Durant les deux dernières décennies, de nombreux efforts ont été apportés pour comprendre le processus de formation des structures de l'Univers jeune. Les avancées dans les technologies observationnelles atteintes aujourd'hui permettent d'observer des galaxies de plus en plus loin, y compris celles responsables de la réionisation cosmique qui a eu lors du premier milliard d'années de l'Univers. L'objectif principal de cette thèse a été de poser des contraintes sur la nature et l'abondance des sources responsables de la réionisation cosmique. Plus précisément, l'étude s'est portée sur les galaxies qui forment des étoiles et qui ont une émission Lyman-alpha (LAE) entre z ~ 3 et 6.7. Il est à noter que cette thèse a été réalisée au sein du consortium MUSE, tout nouvel instrument installé au VLT en janvier 2014 dont nous avons exploité les données du GTO. Ce travail de thèse a permis de confirmer la puissance inégalée de MUSE au niveau de la détection et de l'étude de sources extragalactiques faibles sans aucune présélection. Nous avons observé quatre amas-lentilles dont l'amplification de la lumière permet la détection de sources à faible luminosité, au prix d'une diminution du volume d'Univers observé. Nous nous sommes d'abord focalisés sur l'étude de l'amas de galaxies Abell 1689 afin de structurer une méthodologie applicable aux autres amas. En comparant la densité volumique des LAEs détectés aux différentes fonctions de luminosité (FdL) de la littérature, nous sommes arrivés à la conclusion suivante : la pente de la loi de puissance que décrit la fonction de Schechter pour la partie la plus faible est plus petite que alpha <= -1.5, ce qui signifie que le nombre de LAEs croît extrêment vite vers les faibles luminosités. Nous avons ensuite appliqué cette méthode aux autres amas de notre échantillon observés par MUSE. Les LAEs identifiés et mesurés dans ces amas sont typiquement dix fois plus faibles que ceux observés dans les champs vides (39 < log(Lya) < 42.5). Environ un tiers de ces LAEs n'a pas de contrepartie dans le continuum jusqu'à AB ~ 28 sur les images HST et n'aurait donc jamais été vu sur des relevés pointés. Le catalogue final contient plus de 150 LAEs, ce qui nous a permis d'étudier la contribution des objets les plus faibles, ainsi que l'évolution de la pente en fonction du redshift. Les résultats obtenus semblent confirmer que la pente alpha est proche de -2 pour l'ensemble des LAEs compris entre 2.9 < z < 6.7. On observe aussi une évolution de alpha, qui passe de -1.8 à -1.95 entre z ~ 3-4 et z ~ 5-7, un résultat original et non dépendant des données utilisées pour la partie brillante de la FdL. L'intégration de cette FdL permet ensuite de calculer la densité de photons ionisants émis par ces LAEs et de déterminer leur impact relatif sur la réionisation cosmique. A l'avenir, la profondeur de champ atteinte par les données du James Webb Space Telescope (JWST) va repousser la limite de détection de ces galaxies jusqu'à z ~ 8. L'utilisation de spectrographes dans le proche infrarouge comme MOSFIRE/Keck, KMOS/VLT ou le tout récent EMIR/GTC permettent déjà de confirmer des candidats à z >= 7. Cette thèse a apporté des nouvelles contraintes sur la partie faible de la FdL des LAEs pour un redshift allant jusqu'à z ~ 6, un début donc de ce que l'on va pouvoir faire dans les années à venir pour des redshifts de l'ordre de z ~ 7-8. / Significant efforts have been put for the past two decades to understand the formation process of structure in the early Universe. The recent technological advances in the observational field allow for observing galaxies further and further, even the ones responsible for the cosmic reionization which occurred during the first billion years of the Universe. The main goal of this thesis was to impose constraints on the nature and the abundance of the sources responsible of the cosmic reionization. More specifically, the study was focused on the star-forming galaxies that have a Lyman-alpha emission (LAE) between z ~ 3 and 