Most of the 500 extra-solar planets detected to date have been discovered either by radial velocity measurements or photometric transits but very few by direct techniques. Direct imaging of exoplanets, however, gives access to a wider variety of information on the planet, from its orbital position to its spectrum, but is a difficult task to achieve because of the small angular separation between the star and its planet and the large flux ratio between them. For these two reasons, direct imaging of exoplanets has up to now been limited to the most favorable cases of bright giant exoplanets orbiting at large distances from their host stars. %The present work aims at developing the high dynamic range capabilities of single- and multi-aperture imaging techniques for the detection and characterization of planetary systems. The first part of this work reports studies of adaptive optics-aided ground-based telescopes and their complementarity with space-based facilities for the detection of extra-solar planets. Results obtained with the Well-Corrected Subaperture at Palomar observatory on narrow multiple systems are then used to illustrate this study. The second part of this work is dedicated to stellar and nulling interferometry. We first present a new data reduction technique for interferometry using the statistical distributions of the noise sources to significantly improve the precision of interferometric measurements. This technique is then applied to the Palomar Fiber Nuller instrument at Palomar observatory to constrain the presence of dust and companions in the innermost regions of Vega's stellar environment and to derive stellar angular diameters with very high accuracies. Finally, we introduce an on-going survey that we are pursuing with the AMBER interferometric instrument at the Very Large Telescope (Cerro Paranal, Chile) aiming at detecting sub-stellar companions around young main sequence stars.

The focus of this work is the development and improvement of chemistry schemes in both 1D and 3D atmosphere models, applied to exoplanets. With an ever increasing number of known exoplanets, planets orbiting stars other than the Sun, the diversity in the physical and chemical nature of planets and their atmospheres is becoming more apparent. One of the prime targets, and the focus of many observational and theoretical studies, are the subclass of exoplanets termed hot Jupiters, Jovian sized planets on very short period orbits around their host star. Due to their close orbit, with orbital periods of just a few days, the atmospheres of such planets are heated to very high temperatures (~1000-2000 K) by the intense irradiation from the star. In addition, it is expected that these planets should have synchronised their rotation with their orbital period, a phenomenon called tidal-locking, that leads to a permanently illuminated dayside and a perpetually dark nightside. This combination of intense heating and tidal-locking leads to an exotic type of atmosphere that is without analogue in our own Solar system. Observational constraints suggest that some of these atmospheres may be clear whilst others may be cloudy or contain haze. Some hot Jupiters appear to be inflated with radii larger than is expected for their mass. For the warmest hot Jupiters optical absorbing species TiO and VO are expected to be present, due to the thermodynamical conditions, where they can strongly influence the thermal structure of the atmosphere, yet so far these species have remained elusive in observations. Theoretical simulations of these planets appear to provide poor matches to the observed emission flux from the nightside of the planet whilst providing a much better agreement with the observed dayside flux. These outstanding questions can be tackled in two complimentary ways. Firstly, the number of exoplanets subject to intense observational scrutiny must be increased to improve the statistical significance of observed trends. Secondly, and in tandem, the suite of available theoretical models applied to such atmospheres must be improved to allow for a more comprehensive understanding of the potential physical and chemical processes that occur in these atmospheres, as well as for better comparison of model predictions with observations. In this thesis we present the development and application of one-dimensional (1D) and three-dimensional (3D) models to the atmospheres of hot exoplanets, with a focus on improving the representation of chemistry. One of the concerns of this work is to couple the radiative transfer and chemistry calculations in a one-dimensional model to allow for a self-consistent model that includes feedback between the chemical composition and the thermal structure. We apply this model to the atmospheres of two typical hot Jupiters to quantify this effect. Implications for previous models that do not include this consistency are discussed. Another major focus is to improve the representation of chemistry in the Met Office Unified Model (UM) for exoplanet applications, a three-dimensional model with its heritage in modelling the Earth atmosphere that has recently been applied to exoplanets. We discuss the coupling of two new chemistry schemes that improve both the flexibility and capabilities of the UM applied to exoplanets. Ultimately these developments will allow for a consistent approach to calculate the 3D chemical composition of the atmosphere taking into account the effect of large scale advection, one of the processes currently hypothesised to cause the discrepancy between model predictions and observations of the nightside emission flux of many hot Jupiters.

