Summer Research Programme 2019

Oxford Astrophysics will run a summer research programme for undergraduate physics students again in 2019. We anticipate taking about 5 students. Students will work with a supervisor in the department, usually a postdoctoral researcher or lecturer, on a self-contained research project. There will also be some lectures on current astrophysics topics. Students are encouraged to take part in department life, joining researchers for coffee, discussions and seminars.

The projects run for typically 8 weeks, nominally 1st July through till the end of August. The duration may be adjusted to be shorter or longer, or to accommodate summer travel. Students will be paid as employees of the University, recieving a gross salary of approximately £300 per week (subject to tax and National Insurance deductions). 75% of the salary due for the entire project will be advanced during the first week, and the rest will be paid after completion of the project. The project is full-time but hours can be discussed with your supervisor.


Students currently in the second or third year of a relevant undergraduate degree are eligible to apply. Students who have completed a 3-year undergraduate degree and are now taking a taught Masters course are also eligilbe, as long as they are not in their final year. Applications are welcome from institutes outside of Oxford. Unfortunately, due to UK visa regulations, we are only able to accept applications from candidates within the EU.

How to apply

You should email a one-page-only application, in pdf format, to the Graduate Administrator ( by Friday 8th March 2019, with 'Summer intern application' in the subject line. Students should ask for a short academic reference letter to be emailed to the above email address by the same date. Offers will be made in March.

On your 1-page application please nominate 3 projects that take your interest, stating order of preference. Please tell us why you are interested in the programme and why you are interested in your nominated projects. Your 1-page application should also include your contact details, your year and course, a summary of your undergraduate exam results so far, and contact details (including email) of your academic referee. Please also mention any computer programming experience and any previous research experience. Please note that applications longer than 1 page will not be considered.

You are encouraged to informally contact the supervisor(s) to find out more details about the projects that interest you. For any administrative queries, contact Ashling Gordon on


Measuring the spin of a supermassive black hole

Supervisor: Dr Adam Ingram (

A supermassive black hole -- weighing millions to billions of solar masses -- is thought to be present at the centre of every galaxy. These black holes must have grown to be so massive by accreting gas from their host galaxy and merging with other black holes. The nature of this growth process can be diagnosed by measuring the angular momentum (spin) of nearby supermassive black holes: prolonged accretion from a consistent direction would spin up the black hole as it grows, whereas mergers and "chaotic accretion" from many random directions would leave behind black holes with moderate spins. In active galactic nuclei (AGN) the central black hole accretes gas through an accretion disc, which is so hot that it glows brightly in X-rays. In particular, an iron emission line is observed that is narrow in the rest frame, but the rapid orbital motion of the disc material and the gravitational pull of the black hole distort the shape of the line, resulting in a broadened and skewed line profile in the observer's frame. By modelling the iron line, we can use these distortions in order to measure the spin of the black hole. However, the summer project I ran last year demonstrated on a stellar-mass black hole in our Galaxy that previous spin measurements based on the iron line profile may be biased since they do not include an important piece of disc physics. This year's project will extend this work by testing new, more physical, models on an observation of the AGN NGC 1365 taken by the European Space Agency's XMM-Newton and NASA's NuSTAR. The original analysis of this observation concluded that the black hole must be spinning nearly at the theoretical maximum rate, but inclusion of extra disc physics may dramatically change this conclusion -- with profound implications to our understanding of black hole growth.

The work will be carried out using X-ray analysis software that is fairly straightforward to pick up for anyone with a basic knowledge of programming.

Statistical methods for cosmic microwave background experiments

Supervisor: Dr Maximilian Abitbol (

The cosmic microwave background (CMB) is light from the Big Bang that contains an imprint of fluctuations in the early universe as well as their subsequent evolution. Current experiments aim to detect a faint pattern in the polarization of the CMB, called B-modes, generated by primordial gravitational waves. These gravitational waves are thought to be created by inflation, a rapid exponential expansion of the universe in the first fraction of a second after the Big Bang. Detecting evidence for inflation would be a breakthrough in modern cosmology.

