Cosmological constraints from galaxy clusters rely on accurate measurements of the mass and internal structure of clusters. An important source of systematic uncertainty in cluster mass and structure measurements is the secure selection of background galaxies that are gravitationally lensed by clusters. This issue has been shown to be particular severe for faint blue galaxies. We therefore explore the selection of faint blue background galaxies, by reference to photometric redshift catalogues derived from the Cosmological Evolution Survey (COSMOS) and our own observations of massive galaxy clusters at z similar or equal to 0.2. We show that methods relying on photometric redshifts of galaxies in/behind clusters based on observations through five filters, and on deep 30-band COSMOS photometric redshifts are both inadequate to identify safely faint blue background galaxies with the same 1 per cent contamination level that we have achieved with red galaxies. This is due to the small number of filters used by the former, and absence of massive galaxy clusters at redshifts of interest in the latter. Nevertheless, our least contaminated blue galaxy sample yields stacked weak-lensing results consistent with our previously published results based on red galaxies, and we show that the stacked clustercentric number density profile of these faint blue galaxies is consistent with expectations from consideration of the lens magnification signal of the clusters. Indeed, the observed number density of blue background galaxies changes by similar to 10-30 per cent across the radial range over which other surveys assume it to be flat.

The dark matter paradigm has been a great source of speculation in both the 20th and 21st Centuries. Since its proposed existence in 1933, the mounting evidence has led to this theoretical particle becoming one of the greatest mysteries of modern physics. However, despite its dominant presence in the Universe, little is known about its nature and how it behaves. In this thesis I critically analyse one particular property of dark matter: the self-coupling. The self-interacting dark matter paradigm hypothesises that dark matter is not collisionless as assumed in most cosmological simulations, and in-fact has some probability that it will scatter off itself. Such a self-coupling will resolve many discrepancies that exist between observations and theory, particularly on small, non-linear scales. Moreover, any detection of a self-interaction cross-section will place considerable limitations on the acceptable particle physics models of dark matter and hence has grown to become an important question. In this thesis I develop and implement a method to constrain the self-interaction cross-section of dark matter that exploits continually accreting and merging groups of galaxies as they fall into galaxy clusters. Utilising the ubiquitous nature of accreting substructure, I measure the offsets between dark matter and baryonic gas as they become separated due to their differing interaction properties. Studying this effect over a sample of events, I will be able to make the first ever statistical estimate of the cross-section of dark matter, while averaging over many different unknown merging scenarios. I begin my thesis by deriving an analytical description of sub-halo in-fall, allowing me to constrain dark matter self-interaction models directly from observations. In this study, I find that current archival data should be able to detect a difference in the dynamical behaviour of dark matter and standard model particles at 6σ, and measure the total interaction cross-section σDM/m with 68% confidence limits of ±1 cm2g-1. Having constructed a new method to derive constraints on the cross-section of dark matter I carry out a study into the potential systematics that may affect a measurement. I determine the accuracy of weak gravitational lensing, which is the distortion of light due to intervening mass, as a tool to estimate the positions of substructure in galaxy clusters. I find that the public Lenstool software can measure the position of individual 1:5 x 1013Mʘ peaks with ~ 0:3" systematic bias, as long as they are at least ~ 30" from the cluster centre. Finally, I develop a pipeline that can analyse a sample of inhomogeneous observations from The Hubble Space Telescope and the Chandra X-ray Observatory. By measuring the positions of dark matter, gas and galaxies for 68 individual merging events, from a total of 28 galaxy clusters, I detect a 7:4σ offset between gas and an unobserved dark mass. I make the first ever measurement of cross-section of dark matter from a sample of clusters finding σDM < 0:50cm2/g [95% CL], the best constraints to date. In addition to this I find that the brightest group galaxy in-fact tends to lead the dark matter halo during merging events. Although evidence for the existence of interacting dark matter, I conclude that the astrophysics of the BCG is complicated, and that this apparent directional bias should be considered in all galaxy cluster analyses. Moreover, I show that this technique is easily extendable for future surveys that have larger samples of galaxy clusters, with constraints of σDM < 0:001cm2/g potentially attainable.