Free-floating planets (FFPs) are very difficult to observe directly since they are isolated and intrinsically faint. The gravitational microlensing effect is now major method to observe FFPs, but observing low-mass FFPs is still difficult due to their short duration. We compute simulations for FFP microlensing observations down to Earth-mass using the numerical Besancon Galactic model created by Robin et al. (2012a). These are the first detailed simulation of FFP microlensing using a population synthesis Galactic model incorporating a 3D extinction model, and we also take full account of finite source effects. Firstly, we simulate the microlensing event rate and spatial distribution using three different modes, and for each mode three FFP lens masses (Jupiter, Neptune, and Earth). For the target area of (l, b) =(1, -1.75) which corresponds to the centre of the proposed Euclid ExELS field, our simulations result in 184-920 Jupiter-mass FFPs during the 5 year Euclid mission depending on simulation assumptions. For the Earth-mass FFPs, the rate range is 9-49 FFPs assuming 100% detection efficiency. Next, we compute the rate of parallax detection using a 3D model of the observers. We consider parallax detection by Euclid and WFIRST-AFTA, and by Euclid and LSST. We found that 52 Jupiter-mass FFPs will be detected by a parallax between Euclid and WFIRST-AFTA for two 30-day continuous period around equinoxes if they observe simultaneously. The rate falls to 4 parallax events for Earth-mass FFPs. The parallax detection between Euclid and LSST would be affected by the observation time on the Earth, but it could provide 20 Jupiter-mass FFPs down to 1.4 Earth-mass FFPs.