We are studying a variety of CMB analysis methods to learn about the beginnings of the universe. In order to measure B-modes in the CMB, precise statistical methods are used to identify and isolate the CMB signal from noise and other non-cosmological signals in the foreground of our Galaxy. The main focus of the project(s) will be to develop, apply, and test new statistical methods. The methods include (i) a moment expansion approach to parametric foreground modeling supported by Bayesian evidence or similar criteria and/or (ii) regression with Gaussian processes to describe modeling residuals. We will use existing observations as well as simulations to test our results. Additional projects could include using existing methods on new data at 5 GHz to detect CMB anistropies at this frequency for the first time.

Requirements: Familiarity with Python or similar languages. An interest in statistics would be helpful as well.

The relation between galaxies and dark matter halos

Supervisor: Dr Harry Desmond (

While correlations among galaxy properties are known empirically from surveys, and those among halo properties are known from dark matter-only simulations, the relations between galaxy and halo properties remain unclear. It is these that we expect to hold the key to galaxy formation in a universe dominated by cold dark matter.

A popular technique for inferring the relation between galaxy and halo mass is 'abundance matching': essentially one assigns the nth most massive galaxy observed in a given region to the nth most massive halo produced in a simulation box of the same volume. Traditionally the measure of galaxy mass has been stellar mass, which is relatively easy to measure. However, a sizeable fraction of smaller galaxies' total ("baryonic") mass is contained in cold gas, and the requisite data now exists to upgrade abundance matching to use total galaxy mass (stars + gas). Not only will this provide insight into the physical processes governing the coevolution of baryons and dark matter, but may also shed light on several apparent failings of standard cosmology at the dwarf galaxy scale.

This project will continue ongoing work at Oxford. The student will adapt the abundance matching technique to connect galaxies to halos in a structure formation simulation, generate predictions for the galaxy-halo connection based on this result and test them against observations of galaxy clustering and kinematics. It is expected that all work will be done in Python.

Very-high-energy gamma-ray astrophysics

Supervisor: Prof Garret Cotter (

Very high-energy gamma-ray astrophysics is an exciting field spanning fundamental physics and extreme astrophysical processes. Our investigations range from the nature and variety of particle acceleration around supernovae and black holes, to the nature of matter and physics beyond the Standard Model at energies inaccessible to terrestrial experiments.

The group in Oxford works on both experiment and theory. We are members of the High Energy Stereoscopic System (H.E.S.S.) in Namibia, which is at present the world’s largest gamma-ray observatory, and the Cherenkov Telescope Array (CTA), the next-generation global gamma-ray observatory that will commence operations at Paranal, Chile, and La Palma, Canary Islands, in the next few years.

Students participating in the summer programme will have opportunities to participate in a variety of our activities, such as advanced analysis techniques for the large volumes of data that will be generated when CTA becomes operational, and investigating theoretical models that let us use gamma-ray observations to investigate the physical properties of the relativistic jets and supermassive black holes in quasars and active galaxies.

Superconducting Parametric Amplifiers for Quantum Computations and Astronomical Receivers

Supervisor: Dr Boon Kok Tan (

The best low noise amplifiers used in almost all highly sensitive experiments, such as quantum computations and astronomy, are generally based on the high electron mobility transistors (HEMTs) technology. However, even the state-of-the-art HEMT devices struggle to reach the quantum-limited performance. The emerging new technology of the superconducting parametric amplifiers (SPAs) has shown to be a promising candidate to overcome this limitation. An SPA relies on the nonlinear reactance to convert energy between different frequencies in a mixing process. To reach high gain, the signals must interact strongly with the nonlinearity and this can be achieved by construct an SPA with long propagation length – namely a travelling wave parametric amplifier (TWPA) – which have much larger bandwidth and higher dynamic range than the traditional resonator-based approaches.

At Oxford, we are actively involved in developing these quantum limited TWPAs for their eventual applications in quantum information technologies and astronomical receivers. One of our aims is to explore and compare the various TWPA designs, using analytical models to predict the gain-bandwidth performance of these devices, and seek for the optimal design for a particular application. This project, therefore, involves reviewing research works done in this area, develops the model by writing a software code to solve the coupled differential equations, and analyse the performance of the TWPA using the written software. The project will best suit a student who is interested in learning about astronomical receivers, basic mixing theory, and enjoy developing theoretical/computational software code in relevant to fundamental physics phenomena.

