Find a PhD Project Here
Opportunities for fully funded PhD or EngDoc research projects are available in all fields of research within the School. You may search for current projects on this page. APPLY HERE for a PhD Place.
Search current PhD opportunities in the School of Physics & Astronomy:-
This project qualifies as an STFC studentship in Data-Intensive Science.
Extrasolar planets have proven to be far more diverse than the planets in the solar system. This project will deal with modelling globally circulating atmospheres to study cloud formation and lightning in extrasolar planets.
Cloud formation is a major challenge for understanding planetary atmospheres because clouds determine the spectral appearance of the planet (and also of brown dwarfs) and they influence the dynamic behaviour of the atmosphere. 3D simulations will be used which provide a rich data set for the atmosphere structure as well as for the details of cloud and gas-chemistry. Beside the scientific analysis of the modelling data, the project requires 3D visualisation for data analysis purposes.
Recent studies of the magnetic fields of very low mass stars shows a strange and so far unexplained behaviour. Some have strong, simple magnetic fields, and some have much weaker, complex, solar-like field magnetic fields. We do not fully understand why this difference occurs, but this project involves using the maps of the magnetic fields of these star to explore the physics of their coronae and winds and to examine the impact on any orbiting planets.
-> This project qualifies as an STFC studentship in Data-Intensive Science. <-
Determining the demographics of cool planets by means of microlensing is one of the key science goals of NASA’s WFIRST (Wide Field Infrared Space Telescope) mission. However, current approaches for modelling the acquired data are already unsuitable for managing the much smaller number of gravitational microlensing events that are currently detected by the most advanced ground-based surveys. The major bottleneck to be overcome is the reliance on human judgement in the data analysis process. Any scalable solution needs to be fully-automated (“data-in-model-out”), taking into account accurately the statistics of the photometric data (i.e. not claiming planets from statistical “noise”) and exploring highly-dimensional non-linear parameter space.
Many solar-like stars show cool, dense clouds of gas trapped within the million-degree plasma of their outer atmospheres (or coronae). These so-called ``slingshot prominences'' carry away angular momentum when they are ejected and also are also responsible for mass-loading of the stellar wind. As a result, they may form an important part of the spin-down of young stars, and their impact on orbiting planets may lead to enhanced stripping of the planetary atmosphere.
A main diagnostic of the particle dark matter is its annihilation rate, which depends sensitively on the dark matter density profile. The student will explore various density models of the dark matter, taking into account the effects of black holes and baryonic dynamics.
What is the lowest mass object that can form like a star? And how many massive planets are ejected from their planetary systems during the formation process? These two questions will be tackled in this project.
Over the last years, we have carried out a search for the lowest mass free-floating objects in star forming regions, in a project called SONYC (short for Substellar Objects in Nearby Young Clusters). In SONYC we used the largest existing ground-based telescopes to make ultradeep surveys of the youngest clusters on the sky. While we found plenty of brown dwarfs (with some interesting evidence for environmental difference in the formation of brown dwarfs), we did not find many objects with super-Jupiter masses, the presumed ejected giant planets. If they exist (and we expect that they do), they will be below our mass threshold of 5 Jupiter masses and are still to be discovered.
In the next step of this project (and in this PhD project) we will use the James Webb Space Telescope to explore the domain of free-floating rogue planets with masses between 1 and 5 Jupiter masses. We have good chances to get observing time to get this project started right after the JWST begins operation in 2018. The student will prepare the observations and explore follow-up avenues, with JWST and other facilities, and then be the first to analyse the data. In addition, we will work with the second data release from Gaia to pin down the fundamental parameters of young brown dwarfs. This will lead to new contraints on star formation simulations and insights into the transition from star to planet formation.
This will be a strongly observational project, which requires to learn the details of optical and infrared observations, the physics of ultracool objects, the intricacies of disentangling emission from objects, disks, and accretion, as well as an interest in collaborating with people from the theory side, including atmospheric physics and star/planet formation.