6.7. This thesis has been conducted within the framework of the MUSE consortium, a brand new instrument installed on the VLT in January 2014, as part of the exploitation of the Guaranteed Time (GTO). This thesis work has enabled us to confirm the unrivalled power of MUSE concerning the detection and the study of weak extragalactic sources without any preselection. We have observed four lensing clusters which magnify the incident light and make it possible to detect faint sources, at the expense of a decrease of the volume of the observed Universe. At first we started with the study of the galaxy cluster Abell 1689 in order to build up a methodology we intend to apply on other galaxy clusters. By comparing the volume density of the detected LAEs to the luminosity functions (LF) coming from the literature, we have reached the following conclusion : the slope of the power law from the Schechter function is smaller than alpha <= -1.5, which means that the number of LAEs increases drastically towards the faint luminoities. Then we have applied the new-build method to the other galaxy clusters of our sample observed with MUSE. The LAEs we have detected and measured in this sample are roughly ten times fainter than the ones observed in blank fields thanks to the lensing effect (39 < log(Lya) < 42.5). About one third of them lacks a counterpart in the continuum up to AB ~ 28 on the HST images and couldn't have been seen on targeted surveys. The final catalog includes more than 150 LAEs, this amount has enabled us to study the contribution of the faintest ones and also the evolution of the slope according to the redshift. The results of this work seem to confirm that the slope alpha is close to -2 for all the 2.9 < z < 6.7 LAEs. Furthermore, one can notice the evolution of alpha from -1.8 to -1.95 between z ~ 3-4 and z ~ 5-7, an original result and irrespective of the data set used to complement the present sample towards the bright region of the LF. The integral of the LF allows for working out the ionizing photons density emitted by these LAEs and for determining their relative impact on the cosmic reionization. In the future, the depth of the James Webb Space Telescope (JWST) observations will improve the limits of galaxy detection, certainly up to z ~ 8. The use of near-IR spectrographs such as MOSFIRE/Keck, KMOS/VLT or the very recent EMIR/GTC already provides the confirmation of z >= 7 candidates. This thesis brought new constraints on the faint-end part of the LF of LAEs for a redshift up to z ~ 6, which represents a beginning with respect to all we can do in the coming years for redshifts up to z ~ 7-8.

Une fraction non négligeable de l'histoire de la formation stellaire a lieu dans des environnements très affectés par la poussière. Il est donc naturel de se demander si on arrive à bien contraindre cette activité de formation d'étoiles. En effet, une part importante de cette activité pourrait être manquée due à la présence de poussière. C'est dans ce contexte que s'inscrit le travail que je vais présenter.Dans la première partie de ma thèse, j'ai eu pour but de déterminer la fraction de galaxies lumineuses formant des étoiles à haut redshift (i.e. 1.5<z<3) sélectionnées à partir des observations MIPS-24μm du champs COSMOS manquées par les critères UV/optique que je détaille ci-après. J'ai appliqué à mon échantillon de galaxies les critères BzK et BM/BX, ainsi que la sélection des «IRAC peakers» et celle des sources optiquement faibles mais brillantes en IR (OFIR, de l'anglais «Optically Faint IR-bright sources»). J'ai ensuite quantifié la contribution de ces différentes sous-populations à la fonction de luminosité à 8μm au repos ainsi qu'à la densité de taux de formation d'étoiles à z~2. Les résultats obtenus soulignent le danger d'utiliser des sélections couleurs de ce type lorsque l'on cherche à quantifier la formation stellaire enfouie dans la poussière. En effet, si le critère BzK offre une identification quasi-complète (~90%) de l'échantillon 24μm, il