Depuis la première détection d’une planète extrasolaire autour d’une étoile de type solaire par Mayor et Queloz (1995), plus de 3000 planètes ont été découvertes. La découverte de planètes de type terrestre et la recherche de biomarqueurs dans leur atmosphère sont parmi les principaux objectifs de l'astronomie du XXIeme siècle. Nous nous tournons vers la découverte et la caractérisation des planètes situées dans la zone habitable de leur étoile hôte.La méthode des vitesses radiales (VRs) consiste à mesurer le mouvement réflexe de l'étoile induit par des planètes en orbite. D'autre part, grâce à la photométrie, on peut mesurer la diminution de flux reçu lors du passage d'une planète entre l'étoile ciblée et notre télescope : Il s'agit alors d'un transit. Ces techniques sont complémentaires pour mieux comprendre les systèmes extrasolaires. Cependant, pour atteindre les précisions nécessaires à la détection de Terres ou super-Terres, il est nécessaire de concevoir des instruments très stables, de comprendre les effets systématiques dus à l'atmosphère et tenter de les corriger.La recherche de planètes orbitant autour des étoiles de faibles masses permet d’atteindre dès aujourd'hui des planètes telluriques dans la zone habitable. En effet, en gardant tout autre paramètre égal, le mouvement réflexe (et donc l’amplitude de la variation VR) sera plus grand. De même, un transit sera plus profond si l’étoile centrale est une naine M que pour une étoile de type solaire. De plus, ces étoiles ont une plus faible luminosité que les étoiles de type solaire. Il en résulte que les planètes dans la zone habitable ont des périodes orbitales plus courtes (~50 jours pour les naines M contre ~360 jours pour des étoiles de type solaire).Cette thèse s'inscrit dans une démarche de détections et de caractérisations de planètes en zone habitable de naines M. Pour cela, j'ai observé des naines M avec le spectrographe HARPS, permettant la découverte ou la caractérisation de 24 planètes, qui pourront servir à constituer ou préciser les catalogues de suivi photométriques.En particulier, le projet ExTrA vise à utiliser la photométrie pour détecter des transits en utilisant une nouvelle méthode : la spectrophotométrie différentielle. Elle permet d'améliorer la qualité des courbes de lumière en s'affranchissant d'effets systématiques causés par l'atmosphère. J'introduis l'un d'eux : l'extinction atmosphérique de second ordre, aussi appelé "effet de couleur" et le simule pour la première fois en fonction de divers paramètres d'observations (des conditions atmosphériques aux étoiles ciblés).Je formalise ensuite la technique de spectrophotométrie et simule le gain apporté par la résolution spectrale sur la précision photométrique. Ces simulations prennent en compte les conditions atmosphériques les plus impactantes pour l'effet de couleur (la masse d'air, la quantité de vapeur d'eau) mais également le type d'étoile ciblé (température, gravité, activité) et la résolution spectrale (R<4000).Enfin, il n'existait pas de méthodes numériques spécifiques au projet ExTrA pour traiter les données de spectrophotométrie. Avec l'objectif de corriger les effets systématiques restants dans les courbes de lumière tout en ajustant d'éventuels transits, j'ai développé un nouvel algorithme et j'en expose les premiers résultats. / Since the first detection of an extrasolar planet orbiting a Sun-like star by Mayor and Queloz (1995), more than 3000 have been discovered. Discovering telluric planets and searching for biomarkers in their atmospheres are among the main objectives of the 21st century. Hence, our interest is focused on finding and characterising planets located in the habitable zone of their host star.On one hand, the method known as radial velocities (RV) consists in the measure of the star’s reflex motion induced by orbiting planets. On the other hand, thanks to photometry, we can measure the drop of flux when a planet transits in front of its host star. These techniques are complementary to better understand extrasolar systems. However, in order to reach the precisions necessary to detect an Earth-like planet or a super-Earth, we need very stable instruments as well as the understanding and removal of earth’s atmosphere systematic effects.Searching planets orbiting low mass stars, we already have access to telluric planets in the habitable zone. Indeed, everything else being equal, a lower mass of the host star implies a larger reflex motion, and thus a larger RV amplitude. A transit will be deeper if the central star is a M dwarf compared to a Sun-like star. Moreover, the lower luminosity of M dwarfs implies shorter orbital periods from planets in the habitable zone (~50 days against ~360 days, for M dwarfs compared to solar-type stars, respectively).In this context, this thesis aims to improve the detection and caracterisation of planets in the habitable zone of M dwarfs. I observed some of these stars with the HARPS spectro- graph, leading to the discovery or the caracterisation of 24 planets, which helps us building or precising catalogues of photometric follow-up.In particular, the ExTrA project uses photometry to detect transits using a brand new method : differential spectrophotometry. It improves the light curves quality eliminating sys- tematic effets caused by earth’s atmosphere. I present one these systematics, second order atmospheric extinction also know as color effect, and simulate it for the first time in function of observations parameters (from atmosphere conditions to target and comparison stars).Then, I formalise the technique of spectrophotometry and simulate the gain brought by the addition of spectral resolution to photometric precision. These simulations take in account the atmospheric conditions affecting the color effet (airmass, precipitate water vapor) but also the type of the stars (temperature, gravity, spots), and the spectral resolution (R<4000).Lastly, there were no numerical methods for the treatment of ExTrA’s data at the be- ginning of this work. I developped a new algorithm aiming to correct systematics using the spectral dimension while finding and fitting transits in light curves. I expose the first results obtained from simulations and photometric tests of ExTrA.

Gravitational microlensing occurs when a massive lens (typically a star) deflects light from a more distant source, creating two unresolvable images that are magnified. The effect is transient due to the motions of the lens and source, and the changing magnification gives rise to a characteristic lightcurve. If the lensing object is a binary star or planetary system, more images are created and the lightcurve becomes more complicated. Detection of these lightcurve features allows the lens companion's presence to be inferred. Orbital motion of the binary lens can be detected in some microlensing events, but the expected fraction of events which show orbital motion has not been known previously. We use simulations of orbiting-lens microlensing events to determine the fraction of binary-lens events that are expected to show orbital motion. We also use the simulations to investigate the factors that affect this detectability. Following the discovery of some rapidly-rotating lenses in the simulations, we investigate the conditions necessary to detect lenses that undergo a complete orbit during a microlensing event. We find that such events are detectable and that they should occur at a low but detectable rate. We also derive approximate expressions to estimate the lens parameters, including the period, from the lightcurve. Measurement of the orbital period can in some cases allow the lens mass to be measured. Finally we develop a comprehensive microlensing simulator, MaBμLS, that uses the output of the Besançon Galaxy model to produce synthetic images of Galactic starfields. Microlensing events are added to the images and photometry of their lightcurves simulated. We apply these simulations to a proposed microlensing survey by the Euclid space mission to estimate its planet detection yield.