The University of St Andrews is a founding institutional member of the Wide-Angle Search for Planets (WASP) project, which is a consortium comprising 6 UK universities and 3 overseas observatories. We use two arrays of wide-field camera lenses backed by large-format CCDs to perform high-precision photometry of millions of stars each night, looking for the 1% dips in light that betray gas-giant planets whose orbital planes are close enough to the line of sight that they transit their host stars. With over 166 hot Jupiters confirmed to date by radial-velocity follow-up, WASP is the world’s leading ground-based producer of transiting planets.
Over the last decade, WASP has amassed a database of light curves on some 31 million objects. In the era of NASA’s TESS mission, the WASP data archive will have an important role to play in documenting the past variability history of new planetary systems and variable stars identified from space.
The WASP database has the potential to address many other areas of time-domain astrophysics. We are already applying supervised and unsupervised machine learning methods to distinguish objects of astrophysical interest from false alarms caused by systematics, and to classify objects showing astrophysical variability. This allows likely transiting planets to be distinguished from eclipsing binary stars, rotational variables and pulsating stars. Early validation by rejection of such “astrophysical false positives” increases greatly the efficiency of radial-velocity follow-up. This work also involves cross-matching with other large astronomical databases: 2MASS for infrared colours and angular-diameter estimates, and Gaia for parallaxes and proper motions.
Students working on this Data-Intensive Science project will acquire training and experience in a variety of machine classification techniques, Bayesian parameter estimation with Gaussian-process regression to model systematics and correlated noise, Gaussian mixture models and hierarchical Bayesian inference for extracting population parameters.
Wild, Dr Vivienne - email@example.com
->This project qualifies as an STFC studentship in Data-Intensive Science.<-
How galaxies form and evolve is one of the outstanding questions of modern astrophysics. Extremely large spectroscopic surveys are providing an increasingly detailed census of both local and distant galaxies. Considerable progress is being made on quantifying the changing demographics of the galaxy population over the majority of the age of the Universe, but significant improvements in methods are demanded by the increasingly large samples, and often decreasing quality of the individual observations.
In the last decade a Bayesian approach to the fitting of sophisticated models to high quality spectra and/or multiwavelength photometry has become common place in the analysis of galaxy spectral energy distributions (SEDs) at all redshifts (Walcher et al. 2011). The result are robust physical properties, such as stellar mass and star formation history, as well as well understood degeneracies between fitted parameters. We now know that massive galaxies come in two main types - elliptical/quiescent and spiral/star-forming. Understanding why this is, is a key science goal in astrophysics.
How much information are we missing by treating galaxies as independent entities to determine their physical properties, rather than a population of objects with common origin? How can we use the same fitting approaches for lower quality observations of increasingly large samples?
Hierarchical Bayes techniques have appeared in the astronomical literature to solve problems as diverse as QSO redshift estimation (Bovy et al. 2011), exo-planet orbit analysis (Hogg et al. 2011), properties of SN 1a light curves (Mandel et al. 2009), and photometric redshifts (Leistedt et al. 2016). They differ from standard Bayesian methods by fitting the entire dataset in a coherent manner, instead of single objects as entirely independent entities. By applying these methods to galaxy evolution studies, we will improve our ability to break degeneracies. These methods could be applied to e.g. complete populations of galaxies in spectroscopic or photometric surveys, or entire integral field datacubes of single galaxies.
The school of physics and astronomy at the University of St Andrews is a member of the UK participation group in SDSS-IV (fourth generation of Sloan Digital Sky Surveys, www.sdss.org), a large international collaboration encompassing several astronomical surveys (including MaNGA and eBOSS). Drs Wild, Weijmans, and Tojeiro all have data access. Note that access to data from the SDSS-IV survey is rare in the UK, and guaranteed through the supervisors core involvement in the survey.
The project involves the development of statistical techniques to make them applicable to astronomical datasets. Applications from students with a background in maths, statistics or physics and interest in astrophysics are welcome, as well as from students with a background in astrophysics but strong aptitude for maths and statistics.