We present the HST WFC3/F275W UV imaging observations of A2218-Flanking, a lensed compact dwarf galaxy at redshift z approximate to 2.5. The stellar mass of A2218-Flanking is log(M-*/M-circle dot) = 9.14(-0.04)(+0.07) and SFR is 12.5(-7.4)(+3.8) M-circle dot yr(-1) after correcting the magnification. This galaxy has a young galaxy age of 127. Myr and a compact galaxy size of r(1/2) = 2.4 kpc. The HST UV imaging observations cover the rest-frame Lyman continuum (LyC) emission (similar to 800 angstrom) from A2218-Flanking. We firmly detect (14s) the LyC emission in A2218-Flanking in the F275W image. Together with the HST F606W images, we find that the absolute escape fraction of LyC is f(abs,esc) > 28%-57% based on the flux density ratio between 1700 and 800 angstrom (f(1700)/f(800)). The morphology of the LyC emission in the F275W images is extended and follows the morphology of the UV continuum morphology in the F606W images, suggesting that the f(800) is not from foreground contaminants. We find that the region with a high star formation rate surface density has a lower f(1700)/f(800) (higher f(800)/f(1700)) ratio than the diffused regions, suggesting that LyC photons are more likely to escape from the region with the intensive star-forming process. We compare the properties of galaxies with and without LyC detections and find that LyC photons are easier to escape in low-mass galaxies.

Le principal objectif de ce travail est d'étudier les propriétés multi-longueurs d'onde d'un échantillon de galaxies pour mieux comprendre leur formation et leur évolution. J'ai utilisé les observations du Herschel en complément de données multi-longueurs d'onde dans le champ. J'ai réalisé une extraction des sources ponctuelles. Des catalogues de sources ont été générés en utilisant la méthode de l'ajustement de PSF. J'ai utilisé des catalogues SPIRE pour rechercher des candidats de galaxies SMG subissant les effets d'une lentille gravitationnelle. J'ai identifié 6 sources dans la liste principale et 55 sources dans une liste supplémentaire. En utilisant des ajustements de SEDs, j'ai ensuite estimé la distribution de redshifts de ces sources, et avons ensuite mené une analyse pour contraindre les propriétés des poussières. J'ai trouvé qu'il est très probable que ces sources soient des systèmes lensés. Mentionnons que parmi nos candidates, j'ai découvert une source rouge unique dont la SED en infrarouge lointain croît et qui semble être une galaxie sub-mm (SMG) fortement lensée à haut redshift. Ce genre de sources est rare et la plupart sont découvertes par hasard. J'ai soumis des propositions photométriques et spectroscopiques pour mesurer le redshift de cette source, en utilisant des observations continues avec IRAM-Nika2, NOEMA et SMA et des observations spectroscopiques avec le télescope de 30m de l'IRAM, NOEMA et GEMINI-nord. Les observations ont permis de mesurer le redshift de la lentille et de la galaxie lensée. Une autre observation sera exécutée avant la fin de l'année. Les résultats sont très prometteurs mais j'ai encore besoin de plus de données. / The main aim of this work is to study the multi-wavelength properties of a sample of galaxies to better understand their formation and evolution. I used the new observations from GALEX and \textit{Herschel} in combination with multi-wavelength data available in the field. I made point source catalogs extracted from the observations of GALEX, and \textit{Herschel} SPIRE and PACS bands using the method of the PSF fitting. Then I used Monte Carlo simulations to quantify the quality of the photometry process and the catalogs.Then I used SPIRE catalogs to search for candidate gravitationally lensed SMGs at high redshift. I identified 6 sources sources in a main list, and 55 sources in a supplementary list. Using SED fitting, performed with CIGALE code, I estimated the redshift distributions of these sources, and constrained their dust properties. I found that, it is very likely that we have gravitationally lensing systems. It should be mentioned that within our candidates we have discovered a unique red source with a rising Far-IR SED, which appears to be a strongly lensed submillimeter galaxies at z~5.2. This kind of sources are quite rare on the sky and are serendipitously discovered. I proposed some followup photometric and spectroscopy observation to measure the redshift of this sources, using for instance continuum observations with IRAM-Nika2, NOEMA, and SMA, and spectroscopic observation with IRAM 30m telescope, NOEMA, and GEMINI-North. The GEMINI-North observations succeeded in measuring the redshifts of the lens and of the lensed galaxy. One more will be executed at the end of this year. The results is quite promising, but more data are needed.

