging of ATLBS Survey The ATLBS survey region has been also observed in SDSS r band, specifically for providing information about the galaxies hosting radio sources observed in ATLBS survey as well as galaxies in the neighbourhood of the radio sources. The optical observations were carried out using the CTIO 4 meter Blanco Telescope in Chile and using theMOSAICIIimager,whichisamosaicof8CCDs. In total, 28 optical images were created from the optical data. Each image was formed from a set of 5 dithers, using which spurious sources in the images were rejected. The final images are complete down to a magnitude of 22.75. Radio Source Counts Using the sensitive radio and optical images, a study of radio source counts was carried out. This study made use of some novel strategies and algorithms to generate a source list and correct it for various biases to obtain the radio source counts. More specifically, care was taken to identify sources with low surface brightness by making use of low resolution images for initial identification, and using multiple indicators (including optical images) to identify components of sources. The blending issues inherent in using low resolution images has been avoided using higher resolution images to identify blended sources. Thus, use of low resolution images( beam FWHM =50”′) almost completely removes effects of resolution bias and the use of high resolution images avoids blending issues. These strategies, together with use of optical images to locate candidate galaxy hosts and a careful visual examination of resolved and complex sources instead of automated classification ensures that the ATLBS catalog is a ‘source catalog’ as opposed to a ‘component catalog’. The distinction between ‘sources’ (which are single sources) as opposed to components(parts of a single source appearing separate) is crucial in estimating the true source counts. The source list was used toestimatetheradiosourcecountsdownto0.4 mJy. Comparing the counts with previous work shows that the ATLBS counts are systematically lower and the upturn in sub-mJy source counts has not been found down to the noise limited flux densities probed. The systematically low counts for ATLBS relative to most previous studies are attributed to the ATLBS counts representing sources as opposed to components, as well as corrections for noise bias as well as clustering effects that may affect source counts derived from the small sky coverage typical of deep surveys. This study also demonstrates the substantial difference in counts that result from using component catalogs as opposed to source catalogs: at 1 mJy flux density component counts may be as much as 50% above true source counts. This implies that automated image analysis for counts may be dependent on the ability of the imaging to reproduce connecting emission with low surface brightness as well as the ability of the algorithm to recognize sources, which require that source finding algorithms effectively work with multi-resolution and multi-wavelength data. Galaxy Environments of Extended Radio Sources in ATLBS Survey A study of the galaxy environments of the extended sources in the ATLBS survey was carried out using the optical images. This study of the environments of radio sources from the ATLBS survey is restricted to those that are extended and hence to a subset of the ATLBS-ESS(Extended Source Sample) sources. Briefly, the ATLBS-ESS subsample consists of 119 radio sources that have angular size ex-ceeding0’.5. Applying a red shift cut(to exclude sources with high red shifts whose optical environment may be beyond the depth of the optical images) as well as other constraints(such as availability of optical magnitudes of the host galaxy), a sub-sample of 43 sources was formed, including sources of diverse radio morphologies(FRI/FRII, WATs and HTs)as well as7 radio sources which are highly asymmetric in their radio morphology. For these sources, where no spectroscopic data was available, a red shift estimate was obtained from a magnitude-red shift relation derived from other sources in the ATLBS survey. Using the optical images convolved with a matched filter(following the prescription from Postman et al. (1996))consisting of a radial and magnitude filter, smoothed maps were formed for each source in the sample. These give the likelihood of a cluster being present in a given position in the map (in this case the location of interest being the position of the radio source in the map). Further, five parameters were defined in this study, which give estimates of the angular anisotropy of galaxy density around the axis of the radio source. This method used to quantify environmental asymmetry for the study presented in the thesis is new. The parameters defined thus were used to examine the environments of radio sources in the sample over a wide range in red shift. Specifically a comparison of FRI/FRII environments was made in two different red shift regimes(above and below z = 0.5) and it was found that the FRI and FRII sources inhabit environments of similar richness at low and high red shifts, with no evidence for red shift evolution. The WAT and HT sources were(as expected from earlier studies in literature)found in the most dense environments. Examination of the anisotropy parameters for the asymmetric radio sources clearly showed the influence environment has on radio source morphology, specifically in that the higher density of galaxies was found on the shorter side of the radio sources in almost all cases. Images and Other Resources The radio and optical images are an excellent resource for examining with auto-mated algorithms for source finding, parameter fitting, and morphological classification, and as a resource for testing such algorithms that would be used on upcoming all-sky continuum surveys with the LOFAR and ASKAP/SKA. The techniques and methods developed and presented in the thesis may be used in future studies of radio source populations.