Please contact Vivienne Wild. the lead supervisor for the project, for more information (firstname.lastname@example.org)
Bovy J., Myers A. D., Hennawi J. F. et al. arXiv:1105.3975
Hogg D. W., Myers A. D., Bovy J., 2010, ApJ, 725, 2166
Leistedt B., Mortlock D., Peiris H., 2016, MNRAS, 460, 4258
Mandel K. S.; Wood-Vasey W. M., Friedman A. S., Kirshner R. P., 2009, ApJ, 704, 629
Walcher C. J., Groves B., Budavri T., Dale D., Ap&SS, 2011, 331, 1
Extensive layers of diffuse ionized gas are observed in the Milky Way and other galaxies. This project will study the structure, ionization, heating, and dynamics of diffuse ionized gas using a combination of 3D Monte Carlo radiation transfer codes and recent 3D dynamical models of a supernova driven ISM.
This project qualifies as an STFC studentship in Data-Intensive Science.
Light travel time delays enable micro-arcsecond mapping of accretion discs and broad emission-line regions around the super-massive black holes in the nuclei of active galaxies. Using our share of time on the LCOGT robotic telescope network, along with data from HST, Swift and Kepler satellites, we are monitoring spectral variations of Active Galactic Nuclei to measure black hole masses, accretion rates, and luminosity distances. By decoding information in the reverberating emission-line profiles, we make 2-dimensional velocity-delay maps of broad emission-line regions, mapping the velocity field and ionisation structure of the accretion flows. The student will acquire and analyse such datasets, fitting parameterised models using MCMC methods, image reconstruction using Horne maximum entropy fitting code MEMEcho, and photo-ionisation codes such as Ferland's Cloudy.
Most stars form in clusters, where energetic feedback from massive
(proto)stars--including outflows, ionization, heating, and
winds--shapes the environment and impacts accretion. The relative
importance of different feedback processes is a key outstanding
issue in our understanding of massive star formation.
The aim of this project is to conduct the first large-scale
observational study of the role and physics of feedback in massive
(proto)clusters. This will involve analyzing high-resolution data
from recently-upgraded (sub)mm and cm-wavelength interferometers, in
particular the Submillimeter Array (SMA), the Jansky Very Large Array
(VLA), and, potentially, the Atacama Large Millimeter/sub-millimeter
Array (ALMA). The observational results will be compared with
simulated observations of numerical models of massive star and
Dust grains in protoplanetary discs generally charge up due to photo-effect, electron attachment, and charge exchange reactions with molecular ions. Grain-grain collisions can possibly lead to an additional statistical charging (contact electrification), which has not yet been thoroughly discussed in the disc community yet (see e.g. Muranushi 2010). If grains of different sizes collide, charge up size-dependently, and move selectively (by gravitational settling), a large-scale charge separation could build up, leading to lightning in discs. This scenario has been proposed to explain intra-cloud lightning observed in volcano plumes, as well as lightning in the Earth’s atmosphere and in exo-planets (Helling et al. 2016). Similar effects could take place in protoplanetary discs, causing radio emission and having a long-term impact of the chemical composition of the gas.
* Is frictional charging a key process for midplane ionisation and the MRI in discs?
* Can the gravitational settling of charged grains build up electrostatic fields in discs?
* Can this field overcome the break-down field to cause spontaneous discharge processes (lightning)?
* Where exactly, in the disc, are these processes most likely to occur?
* Could lightning lead to observable signals, like short-term radio variability?
* Could lightning have a long-lasting impact on the chemistry in the planet-forming region?
PhD-student is expected to implement triboelectric charging rates into ProDiMo, using typical turbulent dust velocities from MHD disc models. The resulting charge distribution of the grains will be studied depending on size and location in the disc, and consequences for large-scale electrification and lightning in discs shall be discussed.
GravityCam is a proposed mosaic camera composed of ~100 EMCCDs with a novel design that for the first time combines a wide field with a very fast readout, thereby achieving an angular resolution of 0.15’’ by means of lucky imaging and opening up an entirely new observing paradigm for ground-based astronomy, its only real competition being in space. A core science driver for GravityCam is a Galactic bulge microlensing survey that could go about 4 magnitudes deeper than current efforts for the same signal-to-noise ratio and exposure time, and thereby at the same sensitivity probe cool planets (or satellites) that are 100 times less massive, which gives access to a hitherto uncharted region in planet parameter space extending down to Lunar mass. In addition, as a unique and versatile instrument, GravityCam will be suitable for addressing a wide variety of scientific applications, including in particular studies of dark matter (by means of weak lensing), fast-varying astronomical objects, asteroseismology, variability and astrometry in crowded fields, occultations by small Solar-System bodies, and transiting extra-solar planets, while providing an extensive resource for general data mining of the high-speed variable sky.