The inflationary ΛCDM cosmology currently provides an accurate description of the Universe. It has been tested using several observational techniques over a wide redshift range, and it provides a good fit to most of them. In addition, it is a surprisingly economical model, requiring only six parameters to characterize the background cosmology and its fluctuations. In this model, the Universe is dominated by a cosmological constant Λ driving an accelerated expansion, and by cold dark matter. The strongest constraints on parameters to date come from observations of the temperature and polarization anisotropies of the cosmic microwave background measured by the Planck satellite. There are, however, indications of features in the Planck power spectra, possible differences with high redshift ground-based CMB experiments, and 'tensions' between Planck and low redshift measurements of the Hubble constant and weak gravitational lensing. In this thesis, we review possible tensions and extensions to the Planck cosmology, at both high and low redshifts. We begin with the high redshift analysis, using the Planck data to test models which introduce oscillatory features in the primordial power spectrum. We also study possible departures from slow roll inflation using the generalized slow-roll formalism, which allows for order unity deviations. Although we find models which give marginal improvements on the temperature or polarization power spectra, the combination of temperature and polarization is found to be consistent with a featureless power-law primordial spectrum. We then focus on measurements of the polarized CMB sky by the South Pole Telescope collaboration, who report tension between their measurements and the ΛCDM cosmology and with the cosmological parameters determined by Planck. We find evidence of a high χ2 in the SPTpol spectra which is unlikely to be cosmological. We report consistency between the Planck and SPTpol polarization spectra over the multipoles accessible to Planck (l ∼< 1500). We then investigate tension at low redshifts. We begin with weak gravitational lensing in which a number of surveys have suggested that the amplitude of the fluctuation spectra is lower than the Planck value. We review the small-angle approximations commonly used in galaxy weak lensing analyses and their effect on cosmological parameters. We find that these approximations are perfectly adequate for present and near future experiments. We find internal inconsistencies in the recent KiDS-450 analysis involving photometric redshifts and the KiDS covariance matrix at large scales. Finally, we investigate the difference between measurements of the present day expansion rate of the Universe. We apply a novel parameterization of the inverse distance ladder to determine the present date value of the Hubble parameter H0, which assumes General Relativity but makes no further assumptions about systematic errors or the nature of dark energy. Our analysis uses baryon acoustic oscillation data and Type Ia Supernovae to constrain the expansion history assuming a value of the sound horizon determined from the CMB. Our results are in tension with recent direct determinations of H0. We conclude that this tension, if real, cannot be solved by modifications of the ΛCDM model at late times. Instead, we would require a modification of the theory at early times which reduces the sound horizon. We conclude that at this time there is no compelling evidence that conflicts with the ΛCDM cosmology either at low or at high redshifts.