n'en est pas de même des autres critères. Par exemple, la sélection BM/BX manque 50% des sources considérées et celle des «IRAC peakers» ne sélectionne que 64% de l'échantillon d'étude. Il faut par conséquent être prudent lorsqu'on décide d'utiliser de telles méthodes de sélection qui entraînent nécessairement des extrapolations incertaines, pouvant ainsi fausser notre compréhension de la contribution des galaxies fortement affectées par l'extinction.Dans une seconde partie, je cherche à comprendre la nature composite des sources OFIR les plus brillantes. Cette démarche vise à apporter une compréhension supplémentaire à la connexion AGN/galaxie à flambée de formation stellaire. En se basant sur les données PACS à 100 et 160 μm du satellite herschel, je peux mieux contraindre les distributions spectrales d'énergie de mes sources. Je souhaite déterminer la fraction de la luminosité de ces sources due à la présence d'un AGN ou à la forte activité de formation stellaire dans le but de mieux comprendre le lien entre ces deux phénomènes. Ces sources OFIR brillantes (F_24μm> 1mJy) présentent des couleurs 100/24 et 160/24 plus faibles que les autres sources du champ COSMOS et leur luminosité semble donc provenir majoritairement d'un AGN. Les avancées technologiques et l'exploration des longueurs d'ondes en infra-rouge lointain et en submillimétrique, avec notamment Herschel, SCUBA-2, Alma, JWST, permettront de mieux comprendre la connexion AGN/ flambée de formation stellaire au sein des galaxies jusqu'à des hauts redshifts. / A non-negligible fraction of the star formation across cosmic time occurred within dust-enshrouded environment. One question of the main interest is then do we really know the exact amount of star formation activity. Indeed, this amount could be strongly biased by the effect of dust extinction.This features the context of the work I will discuss here.First of all, I focused my work on determining the number of luminous star-forming galaxies at 1.5<z<3 that are potentially missed by traditional UV/optical selections. I based my work on 24μm sources brighter than 80μJy in the COSMOS field. I applied to this mid-IR selected sample the BzK and BM/BX criteria, as well as the selections of the "IRAC Peakers" and the "Optically Faint-IR bright" galaxies (OFIR). I also quantified the contribution of these sub-populations to the IR luminosity and cosmic star formation density at z~2. I found that the BzK criterion offers an almost complete (~90%) identification of the 24μm sources at 1.4<z<2.5. In contrast, the BM/BX criterion misses 50% of the MIPS sources and the "IRAC Peakers" one only selects 64% of my sample. Color selections of distant star-forming galaxies must be indeed used with a lot of caution given the substantial bias they can suffer. In particular, the effect of dust extinction strongly affects the completeness of identifications at the bright end of the bolometric luminosity function, which could lead to large and uncertain extrapolations in order to account for the contribution of dusty galaxies missed by these selections.In a second time, I was interested in the composite nature of ultra-luminous infra-red galaxies presenting extreme optical/mid-IR colors at z∼2. I here try to better understand the Starburst/AGN connection in the brightest sources of my OFIR sample. Using PACS 100 and 160 μm from the Herschel Telescope, I have better constraints on the spectral energy distributions of the sources. The goal here is to determine the fraction of the IR luminosity due to the AGN and the fraction due to a strong star-forming activity. Theses really bright (F_24μm> 1mJy) OFIR sources present fainter 100/24 and 160/24 colors than the rest of the 24μm-selected sources. Their luminosity might then come from a strong AGN activity. The forthcoming facilities that will operate at long wavelengths (e.g., JWST, AKMA, SCUBA-2, etc.) will allow a better understanding of the link between the AGN activity and the star-forming one, up to high redshifts.

