The transit method is considered to be one of the most promising for discovering extrasolar planets. However, the method requires photometric precision of better than ∼ 1%. If we are able to achieve this kind of accuracy, then we are set to discover extrasolar planets. The uniqueness of my experiment will lead to the discovery of transiting planets around the brightest and most important stars quicker than the competitors in the field. The importance of the transit method stems from being able to supply many more planetary parameters than other methods, which plays a crucial role in testing planet formation theories. This thesis is divided into eight chapters. The first chapter provides a general background about transits and their theory. We discuss other methods of extrasolar planet detection, recent developments, future space missions, and what we have learned so far about properties of hot Jupiters. The second chapter details the theory of signals and noise on CCDs followed by the design of the PASS0 experiment. The third chapter reports on the difference imaging data pipeline that we developed and applied to a set of PASS0 data to search for transiting planets. The fourth chapter shows how we apply the PASS0 pipeline to SuperWASP data and improve on the accuracy obtained with their aperture photometry pipeline. The fifth chapter reports on the search for variable stars from the PASS0 and SuperWASP data sets that we consider in this thesis. In the sixth chapter we perform a transit search on the PASS0 and SuperWASP data sets and report the results. In the seventh chapter we use the PASS0 pipeline to process a full season of observing data from 2007 for two recent planet discoveries, WASP-7b and WASP-8b, that have not yet been announced. We analyse their lightcurves and predict their radii. Finally we conclude in the eighth chapter.

This first part of this thesis concerns the discovery and parameter determination of three hot Jupiter planets, first detected with by the SuperWASP collaboration, and their planetary nature is confirmed with the modelling of radial velocity measurements and further ground-based transit lightcurves. WASP-92b, WASP-93b and WASP-118b are all hot Jupiters with short orbital periods – 2.17, 2.73 and 4.05 days respectively. The analysis in this thesis finds WASP-92b to have R[sub]p = 1.461 ± 0.077 R[sub]J and M[sub]p = 0.805 ± 0.068 M[sub]J; WASP-93b to have R[sub]p = 1.597 ± 0.077 R[sub]J and M[sub]p = 1.47 ± 0.029 M[sub]J, and WASP-118b to have R[sub]p = 1.440 ± 0.036 R[sub]J and M[sub]p = 0.514 ± 0.020 M[sub]J. The second part of this thesis presents three novel approaches to modelling the effect of stellar limb darkening when fitting exoplanet transit lightcurves. The first method trains a Gaussian Process to interpolate between pre-calculated limb darkening coefficients for the non-linear limb darkening law. The method uses existing knowledge of the stellar atmosphere parameters as the constraints of the determined limb darkening coefficients for the host star of the transiting exoplanet system. The second method deploys an artificial neural network to model limb darkening without the requirement of a parametric approximation of the form of the limb profile. The neural network is trained for a specific bandpass directly from the outputs of stellar atmosphere models, allowing predictions to be made for the stellar intensity at a given position on the stellar surface for values of the T[sub]eff , log g and [Fe/H]. The efficacy of the method is demonstrated by accurately fitting a transit lightcurve for the transit of Venus, and for a single transit lightcurve of TRES-2b. The final limb darkening modelling method proposes an adjustment to the neural network model to account for the fact that the stellar radius is not constant across wavelengths. The method also allows the full variation in light at the edge of the star to be modelled by not assuming a sharp boundary at the limb.

To date, more than 300 planets orbiting stars other than our sun have been discovered using a range of observing techniques, with new discoveries occuring monthly. The work in this thesis focused on the detection of exoplanets using the transit method. Planets orbiting close to their host stars have a roughly 10 per cent chance of eclipsing (transiting) the star, with Jupiter?sized planets causing a one per cent dip in the flux of the star over a few hours. A wealth of orbital and physical information on the system can be extracted from these systems, including the planet density which is essential in constraining models of planetary formation. To detect these types of planets requires monitoring tens of thousands of stars over a period of months. To accomplish this, we conduct a wide-field survey using the 0.5-meter Automated Patrol Telescope (APT) at Siding Spring Observatory (SSO) in NSW, Australia. Once candidates were selected from the data?set, selection criteria were applied to separate the likely planet candidates from the false?positives. For this thesis, the methods and instrumentation used in attaining data and selecting planet candidates are discussed, as well as the results and analysis of the planet candidates selected from star fields observed from 2004?2007. Of the 65 planet candidates initially selected from the 25 target fields observed, only two were consistent with a planet transit. These candidates were later determined to be eclipsing binary stars based on follow up observations using the 40-inch telescope, 2.3-m telescope, and the 3.9-m Anglo-Australian Telescope, all located at SSO. Additionally, two planet candidates from the SuperWASP-North consortium were observed on the 40-inch telescope. Both proved to be eclipsing binary stars. While no planets were found, our search methods and results are consistent with successful transit surveys targeting similar fields with stars in a similar magnitude range and using similar methods.