The impact of a non-Gaussian likelihood on cosmological parameter constraints from weak gravitational lensing

Supervisors: Dr Shahab Joudaki ( & Dr Harry Desmond (

Weak gravitational lensing is considered to be one of the most promising probes of the underlying cosmology as a result of its particular sensitivity to both the large-scale structure and expansion of the Universe, and forms a cornerstone in the present efforts to understand the properties of the dark energy that is causing the expansion of the Universe to accelerate. In cosmological parameter inferences from weak lensing, the likelihood is generally assumed to be Gaussian. Using a large suite of N-body simulations this has been shown to be incorrect by up to 30% of the measurement uncertainty for current surveys (Sellentin et al. 2018). We will develop methods to describe the full shape of the likelihood without employing the Gaussian approximation, and propagate this into parameter constraints to quantify the biases this approximation induces given current and future lensing data.

Required skills: General understanding of cosmology, statistics, and programming (ideally Fortran).

Weighing supermassive black holes with molecular gas

Supervisors: Mark Smith ( & Prof Michele Cappellari (

Supermassive black holes (SMBHs), found to reside at the centres of all massive galaxies, are believed to play a significant role in galactic evolution. This is demonstrated by the array of empirical correlations between properties of the host galaxies and the SMBH mass. However, the physics driving these correlations is not yet understood, and the measurements they are derived from have large uncertainties. We cannot currently tell if different types of galaxies follow different correlations, or whether one law holds for all.

The millimetre-Wave Interferometric Survey of Dark Object Masses (WISDOM) project aims to address these questions by making high-precision measurements of the masses of a large sample of such SMBHs. We can do this by exploiting the high angular resolution offered by the Atacama Large Millimetre/submillimetre Array (ALMA; [1]) to resolve the dynamical influence of the black hole on orbiting molecular gas. Since our pioneering Nature paper ([2]), we have made a series of measurements in galaxies of all morphologies (Latest: [3]).

In this summer project, you will work with the WISDOM team to make an independent measurement of the mass of a SMBH using existing ALMA data and the latest generation of our dynamical modelling tools. This measurement will be compared to other measurements of the SMBH in this galaxy to investigate systematic biases between different measuring techniques.

The project will require the use of software tools built in both IDL and Python, so programming experience in any language will be useful.

Probing the initial conditions of the Universe

Supervisors: Supervisors: Dr Shahab Joudaki ( &
Darsh Kodwani (

A wealth of cosmological information has been obtained over the last few decades from measurements of the cosmic microwave background to late-Universe observables that probe the large scale structure, such as weak gravitational lensing and the clustering of galaxies. The theoretical framework to interpret the cosmological data is the so called `Lambda cold dark matter' (LCDM) model, where the initial conditions of the Universe are set by nearly scale-invariant adiabatic Gaussian perturbations. However, these initial conditions are not unique and another class is known as isocurvature/entropy perturbations. While much effort has gone into constraining these types of initial conditions, we will take a model-independent approach in constructing the primordial power spectrum in a non-parametric way that adequately allows the initial conditions to vary across physical scale. By implementing the primordial power spectrum into an existing `Markov Chain Monte Carlo' code that propagates it into observables such as the cosmic microwave background, we can determine the viability of the initial conditions, examine the degeneracies with parameters of the standard cosmological model, and probe novel scale-dependent physics in the early Universe.

Required skills: General understanding of cosmology, statistics, and programming.

Exploiting gravitational lensing to reveal distant galaxies

Supervisor: Dr Rebecca Bowler (

Gravitational lensing has the power to reveal otherwise hidden galaxies. Massive foreground clusters lens the background galaxy population, allowing faint galaxies at high-redshifts to be discovered. In addition to strong lensing by clusters, background galaxies can be lensed by a single massive elliptical galaxy, resulting in an image of the distant object that is both brighter and multiply imaged ([1], [2]). In this project you will search for lensed high-redshift galaxies using our unique photometric dataset in the XMM-LSS field [3] . This field, which covers 4.5 square degrees on the sky, is the perfect combination of depth and area to discover new gravitational lensed systems, and provides a practice dataset for upcoming very wide area datasets (e.g. LSST and Euclid). Using the photometric redshifts and masses for galaxies in the field, you will first select the most massive galaxies (our lenses) in the field. Around these objects you will then search for multiply imaged sources using the photometric data. Once a sample has been established, you will work to understand the properties of the lensed galaxies by compiling the multi-wavelength data available. The lensed galaxies will be perfect candidates for detailed follow-up observations with other telescopes (e.g. Hubble, ALMA).

The project will be computational. Some coding experience will be helpful, but is not essential.