You will have the opportunity to optimise the design of a GravityCam microlensing campaign by means of simulations in order to maximise the science output relating to the arising planet population statistics, which requires carefully balanced choices of exposure time and cadence dictating the total survey area. Moreover, you can take part in current observational efforts (MiNDSTEp and Robonet-II), pioneering crowded-field lucky-imaging photometry and real-time scheduling of microlensing targets across telescope networks.
New spatially resolved observations of protoplanetary discs have revealed so far unseen spatial structures within the discs, such as rings, holes, spiral arms, warps, shadows, and large vortices. They are detected at various wavelengths, in the gas and dust, in scattered light and in thermal emission. These structures are very likely direct signposts of the planet formation process in the discs, yet current hydrodynamical disc models suffer from a very basic uncertainty, namely the poor treatment of radiative transfer and heating/cooling effects in hydrodynamical disc models. The supervisor is an expert in the fields of chemistry, heating & cooling and radiative transfer, but these techniques need to be extended and merged with (magneto) hydrodynamics in 3D to get ready for the new challenges in the era of spatially resolved disc observations.
This project aims at merging current state-of-the-art modelling techniques concerning (magneto) hydrodynamics, chemistry and radiative transfer in protoplanetary discs. Based on the radiation thermo-chemical disc code "ProDiMo" which includes a very detailed treatment of 2D continuum and line radiative transfer, and gas energy balance, we aim at the production of numerical look-up-tables of equilibrium gas and dust temperatures, chemical and ice composition of the gas, and effective heating & cooling rates suitable for hydrodynamical disc simulations.
The task is to build a brigde between thermo-chemical and hydrodynamical disc simulations. The student will study and learn how to run both types of models, calculate the look-up-tables with ProDiMo, and then apply these in hydrodynamical disc simulations.
The mass distribution of the Galaxy is being / will be mapped out in great detail in the next decade with the numerous surveys of the Galaxy, including Segue, RAVE, GAIA, and completed ones like 2MASS, DENIS. A model for the potential and phase space of the galaxy is essential to bring various pieces of information together. The student will develop such models building on experience from existing models.
-> This project qualifies as an STFC studentship in Data-Intensive Science. <-
World-spanning networks of robotic telescopes have opened up new opportunities for the regime of time-domain astronomy. The deployment of such networks comes along with the roll-out of a new generation of powerful surveys that provide “transient factories”, calling for prompt and extensive follow-up monitoring. How can such resources be used efficiently and how can observations be coordinated? While the scheduling of telescopes so far has mostly been discussed from the perspective of service providers, we need to look at this from the perspective of the community of users who want to achieve a scientific objective using a diversity of resources coming from different providers. The selection of a specific target to be observed at a specific time by a specific instrument needs to take into account: 1) the science goals, 2) the acquired data on all potentially relevant targets, 3) the technical specifications of the instrument, 4) the observability of the targets. In particular, target priorities will change as fast as the targets vary themselves and data therefore need to be analysed in quasi-real time in order not to miss critical features. Consequently, a large amount of data need to be pulled together and processed immediately, while continuous monitoring requires uninterrupted operatability. Within larger user communities, objective functions and resulting monitoring strategies also need to consider the ownership of data and both individual and collective benefits. The University of St Andrews has pioneered the implementation of automated target selection strategies for the follow-up of ongoing gravitational microlensing events with the LCOGT/SUPAscope and MiNDSTEp networks. Students will face a technology challenge on data processing, modelling, and management at the intersection of astronomy and computer science, matching the requirement to achieve a fast throughput. Solutions to this sort of problem will likely to be transferable to serve applications in other areas.