To date, more than 300 planets orbiting stars other than our sun have been discovered using a range of observing techniques, with new discoveries occuring monthly. The work in this thesis focused on the detection of exoplanets using the transit method. Planets orbiting close to their host stars have a roughly 10 per cent chance of eclipsing (transiting) the star, with Jupiter?sized planets causing a one per cent dip in the flux of the star over a few hours. A wealth of orbital and physical information on the system can be extracted from these systems, including the planet density which is essential in constraining models of planetary formation. To detect these types of planets requires monitoring tens of thousands of stars over a period of months. To accomplish this, we conduct a wide-field survey using the 0.5-meter Automated Patrol Telescope (APT) at Siding Spring Observatory (SSO) in NSW, Australia. Once candidates were selected from the data?set, selection criteria were applied to separate the likely planet candidates from the false?positives. For this thesis, the methods and instrumentation used in attaining data and selecting planet candidates are discussed, as well as the results and analysis of the planet candidates selected from star fields observed from 2004?2007. Of the 65 planet candidates initially selected from the 25 target fields observed, only two were consistent with a planet transit. These candidates were later determined to be eclipsing binary stars based on follow up observations using the 40-inch telescope, 2.3-m telescope, and the 3.9-m Anglo-Australian Telescope, all located at SSO. Additionally, two planet candidates from the SuperWASP-North consortium were observed on the 40-inch telescope. Both proved to be eclipsing binary stars. While no planets were found, our search methods and results are consistent with successful transit surveys targeting similar fields with stars in a similar magnitude range and using similar methods.

We have attempted adaptive optics (AO) imaging of planets around nearby stars in the L' and M bands, using the Clio instrument on the MMT. The MMT AO system, with its deformable secondary mirror, offers uniquely low background AO-corrected images in these bands. This allowed us to explore a wavelength regime that has not been well utilized in searches for extrasolar planets, but offers some advantages over the more commonly used shorter-wavelength H band regime. We have taken deep L' and M band images of the interesting debris disk stars Vega and ϵ Eri. Our observations of ϵ Eri attain better sensitivity to low mass planets within 3 arcseconds of the star than any other AO observations to date. At 1.7 arcsec, the maximum separation of the known planet ϵ Eri b, our M band sensitivity corresponds to objects only 9-16 times brighter than the predicted brightness of this planet. M is by far the most promising band for directly imaging this planet for the first time, though Clio would require a multi-night integration. We have carried out a survey of 50 nearby stars, using mostly the L' band. The survey objective was to determine whether power law fits to the statistics of planet mass m and orbital semimajor axis a from radial velocity (RV) surveys apply when extrapolated to orbital radii beyond the outer limits of RV sensitivity. Given dN/dm ~ m^{-1.44}, our survey null result rules out dN/da ~ a^{-0.2} extending beyond 155 AU, or dN/da constant extending beyond 70 AU, at the 95% confidence level. We have not placed as tight constraints on the planet distributions as the best H band surveys. However, we have probed older planet populations and by using a different wavelength regime have helped diversify results against model uncertainties. We have developed careful and well-tested observing, image processing, sensitivity analysis, and source detection methods, and helped advance L' and M band AO astronomy. These wavelengths will become increasingly important with the advent of new giant telescopes sensitive to interesting, low-temperature planets with red H-L' and H-M colors.