We use cosmological hydrodynamic simulations to study the impact of out-flows and radiative feedback on high-redshift galaxies. For outflows, we consider simulations that assume (i) no winds, (ii) a .constant-wind. model in which the mass-loading factor and outflow speed are constant, and (iii) "momentum driven" winds in which both parameters vary smoothly with mass. In order to treat radiative feedback, we develop a moment-based radiative transfer technique that operates in both post-processing and coupled radiative hydrodynamic modes. We first ask how outflows impact the broadband spectral energy distributions (SEDs) of six observed reionization-epoch galaxies. Simulations reproduce five regardless of the outflow prescription, while the sixth suggests an unusually bursty star formation history. We conclude that (i) simulations broadly account for available constraints on reionization-epoch galaxies, (ii) individual SEDs do not constrain outflows, and (iii) SED comparisons efficiently isolate objects that challenge simulations. We next study how outflows impact the galaxy mass metallicity relation (MZR). Momentum-driven outflows uniquely reproduce observations at z = 2. In this scenario, galaxies obey two equilibria: (i) The rate at which a galaxy processes gas into stars and outflows tracks its inflow rate; and (ii) The gas enrichment rate owing to star formation balances the dilution rate owing to inflows. Combining these conditions indicates that the MZR is dominated by the (instantaneous) variation of outflows with mass, with more-massive galaxies driving less gas into outflows per unit stellar mass formed. Turning to radiative feedback, we use post-processing simulations to study the topology of reionization. Reionization begins in overdensities and then .leaks. directly into voids, with filaments reionizing last owing to their high density and low emissivity. This result conflicts with previous findings that voids ionize last. We argue that it owes to the uniqely-biased emissivity field produced by our star formation prescriptions, which have previously been shown to reproduce numerous post-reionization constraints. Finally, preliminary results from coupled radiative hydrodynamic simulations indicate that reionization suppresses the star formation rate density by at most 10.20% by z = 5. This is much less than previous estimates, which we attribute to our unique reionization topology although confirmation will have to await more detailed modeling.

We present multi-wavelength detections of nine candidate gravitationally lensed dusty starforming galaxies (DSFGs) selected at 218 GHz (1.4 mm) from the Atacama Cosmology Telescope (ACT) equatorial survey. Among the brightest ACT sources, these represent the subset of the total ACT sample lying in Herschel SPIRE fields, and all nine of the 218 GHz detections were found to have bright Herschel counterparts. By fitting their spectral energy distributions (SEDs) with a modified blackbody model with power-law temperature distribution, we find the sample has a median redshift of z = 4.1(-1.0)(+1.1) (68 per cent confidence interval), as expected for 218 GHz selection, and an apparent total infrared luminosity of log10(mu LIR/L-circle dot) = 13.86(-0.30)(+0.33), which suggests that they are either strongly lensed sources or unresolved collections of unlensed DSFGs. The effective apparent diameter of the sample is root mu d = 4.2(-1.0)(+1.7) kpc, further evidence of strong lensing or multiplicity, since the typical diameter of DSFGs is 1.0-2.5 kpc. We emphasize that the effective apparent diameter derives from SED modelling without the assumption of optically thin dust (as opposed to image morphology). We find that the sources have substantial optical depth (tau = 4.2(-1.9)(+3.7)) to dust around the peak in the modified blackbody spectrum (lambda(obs) <= 500 mu m), a result that is robust to model choice.

We directly detect dust emission in an optically detected, multiply imaged galaxy lensed by the Frontier Fields cluster MACSJ0717.5+3745. We detect two images of the same galaxy at 1.1 mm with the AzTEC camera on the Large Millimeter Telescope leaving no ambiguity in the counterpart identification. This galaxy, MACS0717_Az9, is at z > 4 and the strong lensing model (mu = 7.5) allows us to calculate an intrinsic IR luminosity of 9.7 x 10(10) L-circle dot and an obscured star formation rate of 14.6 +/- 4.5 M-circle dot yr(-1). The unobscured star formation rate from the UV is only 4.1 +/- 0.3 M-circle dot yr(-1), which means the total star formation rate (18.7 +/- 4.5 M-circle dot yr(-1)) is dominated (75%-80%) by the obscured component. With an intrinsic stellar mass of only 6.9 x 10(9) M circle dot, MACS0717_Az9 is one of only a handful of z. >. 4 galaxies at these lower masses that is detected in dust emission. This galaxy lies close to the estimated star formation sequence at this epoch. However, it does not lie on the dust obscuration relation (IRX-beta) for local starburst galaxies and is instead consistent with the Small Magellanic Cloud attenuation law. This remarkable lower mass galaxy, showing signs of both low metallicity and high dust content, may challenge our picture of dust production in the early universe.