The direct detection and detailed characterization of exoplanets using extreme adaptive optics (ExAO) is a key science case of both current and future telescopes. However, both quasi-static and residual atmospheric wavefront errors currently limit the sensitivity of this endeavour, generating “speckles” in a coronagraphic image that initially obscure any faint exoplanet(s) from detection. I first demonstrate the current limits of exoplanet imaging using datasets taken with the Gemini Planet Imager and Subaru Coronagraphic ExAO systems. Even when using advanced post-processing algorithms, speckle evolution over time and wavelength is shown to limit the final contrasts that can be reached with current state- of-the-art instruments. A new approach is thus needed to detect fainter exoplanets below these limits. I then illustrate a path forward to reach contrasts near the fundamental photon noise limit: fast focal plane wavefront sensing of both quasi-static and atmospheric speckles. My new method, called the Fast Atmospheric Self-coherent camera Technique (FAST), deploys new hardware and software to overcome these limitations. Looking toward the future, the contrast improvements from fast focal plane wave- front sensing techniques such as FAST are expected to play an essential role in the ground-based detection and characterization of lower mass exoplanets. / Graduate

Dust grain properties were investigated in two very different Galactic environments: the interstellar medium and an exoplanetary system. Two sounding rocket missions were developed to study these regions. Wide-field observations of the Orion OB stellar association were performed in the far-ultraviolet using the Spectrograph for Photometric Imaging with Numeric Reconstruction (SPINR) sounding rocket. These observations reveal the diffuse signature of starlight scattering off interstellar dust grains. The spectral-imaging data were used along with a three-dimensional radiative transfer model to measure the dust scattering parameters: the grain albedo (a) and the scattering asymmetry (g). The measured parameters are consistent with previous measurements made toward Orion. A sharp increase in albedo was measured at 〜1330 A. This feature is not explained by current grain models. The constructed three-dimensional model of Orion includes a two-component dust distribution. The foreground distribution is responsible for the small amount of visible reddening measured toward the bright stars in the Orion constellation.The background distribution represents the Orion Molecular Cloud, which dominates observations of dust emission in the infrared. This model was used to show that backscattered light from the molecular cloud alone cannot produce the observed scattered light distribution. The foreground dust, though optically thin in the visible, significantly contributes to the scattered light in the far-ultraviolet. This suggests that observations of Orion in the infrared and far-ultraviolet may probe entirely different dust populations. The Planetary Imaging Concept Testbed Using a Rocket Experiment (PICTURE) sounding rocket was developed to characterize dust grains in the nearby Epsilon Eridani exoplanetary system. This is a young, dusty system with a Jupiter-massed planet orbiting at 〜3.4 AU (astronomical units). PICTURE sought to capture a direct, visible-light image of dust-scattered starlight in this system with the aid of a high-contrast nulling coronagraph. The design and laboratory testing of the PICTURE science payload is presented. Although the mission returned no science data, several important technological advances were made to enable future direct imaging missions. Most notably, PICTURE demonstrated 5.1 milliarcsecond pointing stability using a fast optical tracking system.

In this work, we investigate the presence of stellar companions around hot Jupiter systems using data sets from the Clio and VISAO instruments on the Magellan Telescope. We observed eighteen targets of which eleven have known spin-orbit obliquity measurements. We detected eleven candidate companions of which five are new discoveries, five involved the validation and confirmation of previous studies, and one candidate proved to be a background star not bound to the transiting planet system. Out of eleven systems with known spin-orbit obliquity, seven systems have candidate companions. Due to the size of the sample, we could not find any correlation between the spin-orbit obliquity and the presence of a stellar companion. As future work, we will do follow up observations on the targets with candidate companions. We will increase our sample to one hundred systems to investigate if there is a correlation between spin-orbit obliquity and the presence of a distant stellar companion.