ISIS is a program that specializes in detecting variable stars within clusters, both open and globular. ISIS has historically been used in surveys searching for variable stars. However, recent work has identified new uses for ISIS, such as the detection of exoplanet transits in clusters of stars. This thesis will evaluate ISIS, both how the program functions and for which objects it is most efficient and appropriate. I examined five clusters: three open, and two globular, for one night surveys. With this survey, I identified twenty-five variable stars. Twenty-two variables were previously known, and three variable stars were previously undiscovered. In total, from my short one night surveys, I confirmed twenty-five previously known variable stars and four unknown variable stars. Furthermore, ISIS has detected several δ Scuti stars whose amplitude of variation is on par with an exoplanet transit. These data suggest that ISIS can be used for the detection of exoplanet transits. / Department of Physics and Astronomy

The study of extrasolar planets, or exoplanets for short, has developed rapidly over the last decade. While we have spent much effort building both ground-based and space telescopes to search for exoplanets, it is even more important that we use the observational data wisely to understand them. Exoplanets are of great interest to both astronomers and the general public because they have shown varieties of characteristics that we couldn't have anticipated from planets within our Solar System. To properly analyze the exoplanet populations, we need the tools of statistics. Therefore, in Chapter 1, I describe the science background as well as the statistical methods which will be applied in this thesis. In Chapter 2, I discuss how to train a hierarchical Bayesian model in detail to fit the relationship between masses and radii of exoplanets and categorize exoplanets based on that. A natural application that comes with the model is to use it for future observations of mass/radius and predict the other measurement. Thus I will show two application cases in Chapter 3. Composition of an exoplanet is also very much constrained by its mass and radius. I will show an easy way to constrain the composition of exoplanets in Chapter 4 and discuss how more complicated methods can be applied in future works. Of even greater interest is whether there is life elsewhere in the Universe. Although the future discovery of extraterrestrial life might be totally a fluke, a clear sketched plan always gives us some directions. Research in this area is still very preliminary. Fortunately, besides directly searching for extraterrestrial life, we can also apply statistical reasoning to first estimate the rate of abiogenesis, which will give us some clue on the question of whether there is extraterrestrial life in a probabilistic way. In Chapter 5, I will discuss how different methods can constrain the abiogenesis rate in an informatics perspective. Finally I will give a brief summary in Chapter 6.

No description available.

We analyse the properties of extrasolar planets, other close companions and their hosts. We start by identifying a sample of the detected extrasolar planets that is minimally affected by the selection effects of the Doppler detection method. With a simple analysis we quantify trends in the surface density of this sample in the Msini-period plane. A modest extrapolation of these trends puts Jupiter in the most densely occupied region of this parameter space, thus suggesting that Jupiter is a typical massive planet rather than an outlier. We then examine what fraction of Sun-like (~ FGK) stars have planets. We find that at least ~25% of stars possess planets when we limit our analysis to stars that have been monitored the longest and whose low surface activity allow the most precise radial velocity measurements. The true fraction of stars with planets may be as large as ~100%. We construct a sample of nearby Sun-like stars with close companions (period < 5 years). By using the same sample to extract the relative numbers of stellar, brown dwarf and planetary companions, we verify the existence of a very dry brown dwarf desert and describe it quantitatively. Approximately 16% of Sun-like stars have close companions more massive than Jupiter: 11% +- 3% are stellar, <1% are brown dwarf and 5% +- 2% are giant planets. A comparison with the initial mass function of individual stars and free-floating brown dwarfs, suggests either a different spectrum of gravitational fragmentation in the formation environment or post-formation migratory processes disinclined to leave brown dwarfs in close orbits. Finally we examine the relationship between the frequency of close companions and the metallicity of their Sun-like hosts. We confirm and quantify a ~4 sigma positive correlation between host metallicity and planetary companions. In contrast we find a ~2 sigma anti-correlation between host metallicity and the presence of a stellar companion. Upon dividing our sample into FG and K sub-samples, we find a negligible anti-correlation in the FG sub-sample and a ~3 sigma anti-correlation in the K sub-sample. A kinematic analysis suggests that this anti-correlation is produced by a combination of low-metallicity, high-binarity thick disk stars and higher-metallicity, lower-binarity thin disk stars.