Rotation is a fundamental property of stars. The angular momentum regulation of stars is linked with the evolution of disks, the physics of magnetically driven winds, and the interior structure. Stars like the Sun start with a period of a few days, but spin down to periods of weeks and months over the course of billions of years. This project is focused on investigating the spindown of very low mass stars, the most abundant type of stars in our Galaxy, which present a serious challenge for our current understanding of stellar rotation. In contrast to solar-mass stars, they have long spindown timescales of ~1 gigayear or more. The extreme case are brown dwarfs, which do not seem to spin down over a Hubble time, comparable to giant planets. All this is probably related to the atmospheric physics, particularly the magnetic properties. We are therefore particularly interested in probing the link between rotation and magnetic activity. We have been granted observing time with the Kepler-2 mission to get lightcurves for very low mass stars and brown dwarfs are various stages of their evolution. Since these stars are magnetically very active, star spots cause a periodic modulation of the flux from which the rotation period can be measured with high accuracy. The same lightcurves also give information about magnetic spots. The archive of the Super-WASP planet search will also be used to study the longest timescales. Another dataset from the Very Large Telescope will be used to examine rotation rates in young free-floating planetary mass objects. There is scope for new observations carried out with large telescopes. We will measure rotation periods, probe the period-activity correlation, compare with new models for the stellar spindown, and investigate the possibility of gyrochronology (i.e. estimating ages from rotation rates) for red and brown dwarfs.
This project is to develop the first models of resolved star formation on galactic scales. This will involve modelling a full galactic potential and how it drives the formation of molecular clouds and the onset of gravitational collapse and star formation. feedback from ionisation and supernova will be included to assess molecular cloud lifetimes and star formation efficiencies.
Tau Boo is the only star for which we have been able to track the full cyclic reversal of the stellar magnetic field. This system is also well-known, however, because it hosts a Hot Jupiter that is so close to the star that it may lie within the stellar corona. What is the nature of the interaction between the star and planet in this case and is it related to the puzzling nature of the very short magnetic cycle? This project will investigate tau Boo and other similar star-planet systems.
When a newborn solar-like star emerges from its natal cloud it is still surrounded by a substantial disk of dust and gas. At this stage of pre-main sequence evolution the star interacts with the inner disk via its large-scale magnetic field, which channels gas onto the stellar surface at high velocity. Recent large observing programs have begun to reveal how their magnetic geometries are linked to their location in the Hertzsprung-Russell diagram. Tentatively, it appears as though solar-like stars are born with simple axisymmetric magnetic fields that become more multipolar/complex and non-axisymmetric as the stellar interior structure varies from fully to partially convective. Should this stellar structure change occur before the disk has dispersed it will have implications for the magnetic star-disk interaction, the coronal evolution of the star itself, the balance of torques in the star-disk system, and the rotation rate of the star. Using the latest observational data as a basis, the student will model the star-disk interaction and coronal magnetic evolution as stars evolve across the pre-main sequence.
For more evolved pre-main sequence stars, where the disk has dispersed but the star is still contracting under gravity, it has been observed that the scatter in X-ray luminosities decreases for stars in older star-forming regions, approaching main sequence cluster levels by about ~30 Myr (roughly the pre-main sequence lifetime of a solar mass star). Using new data currently being acquired as part of a large program at the Canada-France-Hawai'i Telescope we will model the coronal evolution of pre-main sequence stars, and produce a theoretical grounding for the observed evolution of their rotation-activity relationship.
There are several outstanding issues in current models of star formation. One of these is the role of feedback from young stars in producing subsequent generations of young stars. Triggering of star formation through supernova events is likely to be the dominant mechanism. Numerical simulations of SNII impacting on molecular clouds and the triggering of star formation will be used to develop physical models, and ultimately observational predictions and tests of the process.
Weijmans, Dr Anne-Marie - email@example.com
Wild, Dr Vivienne - firstname.lastname@example.org
-> This project qualifies as an STFC studentship in Data-Intensive Science. <-
The Dark Energy Spectroscopic Instrument survey will gather spectra for many millions of galaxies over 14,000 sq. degrees of sky. The size and depth of DESI, particular the Bright Galaxy Survey, will allow us to study the large-scale structure and detailed local environment of each galaxy with unprecedented accuracy. Combined with careful spectral analysis of galaxies, which tells us about its formation history, chemical enrichment and dust content, we will be able to link the evolution of galaxies with their cosmological and local environment with unprecedented clarity.