<p> Planetary migration has become one of the most important processes in planet formation since the first discovery of an exoplanet around 51Peg. A decade after the discovery, the total number of exoplanets has increased to about three hundred. Theoretical work has shown that the disk configuration in which planets are formed strongly affects the subsequent migration of planets within them. Disks evolve and their shape transits from flared to fiat. This is thought to arise because of dust settling. We take this effect into account in our models of planet migration in protoplanetary disks that are heated by the radiation of their central stars. In particular we solve the radiative transfer equation for disks by means of the Monte Carlo method, and then consider planetary migration. We focus on planets around very low mass stars (VLMSs). </p> <p> Our calculations reproduce the disk configurations of Chiang & Goldreich (1997). As dust settles, the superheated and inner layer declines toward the mid-plane. At the same time, dust settling causes the temperature of the upper layer to increase and that of the inner layer to decrease. In order to calculate the migration time accurately, we include the gravity of planets, which causes the density around them to be compressed. This results in shadowing (in front of the planet) and illumination (behind the planet) regions. We included disk evolution by taking into account the effect of dust settling. We found that dust settling itself (without planetary gravity) can reduce the migration time by a factor of 8. When we included the gravity of planets, the effect of dust settling is somewhat washed out. This is because the effect of dust settling on migration acts in a similar way to that of planetary gravity. Thus, when the migration time without dust settling is compared to the case of dust settling (including planetary gravity), dust settling can reduce the migration time by a factor of 2. </p> <p> We also found that the migration time of massive planets(> 5MEB) in such low mass disks, for both cases, is comparable to the disk life time ( rv 107 years). This suggests that planets around VLMS do not plunge into the star within a disk lifetime. This finding is consistent with the discovery of the super-Earth (rv 5.5MEB) at 2.6 AU around M dwarf (Beaulieu et al., 2006). For lower mass planets, the migration time is about two orders of magnitude longer than the disk life time. Thus, the long planetary migration around VLMS does not cause any serious time mismatch problem as in the case of classical T Tauri star system. </p> / Thesis / Master of Science (MSc)

The detection of exoplanets has rapidly evolved to one of the most important frontiers of astronomical and astrophysical research. The recent decades have seen the development of various techniques for detecting exoplanets. Of these approaches the transit method has received particular interest and has lead to the largest number of discoveries to date. The Kepler K2 mission is an ongoing observational survey, which has generated light curves for thousands of stars, a large fraction of which have yet to be fully explored. To discover and characterize the transiting planets hosted by the respective stars, extensive transit screens are required. However, implementing a pipeline for transit analyses is not straight forward, considering the light curve properties of different survey, the rapid changes brought by technological advancements, and the apparent lack of a golden standard with respect to the applied methodology. The project has reviewed several aspects of exoplanet detection via the transit method. Particular focus was placed on the identification of a suitable workflow covering the relevant steps to move from raw light curve files to a final prediction and characterization of transiting planetary candidates. Adhering to the identified strategy, the major part of the project then dealt with the implementation of a pipeline that integrates and executes all the different steps in a streamlined fashion. Of note, primary focus was placed on the actual selection and implementation of methods into an operational pipeline, but due to the given time constraints extensive optimizations of each individual processing step was outside the scope of this project. Nevertheless, the pipeline was employed to predict transit candidates for K2 campaigns C7, C8, C10, C11, and C12. A comparsion of the most conservative predictions from campaigns C7 and C10 with previously reported exoplanet candidates demonstrated that the pipeline was highly capable of discovering reliable transit candidates. Since campaigns C11 and C12 have not yet been fully explored, the respective candidates predicted for those campaigns in the current project might thus harbour novel planetary transit candidates that would be suitable for follow-up confirmation runs. In summary, the current project has produced a pipeline for performing transiting exoplanet searches in K2 data, which integrates the steps from raw light curve processing to transit candidate selection and characterization. The pipeline has been demonstrated to predict credible transit candidates, but future work will have to focus on additional optimizations of individual method parameters and on the analysis of transit detection efficiencies.