When Galactic microlensing events of stars are observed, one usually measures a symmetric light curve corresponding to a single lens, or an asymmetric light curve, often with caustic crossings, in the case of a binary lens system. In principle, the fraction of binary stars at a certain separation range can be estimated based on the number of measured microlensing events. However, a binary system may produce a light curve which can be fitted well as a single lens light curve, in particullary if the data sampling is poor and the errorbars are large. We investigate what fraction of microlensing events produced by binary stars for different separations may be well fitted by and hence misinterpreted as single lens events for various observational conditions. We find that this fraction strongly depends on the separation of the binary components, reaching its minimum at between 0.6 and 1.0 Einstein radius, where it is still of the order of 5% The Einstein radius is corresponding to few A.U. for typical Galactic microlensing scenarios. The rate for misinterpretation is higher for short microlensing events lasting up to few months and events with smaller maximum amplification. For fixed separation it increases for binaries with more extreme mass ratios.<br><br> Problem of degeneracy in photometric light curve solution between binary lens and binary source microlensing events was studied on simulated data, and data observed by the PLANET collaboration. The fitting code BISCO using the PIKAIA genetic algorithm optimizing routine was written for optimizing binary-source microlensing light curves observed at different sites, in I, R and V photometric bands. Tests on simulated microlensing light curves show that BISCO is successful in finding the solution to a binary-source event in a very wide parameter space. Flux ratio method is suggested in this work for breaking degeneracy between binary-lens and binary-source photometric light curves. Models show that only a few additional data points in photometric V band, together with a full light curve in I band, will enable breaking the degeneracy. Very good data quality and dense data sampling, combined with accurate binary lens and binary source modeling, yielded the discovery of the lowest-mass planet discovered outside of the Solar System so far, OGLE-2005-BLG-390Lb, having only 5.5 Earth masses. This was the first observed microlensing event in which the degeneracy between a planetary binary-lens and an extreme flux ratio binary-source model has been successfully broken. For events OGLE-2003-BLG-222 and OGLE-2004-BLG-347, the degeneracy was encountered despite of very dense data sampling. From light curve modeling and stellar evolution theory, there was a slight preference to explain OGLE-2003-BLG-222 as a binary source event, and OGLE-2004-BLG-347 as a binary lens event. However, without spectra, this degeneracy cannot be fully broken.<br><br> No planet was found so far around a white dwarf, though it is believed that Jovian planets should survive the late stages of stellar evolution, and that white dwarfs will retain planetary systems in wide orbits. We want to perform high precision astrometric observations of nearby white dwarfs in wide binary systems with red dwarfs in order to find planets around white dwarfs. We selected a sample of observing targets (WD-RD binary systems, not published yet), which can possibly have planets around the WD component, and modeled synthetic astrometric orbits which can be observed for these targets using existing and future astrometric facilities. Modeling was performed for the astrometric accuracy of 0.01, 0.1, and 1.0 mas, separation between WD and planet of 3 and 5 A.U., binary system separation of 30 A.U., planet masses of 10 Earth masses, 1 and 10 Jupiter masses, WD mass of 0.5M and 1.0 Solar masses, and distances to the system of 10, 20 and 30 pc. It was found that the PRIMA facility at the VLTI will be able to detect planets around white dwarfs once it is operating, by measuring the astrometric wobble of the WD due to a planet companion, down to 1 Jupiter mass. We show for the simulated observations that it is possible to model the orbits and find the parameters describing the potential planetary systems. / Bei von Sternen verursachten Mikrolinsen-Ereignissen beobachtet man meist symmetrische Lichtkurven einer einzelnen Linse oder asymmetrische Lichtkurven (oftmals