Weak gravitational lensing has developed to be one of the most powerful tools for studying the (dark) matter distribution in the Universe. Most weak lensing studies thus far were con- ducted in the optical and near infrared. Measuring weak lensing in the radio though, provided it is feasible, can be very advantageous. One can exploit the well known and deterministic beam pattern of a radio telescope and the polarisation information in radio data to reduce shape biases and intrinsic alignment effects respectively. Combining the information from an optical and a radio survey can also help remove systematics from both datasets. This has motivated this study that uses archival radio and optical data to treat telescope systematics and measure an unbiased weak lensing signal using shape information derived from radio observations. Using simulations I have shown that an unbiased convergence cross power spectrum can be measured in the presence of the large scale (θ > 1◦) systematics detected in FIRST and SDSS. The method however amplifies the uncertainties by a factor ∼2.5 compared to the errors due to cosmic variance and noise due to galaxy intrinsic shape alone. Using the shape information from the two surveys I measure a Ckappakappa spectrum signal that is inconsistent with zero at the 2.7sigma. The placed constraints are consistent with the expected signal in the concordance cosmological model assuming recent estimates of the cosmological parameters from the Planck satellite and literature values for the median redshifts of SDSS and FIRST.Through simulations I also show that I can successfully remove position based small scale systematics (θ5). Using the deconvolved information for the resolved sources I calculate a FWHM median size and flux density of 0.5'' and 300μJy respectively. Comparing the source number density and RMS noise of the study with those of FIRST, I extrapolate to predict that the number density of sources at > 5sigma will be ∼5arcmin-2, assuming the target noise threshold for the survey is reached.

La formation et l'évolution des galaxies reste à ce jour un des mystères de l'Univers observable. Dans le but d'améliorer notre connaissance dans ce domaine, la recherche a utilisé les différentes campagnes d'observation pour caractériser les relations d'échelle des propriétés physiques dans le but de mieux contraindre et comprendre les populations de galaxies aux différentes étapes de leur vie au cours de l'histoire de l'Univers. Depuis les dernières décennies, les études tentent d'étendre ces relations d'échelles dans l'espace des paramètres. C'est dans ce mouvement là que s'inscrit ce projet de thèse. La photométrie des galaxies à haut redshift contient la signature des propriétés physiques comme la masse stellaire, le taux de formation stellaire et l'extinction. Dans cette étude, j’ai réalisé une analyse SED des galaxies amplifiées à z>3 en utilisant les images profondes de Hubble, Bande-K et IRAC des Fontier Fields. Nous avons réalisé la décontamination de ces images en ajustant automatiquement les galaxies avec GALFIT, en utilisant un script Python développé qui prend en compte les niveaux de contamination relatif de toutes les galaxies du champ. Nous avons ensuite ajusté les SEDs décontaminées en utilisant des synthèses de populations stellaires.J’ai appliqué cette méthode pour obtenir les SFR, les SM et les tailles d'un échantillon de 63 galaxies à z>3 détectées dans les champs de A2744 et MACS0416, spectroscopiquement confirmées par MUSE. L'amplification très forte de ces amas nous a permis de collecter un échantillon robuste de galaxies de faibles masses/faibles luminosité, permettant de contraindre les relations d'échelles dans des zones encore in-explorées / Galaxy formation and evolution is one of the most challenging mysteries in the observable Universe. In order to improve our knowledge in this field, the research make use of different observation programs to characterize scaling relations of physical properties, to better constrain and understand galaxy population at different stages of their lives throughout the history of the Universe. Since the past decades, studies are trying to extend those scaling relations in the parameter space. It is in this movement that this thesis project fits. The Spectral Energy Distribution(SED) of high redshift galaxies contains the signature of physical properties such as stellar mass, SFR and extinction. In this work, we perform a SED analysis of magnified galaxies at z>3 using deep Hubble, VLT and Spitzer/IRAC images of the Frontier Fields galaxy clusters. Due to the size of the Kband PSF and specially IRAC PSF and the high density of bright cluster members, it is crucial to deblend Kband and IRAC images to get a reliable SED. We do this by automatically fitting the contaminating galaxies with GALFIT, using a custom Python script which accounts for the relative levels of contamination from each cluster member. We model the decontaminated SED using stellar population models. We apply this method to derive SFR, masses and sizes of a sample of 63 galaxies at z>3 detected in the A2744 and MACS0416 fields, spectroscopically confirmed with MUSE. The very strong amplification of these clusters allow us to collect a robust sample of low-mass galaxies (108 M?), probing the low-luminosity part of scaling relations between stellar mass & size and stellar mass & stellar formation rate