Very few low-ionization broad absorption line (LoBAL) QSOs have been found at high redshifts, to date. One high-redshift LoBAL QSO, J0122+1216, was recently discovered by the Lijiang 2.4 m Telescope, with an initial redshift determination of 4.76. Aiming to investigate its physical properties, we carried out follow-up observations in the optical and near-IR spectroscopy. Near-IR spectra from UKIRT and P200 confirm that it is a LoBAL, with a new redshift determination of 4.82 +/- 0.01 based on the Mg II emission-line. The new Mg II redshift determination reveals strong blueshifts and asymmetry of the high-ionization emission lines. We estimate a black hole mass of similar to 2.3 x 10(9) M-circle dot and Eddington ratio of similar to 1.0 according to the empirical Mg II-based single-epoch relation and bolometric correction factor. It is possible that strong outflows are the result of an extreme quasar environment driven by the high Eddington ratio. A lower limit on the outflowing kinetic power (>0.9% L-Edd) is derived from both emission and absorption lines, indicating that these outflows play a significant role in the feedback process that regulates the growth of its black hole, as well as host galaxy evolution.

We present the results of the first, deep Atacama Large Millimeter Array ( ALMA) imaging covering the full similar or equal to 4.5 arcmin(2) of the Hubble Ultra Deep Field ( HUDF) imaged with Wide Field Camera 3/IR on HST. Using a 45-pointing mosaic, we have obtained a homogeneous 1.3-mm image reaching sigma 1.3 similar or equal to 35 mu Jy, at a resolution of similar or equal to 0.7 arcsec. From an initial list of similar or equal to 50 > 3.5 sigma peaks, a rigorous analysis confirms 16 sources with S-1.3 > 120 mu Jy. All of these have secure galaxy counterparts with robust redshifts (< z > = 2.15). Due to the unparalleled supporting data, the physical properties of the ALMA sources are well constrained, including their stellar masses ( M-*) and UV+FIR star formation rates ( SFR). Our results show that stellar mass is the best predictor of SFR in the high-redshift Universe; indeed at z = 2 our ALMA sample contains seven of the nine galaxies in the HUDF withM(*) = 2 x 10(10)M circle dot, and we detect only one galaxy at z > 3.5, reflecting the rapid drop-off of high-mass galaxies with increasing redshift. The detections, coupled with stacking, allow us to probe the redshift/mass distribution of the 1.3-mm background down to S1.3 similar or equal to 10 mu Jy. We find strong evidence for a steep star-forming `main sequence' at z similar or equal to 2, with SFR. M* and a mean specific SFR similar or equal to 2.2 Gyr(-1). Moreover, we find that similar or equal to 85 per cent of total star formation at z similar or equal to 2 is enshrouded in dust, with similar or equal to 65 per cent of all star formation at this epoch occurring in high-mass galaxies ( M-* > 2 x 10(10)M circle dot), for which the average obscured: unobscured SF ratio is similar or equal to 200. Finally, we revisit the cosmic evolution of SFR density; we find this peaks at z similar or equal to 2.5, and that the star-forming Universe transits from primarily unobscured to primarily obscured at z similar or equal to 4.