However, current spectral analysis techniques, although successful, rely on much higher signal to noise data than will be delivered by DESI.
In the regime where we have very many N spectra of low S/N, you will explore forward-modelling techniques, whereby model populations of galaxies are forward-modelled through a survey’s window function, which includes selection and observational effects such as target selection, photon-noise, foregrounds, etc. In that regime, we are interested in recovering the mean physical properties of a set of M galaxy populations, where M << N (see Montero-Dorta et al. 2016 for a simplified application to another dataset). The potential of this technique lies in the fact that the redshift evolution of the galaxy populations may be measured in a completely self-consistent way, using a hierarchical Bayesian approach. In addition you will explore simpler, fully data-driven techniques to recover a mean signal from noisy data. This is a well-studied problem in applied statistics. Often called denoising or signal reconstruction algorithms, such methods work by either filtering (e.g., using wavelets); Gibbs sampling (e.g., Wandelt 2004 for a CMB-motivated approach); or noise subtraction (a common technique in speech recognition).
You will then be able to study the detailed evolution of galaxies within the intricate context of their environment, which will be beautifully characterised by DESI.
This project will suit a student with a keen interest in data analysis, algorithm development, applied statistics and data science.
Please contact Dr Rita Tojeiro, the lead supervisor for this project, for further information.
The DESI survey: https://arxiv.org/abs/1611.00036v1
Montero-Dorta et al. 2016, MNRAS, 461, 1131
Wandelt 2004, arXiv:astro-ph/0401623
Currently popular models for the formation of stars like the Sun invoke accretion along magnetic field lines from a protoplanetary disk onto hot spots on the stellar surface. This star-disk model can explain the observed infrared emission from disks and also the ultraviolet excess emission produced from shocks of accreting material impacting the stellar surface. Young stars also display multi-wavelength variability on a wide range of timescales, again attributed to the accretion geometry. The on-going YSOVARS observing campaign is obtaining vast optical and infrared datasets on young stars revealing the complexities of their temporal variability. The goal of this PhD project is to use detailed three dimensional Monte Carlo radiation transfer simulations to model the observed multiwavelength data from the YSOVARs project. Working together with Kenny Wood and Aleks Scholz at St Andrews and in collaboration with members of the YSOVARS team in California, the student will explore different classes of variability and the different star-disk-magnetic field configurations that produce the observed light curves. By modeling data from the vast YSOVARs archive, we will learn about magnetic accretion geometries, disk warping, and the accompanying variability on a range of different time and spatial scales, all contributing to a greater understanding of the star formation process.
Informal enquiries to Kenny Wood: email@example.com
Carr & Najita (2008) have established that class II T Tauri stars usually exhibit rich molecular emission spectra of H2O, OH, CO2, HCN and C2H2. These emissions are often superpositions of many (up to hundreds of) individual emission lines. JWST/MIRI and, in the future SPICA/SMI, will observe protoplanetary discs with unprecedented spectral resolution and signal/noise.
In the frame of the supervisor's FP7 project "DIANA" we have developed the new fast line tracer FLiTs (yet unpublished work) which can compute formal solutions of the line radiative transfer problem for tens of thousands of spectral lines simultaneously, including the Keplerian velocity fields and physical line overlaps. These line radiative transfer calculations are based on ProDiMo (Woitke et al.2009) thermo-chemical disc models, which compute the chemical abundances and temperatures of gas and dust.
These two developments allow us to harvest future JWST and SPICA line observations of discs. Our models predict these lines fully consistently with the calculated 2D disc structures, which is a much
more powerful approach than previously used parametric LTE-slab models.
The science questions are
* What are the element abundances, and what is the molecular composition of the gas in the planet-forming regions of protoplanetary discs?
* Why do some T Tauri stars show strong molecular emission lines whereas others don’t? Why do Herbig Ae stars show weaker line emissions?
* Can we use IR molecular emission lines to determine the spatial disc structure and diagnose disc anomalies such as gaps, vortices and spiral waves at radial distances of a few AU?
* Can we conclude about dust opacities and gas/dust ratios in the planet forming region?