In this work I used a code of galaxy formation (GALICS) to explore three different points that are relevant to the the problem of galaxy formation. The first deals with the method of simulation itself, the second considers the simulation of infrared galaxies within GALICS and the third investigated the role of galaxy formation in measurements of weak gravitational lensing.<br /><br />I examined the predictability of models such as GALICS, using an exploratory test, based on the response of the model to variations in the parameters controlling star formation. I have also proposed a new description, in the astrophysical context, of merger trees.<br /><br />I helped to implement new physical prescriptions into GALICS to bring the model into better agreement with the available observations of infrared galaxies. <br /><br />I wrote a code (LEMOMAF) to simulate the weak lensing effect, taking advantage of the coevolution of galaxies and dark matter included in GALICS.

Die Untersuchung mikrogelinster astronomischer Objekte ermöglicht es, Informationen über die Größe und Struktur dieser Objekte zu erhalten. Im ersten Teil dieser Arbeit werden die Spektren von drei gelinsten Quasare, die mit dem Potsdamer Multi Aperture Spectrophotometer (PMAS) erhalten wurden, auf Anzeichen für Mikrolensing untersucht. In den Spektren des Vierfachquasares HE 0435-1223 und des Doppelquasares HE 0047-1756 konnten Hinweise für Mikrolensing gefunden werden, während der Doppelquasar UM 673 (Q 0142--100) keine Anzeichen für Mikrolensing zeigt. Die Invertierung der Lichtkurve eines Mikrolensing-Kausik-Crossing-Ereignisses ermöglicht es, das eindimensionale Helligkeitsprofil der gelinsten Quelle zu rekonstruieren. Dies wird im zweiten Teil dieser Arbeit untersucht. Die mathematische Beschreibung dieser Aufgabe führt zu einer Volterra'schen Integralgleichung der ersten Art, deren Lösung ein schlecht gestelltes Problem ist. Zu ihrer Lösung wird in dieser Arbeit ein lokales Regularisierungsverfahren angewendet, das an die kausale Strukture der Volterra'schen Gleichung besser angepasst ist als die bisher verwendete Tikhonov-Phillips-Regularisierung. Es zeigt sich, dass mit dieser Methode eine bessere Rekonstruktion kleinerer Strukturen in der Quelle möglich ist. Weiterhin wird die Anwendbarkeit der Regularisierungsmethode auf realistische Lichtkurven mit irregulärem Sampling bzw. größeren Lücken in den Datenpunkten untersucht. / The study of microlensed astronomical objects can reveal information about the size and the structure of these objects. In the first part of this thesis we analyze the spectra of three lensed quasars obtained with the Potsdam Multi Aperture Spectrophotometer (PMAS). The spectra of the quadrupole quasar HE 0435--1223 and the double quasar HE 0047--1756 show evidence for microlensing whereas in the double quasar UM 673 (Q 0142--100) no evidence for microlensing could be found. By inverting the lightcurve of a microlensing caustic crossing event the one dimensional luminosity profile of the lensed source can be reconstructed. This is investigated in the second part of this thesis.The mathematical formulation of this problem leads to a Volterra integral equation of the first kind, whose solution is an ill-posed problem. For the solution we use a local regularization method which is better adapted to the causal structure of the Volterra integral equation compared to the so far used Tikhonov-Phillips regularization. Furthermore we show that this method is more robust on reconstructing small structures in the source profile. We also study the influence of irregular sampled data and gaps in the lightcurve on the result of the inversion.

LAMBDA was an exploration of science and art in two components: a performance-lecture and an art installation. This project asserted that performance and art are effective methods of delivering scientific information to a general audience. The first component was a solo performance-lecture that used lights, projection, costumes, props, and sound in a hybrid classroom space. The performance-lecture covered the topics of light, the Aurora Borealis, gravitational lensing, and the cosmic microwave background. The art installation used light emitting objects to visually represent the gravitational disturbances that exist within a section of space-time. / text