This PhD thesis, written in english, studies the non-Gaussianity (NG) of extragalactic foregrounds to the Cosmic Microwave Background (CMB), the latter being one of the golden observables of today's cosmology. In the last decade has emerged research for deviations of the CMB to the Gaussian law, as they would discriminate the models for the generation of primordial perturbations. However the CMB measurements, e.g. by the Planck satellite, are contaminated by several foregrounds. I studied in particular the extragalactic foregrounds which trace the large scale structure of the universe : radio and infrared point-sources and the thermal Sunyaev-Zel'dovich effect (tSZ). I hence describe the statistical tools to characterise a random field : the correlation functions, and their harmonic counterpart : the polyspectra. In particular the bispectrum is the lowest order indicator of NG, with the highest potential signal to noise ratio (SNR). I describe how it can be estimated on data, accounting for a potential mask (e.g. galactic), and propose a method to visualise the bispectrum, which is more adapted than the already existing ones. I then describe the covariance of a polyspectrum measurement, a method to generate non-Gaussian simulations, and how the statistic of a 3D field projects onto the sphere when integrating along the line-of-sight. I then describe the generation of density perturbations by the standard inflation model and their possible NG, how they yield the CMB anisotropies and grow to form the large scale structure of today's universe. To describe this large scale structure, I present the halo model and propose a diagrammatic method to compute the polyspectra of the galaxy density field and to have a simple and powerful representation of the involved terms. I then describe the foregrounds to the CMB, galactic as well as extragalactic. I briefly describe the physics of the thermal Sunyaev-Zel'dovich effect and how to describe its spatial distribution with the halo model. I then describe the extragalactic point-sources and present a prescription for the NG of clustered sources. For the Cosmic Infrared Background (CIB) I introduce a physical modeling with the halo model and the diagrammatic method. I compute numerically the 3D galaxy bispectrum and produce the first theoretical prediction of the CIB angular bispectrum. I show the contributions of the different terms and the temporal evolution of the galaxy bispectrum. For the CIB angular bispectrum, I show its different terms, its scale and configuration dependence, and how it varies with model parameters. By Fisher analysis, I show it allows very good constraints on these parameters, complementary to or better than those coming from the power spectrum. Finally, I describe my work on measuring NG. I first introduce an estimator for the amplitude of the CIB bispectrum, and show how to combine it with similar ones for radio sources and the CMB, for a joint constraint of the different sources of NG. I quantify the contamination of extragalactic point-sources to the estimation of primordial NG ; for Planck it is negligible for the central CMB frequencies. I then describe my measurement of the CIB bispectrum on Planck data ; it is very significantly detected at 217, 353 and 545 GHz with SNR ranging from 5.8 to 28.7. Its shape is consistent between frequencies, as well as the intrinsic amplitude of NG. Ultimately, I describe my measurement of the tSZ bispectrum, on simulations and on Compton parameter maps estimated by Planck, validating the robustness of the estimation thanks to realist foreground simulations. The tSZ bispectrum is very significantly detected with SNR~200. Its amplitude and its scale and configuration dependence are consistent with the projected map of detected clusters and tSZ simulations. Finally, this measurement allows to put a constraint on the cosmological parameters : sigma_8*(Omega_b/0.049)^0.35 = 0.74+/-0.04 in agreement with other tSZ statistics.

Cette thèse, écrite en anglais, étudie la non-Gaussianité (NG) des avant-plans extragalactiques au fond de rayonnement fossile (FDC), celui-ci étant une des observables de choix de la cosmologie actuelle. Ces dernières années a émergé la recherche de déviations du FDC à la loi Gaussienne, car elles permettraient de discriminer les modèles de génération des perturbations primordiales. Cependant les mesures du FDC, e.g. par le satellite Planck, sont contaminées par différents avant-plans. J'ai étudié en particulier les avant-plans extragalactiques traçant la structure à grande échelle de l'univers: les sources ponctuelles radio et infrarouges et l'effet Sunyaev-Zel'dovich thermique (tSZ). Je décris donc les outils statistiques caractérisant un champ aléatoire : les fonctions de corrélations, et leur analogue harmonique : les polyspectres. En particulier le bispectre est l'indicateur de plus bas ordre de NG avec le plus fort rapport signal sur bruit (SNR) potentiel. Je décris comment il peut être estimé sur des données en tenant compte d'un masque (e.g. galactique), et propose une méthode de visualisation du bispectre plus adaptée que les préexistantes. Je décris ensuite la covariance d'une mesure de polyspectre, une méthode pour générer des simulations non-Gaussiennes, et comment la statistique d'un champ 3D se projette sur la sphère lors de l'intégration sur la ligne de visée. Je décris ensuite la genèse des perturbations de densité par l'inflation standard et leur possible NG, comment elles génèrent les anisotropies du FDC et croissent pour former la structure à grande échelle de l'univers actuel. Pour décrire cette dite structure, j'expose le modèle de halo et propose une méthode diagrammatique pour calculer les polyspectres du champ de densité des galaxies et avoir une représentation simple et puissante des termes impliqués. Puis je décris les avant-plans au FDC, tant galactiques que extragalactiques. J'expose la physique de l'effet tSZ et comment décrire sa