We report the detection of diffuse Ly alpha emission, or Lya halos (LAHs), around star-forming galaxies at z approximate to 3.78 and 2.66 in the NOAO Deep Wide-Field Survey Bootes field. Our samples consist of a total of similar to 1400 galaxies, within two separate regions containing spectroscopically confirmed galaxy overdensities. They provide a unique opportunity to investigate how the LAH characteristics vary with host galaxy large-scale environment and physical properties. We stack Ly alpha images of different samples defined by these properties and measure their median LAH sizes by decomposing the stacked Ly alpha radial profile into a compact galaxy-like and an extended halo-like component. We find that the exponential scale-length of LAHs depends on UV continuum and Ly alpha luminosities, but not on Ly alpha equivalent widths or galaxy overdensity parameters. The full samples, which are dominated by low UV-continuum luminosity Lya emitters (M-UV greater than or similar to -21), exhibit LAH sizes of 5-6 kpc. However, the most UV- or Ly alpha-luminous galaxies have more extended halos with scale-lengths of 7-9 kpc. The stacked Ly alpha radial profiles decline more steeply than recent theoretical predictions that include the contributions from gravitational cooling of infalling gas and from low-level star formation in satellites. However, the LAH extent matches what one would expect for photons produced in the galaxy and then resonantly scattered by gas in an outflowing envelope. The observed trends of LAH sizes with host galaxy properties suggest that the physical conditions of the circumgalactic medium (covering fraction, H I column density, and outflow velocity) change with halo mass and/or star formation rates.

We report new IRAM/PdBI, JCMT/SCUBA-2, and VLA observations of the ultraluminous quasar SDSS J010013.02+280225.8 (hereafter, J0100+2802) at z =. 6.3, which hosts the most massive supermassive black hole (SMBH), 1.24 x 10(10) M circle dot, that is known at z > 6. We detect the [C II] 158 mu m fine structure line and molecular CO(6-5) line and continuum emission at 353, 260, and 3 GHz from this quasar. The CO(2-1) line and the underlying continuum at 32 GHz are also marginally detected. The [C II] and CO detections suggest active star formation and highly excited molecular gas in the quasar host galaxy. The redshift determined with the [C II] and CO lines shows a velocity offset of similar to 1000 km s(-1) from that measured with the quasar Mg II line. The CO (2-1) line luminosity provides a direct constraint on the molecular gas mass, which is about (1.0 +/- 0.3) x 10(10) M circle dot We estimate the FIR luminosity to be (3.5 +/- 0.7) x 10(12) L circle dot, and the UV-to-FIR spectral energy distribution of J0100 +2802 is consistent with the templates of the local optically luminous quasars. The derived [C II]-to-FIR luminosity ratio of J0100+2802 is 0.0010 +/- 0.0002, which is slightly higher than the values of the most FIR luminous quasars at z similar to 6. We investigate the constraint on the host galaxy dynamical mass of J0100 vertical bar 2802 based on the [C II] line spectrum. It is likely that this ultraluminous quasar lies above the local SMBH-galaxy mass relationship, unless we are viewing the system at a small inclination angle.

We present Atacama Large Millimeter Array [C-I](1 - 0) (rest frequency 492 GHz) observations for a sample of 13 strongly lensed dusty star-forming galaxies (DSFGs) originally discovered at 1.4 mm in a blank-field survey by the South Pole Telescope (SPT). We compare these new data with available [C-I] observations from the literature, allowing a study of the interstellar medium (ISM) properties of similar to 30 extreme DSFGs spanning a redshift range 2 < z < 5. Using the [C-I] line as a tracer of the molecular ISM, we find a mean molecular gas mass for SPT-DSFGs of 6.6 x 10(10) M-circle dot. This is in tension with gas masses derived via low-J (CO)-C-12 and dust masses; bringing the estimates into accordance requires either (a) an elevated CO-to-H-2 conversion factor for our sample of alpha(CO) similar to 2.5 and a gas-to-dust ratio similar to 200, or (b) an high carbon abundance X-CI similar to 7 x 10(-5). Using observations of a range of additional atomic and molecular lines (including [C-I], [C-II] and multiple transitions of CO), we use a modern photodissociation region code (3(D)-PDR) to assess the physical conditions (including the density, UV radiation field strength and gas temperature) within the ISM of the DSFGs in our sample. We find that the ISM within our DSFGs is characterized by dense gas permeated by strong UV fields. We note that previous efforts to characterize photodissociation region regions in DSFGs may have significantly under-estimated the density of the ISM. Combined, our analysis suggests that the ISM of extreme dusty starbursts at high redshift consists of dense, carbon- rich gas not directly comparable to the ISM of starbursts in the local Universe.