mit Kaustik-Crossing), die durch Doppel-Linsen hervorgerufen werden. Im Prinzip kann aus der Zahl der gemessenen unsymmetrischen Ereignisse der Anteil der Doppelstern-Systeme in Abhängigkeit vom Winkelabstand abgeschätzt werden. Allerdings kann auch ein Doppelsystem Lichtkurven erzeugen, die gut mit einer Einzellinsen-Lichtkurve gefittet werden können. Die gilt insbesondere bei lückenhafter Messung oder grossen Messfehlern. In dieser Arbeit wird für verschiedene Beobachtungsbedingungen untersucht, wie häufig Lichtkurven, die von Doppellinsen mit unterschiedlichen Abständen erzeugt werden, gut mit Einzellinsen-Lichtkurven gefittet werden können und damit fehlinterpretiert werden. Es wurde herausgefunden, dass der Anteil fehlinterpretierter Lichtkurven stark von der Separation der Komponenten abhängig ist: das Minimum liegt zwischen 2 A.E. and 5 A.E. / wobei der Anteil immer noch 5% beträgt. Die Rate der Fehlinterpretationen ist höher für kurze Mikrolinsen-Ereignisse (bis zu wenigen Monaten) und für Ereignisse mit geringer Maximalverstärkung. Bei gleicher Separation steigt die Rate mit extremeren Massenverhältnissen an.<br><br> Das Problem der Degenerierung zwischen den Lichtkurven für doppelte Linsensysteme und doppelte Hintergrund-Quellen wurde anhand simulierter Daten und mit Beobachtungsdaten des PLANET Projekts untersucht. Der Fit-Code BISCO, der den genetischen Algorithmus PIKAIA nutzt, wurde geschrieben, um Doppel-Linsen Lichtkurven, die von verschiedenen Observatorien in den photometrischen Bändern I, B, und V gemessen wurden, zu modellieren. Tests mit simulierten Daten haben gezeigt, dass BISCO in der Lage ist, in einem sehr weiten Parameterbereich die korrekte Lösung für die Lichtkurve einer Doppel-Linsen zu finden. In dieser Arbeit wird die Flussverhältnis-Methode empfohlen, um die Degenerierung zwischen Doppel-Linse und Doppel-Quelle aufzulösen. Modellierungen zeigen, dass nur wenige zusätzliche Datenpunkte im V-Band genügen, um zusammen mit einer vollständigen Lichtkurve im I-Band die Degenerierung aufzubrechen. Mit sehr guter Datenqualität und zeitlich dichten Messungen, kombiniert mit genauer Modellierung von Doppel-Linsen und Doppel-Quellen, gelang die Entdeckung des bisher masseärmsten Planeten ausserhalb des Sonnensystems: OGLE-2005-BLG-390Lb, mit nur 5.5 Erdmassen. Dies war das erste Mikrolinsen-Ereignis, bei dem die Degenerierung zwischen plantarer Doppel-Linse und einer Doppel-Quelle mit extremem Flussverhältnis erfolgreich aufgelöst wurde. Für die Ereignisse OGLE-2003-BLG-222 und OGLE-2004-BLG-347 besteht die Degenerierung trotz sehr dichter Messungen. Aufgrund der Lichtkurvenmodellierung und Argumenten aus der Theorie der Sternentwicklung ist die Erklärung von OGLE-2003-BLG-222 als Doppel-Quelle und OGLE-2004-BLG-347 als Doppel-Linsen Ereignis vorzuziehen. Allerdings kann die Degenerierung ohne spektrale Daten nicht vollständig aufgelöst werden.<br><br> Bisher wurde kein Planet als Begleiter eines Weissen Zwerges gefunden, obwohl es möglich sein sollte, dass jupiterähnliche Planeten die Spätstadien der Sternentwicklung überleben und dass sich Weisse Zwerge Planetensysteme mit weiten Umlaufbahnen erhalten können. Wir planen hochgenaue astrometrische Beobachtungen von nahen Weissen Zwergen in weiten Doppelsystemen, um Planeten um Weisse Zwerge zu finden. Wir haben eine Stichprobe von Systemen zusammengestellt, in denen möglicherweise Planeten gefunden werden könnten. Wir haben synthetische astrometrische Orbits modelliert, die für diese Systeme mit existierenden und zukünftigen astrometrischen Instrumenten beobachtbar sind. Die Modellierungen wurden für astrometrische Genauigkeiten von 0.01, 0.1, 1.0 Mikrobogensekunden gerechnet. Als Abstände zwischen weissem Zwerg und Planet wurden 3, 5 und 10 Astronomische Einheiten angenommen, für den Abstand zwischen den Doppelsternkomponenten 30 A.E. Als Planetenmassen wurden 10 Erdmassen, bzw. 1 und 10 Jupitermassen gewählt, als Masse für den weissen Zwerg 0.5 und 1.0 Sonnenmassen. Die Distanzen zum System betragen 10 und 20 parsec. Als Resultat dieser Untersuchung wurde herausgefunden, dass das PRIMA Instrument am VLTI in der Lage sein wird, die astrometrischen Oszillationen, die ein Planet ab einer Jupitermasse verursacht, zu detektieren. Wir zeigen, dass es möglich sein wird, die Umlaufbahnen solcher Planeten zu modellieren und damit die Parameter dieser Planetensysteme zu bestimmen.