distribution spatiale avec le modèle de halo. Puis je décris les sources extragalactiques et présente une prescription pour la NG de sources corrélées. Pour le fond diffus infrarouge (FDI) j'introduis une modélisation physique par le modèle de halo et la méthode diagrammatique. Je calcule numériquement le bispectre 3D des galaxies et obtiens la première prédiction du bispectre angulaire FDI. Je montre les différentes contributions et l'évolution temporelle du bispectre des galaxies. Pour le bispectre du FDI, je montre ses différents termes, sa dépendence en échelle et en configuration, et comment il varie avec les paramètres du modèle. Par analyse de Fisher, je montre qu'il apporte de fortes contraintes sur ces paramètres, complémentaires ou supérieures à celles venant du spectre. Enfin, je décris mon travail de mesure de la NG. J'introduis d'abord un estimateur pour l'amplitude du bispectre FDI, et montre comment le combiner avec de similaires pour les sources radio et le FDC, pour une contrainte jointe des différentes sources de NG. Je quantifie la contamination des sources ponctuelles à l'estimation de NG primordiale ; pour Planck elle est négligeable aux fréquences centrales du FDC. Je décris ensuite ma mesure du bispectre FDI sur les données Planck ; il est détecté très significativement à 217, 353 et 545 GHz, avec des SNR allant de 5.8 à 28.7. Sa forme est cohérente entre les différentes fréquences, de même que l'amplitude intrinsèque de NG. Enfin, je décris ma mesure du bispectre tSZ, sur des simulations et sur les cartes tSZ estimées par Planck, validant la robustesse de l'estimation via des simulations d'avant-plans. Le bispectre tSZ est détecté avec un SNR~200. Son amplitude et sa dépendence en échelle et en configuration sont cohérentes avec la carte des amas détectés et avec les simulations. Enfin, cette mesure place une contrainte sur les paramètres cosmologiques : sigma_8 (Omega_b/0.049)^0.35 = 0.74+/-0.04 en accord avec les autres statistiques tSZ. / This PhD thesis, written in english, studies the non-Gaussianity (NG) of extragalactic foregrounds to the Cosmic Microwave Background (CMB), the latter being one of the golden observables of today's cosmology. In the last decade has emerged research for deviations of the CMB to the Gaussian law, as they would discriminate the models for the generation of primordial perturbations. However the CMB measurements, e.g. by the Planck satellite, are contaminated by several foregrounds. I studied in particular the extragalactic foregrounds which trace the large scale structure of the universe : radio and infrared point-sources and the thermal Sunyaev-Zel'dovich effect (tSZ). I hence describe the statistical tools to characterise a random field : the correlation functions, and their harmonic counterpart : the polyspectra. In particular the bispectrum is the lowest order indicator of NG, with the highest potential signal to noise ratio (SNR). I describe how it can be estimated on data, accounting for a potential mask (e.g. galactic), and propose a method to visualise the bispectrum, which is more adapted than the already existing ones. I then describe the covariance of a polyspectrum measurement, a method to generate non-Gaussian simulations, and how the statistic of a 3D field projects onto the sphere when integrating along the line-of-sight. I then describe the generation of density perturbations by the standard inflation model and their possible NG, how they yield the CMB anisotropies and grow to form the large scale structure of today's universe. To describe this large scale structure, I present the halo model and propose a diagrammatic method to compute the polyspectra of the galaxy density field and to have a simple and powerful representation of the involved terms. I then describe the foregrounds to the CMB, galactic as well as extragalactic. I briefly describe the physics of the thermal Sunyaev-Zel'dovich effect and how to describe its spatial distribution with the halo model. I then describe the extragalactic point-sources and present a prescription for the NG of clustered sources. For the Cosmic Infrared Background (CIB) I introduce a physical modeling with the halo model and the diagrammatic method. I compute numerically the 3D galaxy bispectrum and produce the first theoretical prediction of the CIB angular bispectrum. I show the contributions of the different terms and the temporal evolution of the galaxy bispectrum. For the CIB angular bispectrum, I show its different terms, its scale and configuration dependence, and how it varies with model parameters. By Fisher analysis, I show it allows very good constraints on these parameters, complementary to or better than those coming from the power spectrum. Finally, I describe my work on measuring NG. I first introduce an estimator for the amplitude of the CIB bispectrum, and show how to combine it with similar ones for radio sources and the CMB, for a joint constraint of the different sources of NG. I quantify the contamination of extragalactic point-sources to the estimation of primordial NG ; for Planck it is negligible for the central CMB frequencies. I then describe my measurement of the CIB bispectrum on Planck data ; it is very significantly detected at 217, 353 and 545 GHz with SNR ranging from 5.8 to 28.7. Its shape is consistent between frequencies, as well as the intrinsic amplitude of NG. Ultimately, I describe my measurement of the tSZ bispectrum, on simulations and on Compton parameter maps estimated by Planck, validating the robustness of the estimation thanks to realist foreground simulations. The tSZ bispectrum is very significantly detected with SNR~200. Its amplitude and its scale and configuration dependence are consistent with the projected map of detected clusters and tSZ simulations. Finally, this measurement allows to put a constraint on the cosmological parameters : sigma_8*(Omega_b/0.049)^0.35 = 0.74+/-0.04 in agreement with other tSZ statistics.