Black holes have emerged in recent years as key players in many areas of astrophysics. They hold the key to understanding extreme physical processes which cannot be replicated in Earth laboratories, and are the gateway to the invisible universe of gravitational waves. But their birth, growth, evolution and deaths all remain hotly debated.
 
This PhD project aims to substantially advance our understanding of the Milky Way population of black holes, and use this to infer the wider properties of black holes in other galaxies. The student will work at the interface of observational and theoretical astrophysics, analysing state-of-the-art data from survey facilities such as Gaia, the Vera Rubin Observatory (LSST), eROSITA/SRG, and VISTA/4MOST.
 
Key goals include:
 
• Constraining the demography of black holes in the Milky Way, pushing sensitivity limits by orders-of-magnitude beyond previous studies;
• Exploring machine-learning methods to accelerate parameter inference;
•Identifying and characterising new black hole candidates, including follow-up spectroscopy for dynamical mass and kinematic measurements;
•Using these results to predict signals for next-generation gravitational-wave detectors and to inform models of supermassive black hole growth in other galaxies.
 
The project offers fully-rounded astrophysics training and skill development in:
 
•Data analysis and coding (including high-performance computing, if desired);
•Observational techniques using world-class astronomical facilities;
•Critical evaluation of theoretical models;
•Scientific communication, presentation, and publication.
 
Collaborations are anticipated with international teams in the USA, South Africa, Chile, and Europe. Training will be provided in all aspects of observational astronomy and data science. Applicants should be comfortable with at least one programming language; additional coding training will be available. Informal enquiries are welcome to Prof. Poshak Gandhi.
 

Type I X-ray bursts-also known as thermonuclear bursts-occur on the surfaces of neutron stars in low-mass X-ray binary systems. As gas from a companion star accretes onto the neutron star, it spreads across the surface and forms a layer of fresh material. Under the star's immense gravity, this layer becomes compressed and heated until it reaches the critical temperature and density required for thermonuclear ignition.
 
At that point, nuclear fusion of hydrogen begins, rapidly converting hydrogen into helium and later to heavier elements. The resulting runaway nuclear burning releases an enormous amount of energy in the form of X-rays-an a burst lasting only seconds to minutes, yet radiating more energy than the Sun emits in a week. These bursts provide a unique observational window into the extreme physical conditions found on neutron star surfaces.
 
In this project, the student will develop and run numerical simulations to model how the burning front spreads across the neutron star once ignition occurs. The results will be compared with observational data and theoretical predictions to refine our understanding of thermonuclear processes in compact stars.
 
Beyond neutron stars, the student will also explore how the computational tools and methods developed here can be applied to other astrophysical and planetary contexts-such as flame propagation in other stellar environments and atmospheric dynamics on exoplanets.
 
The student will join the High-Energy Astrophysics Group at the University of Southampton and collaborate with leading researchers worldwide. We are looking for a candidate who enjoys programming and has strong coding skills, with enthusiasm for using computational methods to tackle challenging astrophysical problems.

Are you interested in black holes and the early universe? Did you ever wonder how our universe came to be as we see it today? Then this project is for you! Recent observations suggest that quasars in the early universe have SMBHs that are too massive to form in the short time since the big bang, shaking the foundations of cosmology and our understanding of black hole growth. The main problem is that we cannot be sure that the methods employed to estimate those SMBH masses are reliable in the conditions of the early universe. GRAVITY+ will deliver transformational capabilities to measure spatially resolved, dynamical supermassive black hole (SMBH) masses up to redshift 4.5 and possibly beyond. Our first data from such a high redshift quasar indicate that current mass estimates may indeed be out by a factor of 10. Using these new GRAVITY+ measurements, we will be able to recalibrate early universe measurements and deliver precise and accurate SMBH masses. As part of this project, you will work on GRAVITY+ data of quasars between redshift 1 < z < 4.5, perform dynamical modelling to determine the SMBH masses, and compare those measurements to estimates from more generic methods used in the early universe. With this comparison, you will be able to characterise how high-luminosity, fast growing objects are offset from local sources and how early universe mass estimates need to be corrected. You will work in a team funded through an ERC Advanced Grant-selected project and will be embedded in an international collaboration with institutions from across Europe, Australia and China. This will provide ample opportunities to perform research visits abroad and contribute to the wider GRAVITY+ programme.

All quasars are powered by the same central engine: a supermassive black hole that is surrounded and fed by a luminous accretion disk. Approximately 15% of all quasars exhibit clear evidence for powerful outflows driven from these disks, in the form of broad, blue-shifted absorption lines. However, these so-called "broad absorption line quasars" (BALQSOs) are just the tip of the iceberg: since disk-driven winds cannot be spherical, BALQSOs are just the sub-set of quasars viewed at a particularly favourable orientation. In reality, all quasars are likely to drive such winds. This is important, because these outflows provide a key feedback mechanism: they can remove significant amounts of mass, energy and angular momentum from the quasar and inject it into the surrounding (inter-)galactic medium. However, despite their importance, we know almost nothing about these accretion disk winds. For example, the geometry, kinematics, and even the basic driving mechanism responsible for launching them are still basically unknown.
 
The aim of this PhD project will be to remedy this situation by modelling the wind-formed observational signatures of quasars. This work will be carried out in the context of an established collaboration (which includes two other PhD students and one postdoctoral fellow at Southampton) and will use an existing, state-of-the-art Monte Carlo radiative transfer code. The ultimate goal we are pursuing is to determine the fundamental parameters of quasar accretion disk winds and thus shed light on how they regulate the fueling of supermassive black holes and the feedback of energy into their environment. In addition, we aim to shed light on quasar unification: is it possible that most observational signatures we associate with (even non-BAL) quasars are actually shaped by disk winds?
 

The detection of gravitational waves (GWs) from merging black holes and neutron stars has been one of the greatest breakthroughs in (astro)physics in recent years. The next huge milestone in this field will be the launch of the space-based LISA GW observatory (planned for 2035). LISA's sensitivity to low-frequency GWs will allow it to detect completely new and different source populations compared to current ground-based observatories, including (for the first time) close binary stars in the Milky Way.
 
Such binaries are astrophysically important in their own right, but they can also form a GW background that may make it harder to detect distant and/or fainter sources. Since close binaries are critical for LISA mission planning, huge effort is currently being dedicated to estimating the number and properties of these systems. So far, essentially all of this effort has so far been dedicated to compact binaries -- i.e. systems in which one of both components are black holes, neutron stars or white dwarfs. What has been overlooked, however, is that there is one important population of close binaries in which *neither* component is compact: the contact binaries. These are main-sequence binaries in which the two components are so close together that they share a single dumbbell-shaped envelope.
 
In this project, we will first construct the most up-to-date compilation of contact binaries and determine out to what distance our catalogue is complete (i.e. is not missing a significant number of objects). We will then carry out simulations to predict the expected gravitational wave signals, both from individual systems and from the overall population. This will allow us to determine which contact binaries may be detectable directly by LISA, and to what extent the GW background contributed by contact binaries will affect LISA's ability to detect other types of sources. Finally, we will also re-determine the space density of contact binaries, which is important astrophysically for our understanding of binary evolution.

Most stars are members of binary systems. This can dramatically alter their evolution, and many of the most interesting astrophysical systems - from Type Ia supernovae to the black-hole mergers observed by LIGO - only exist as products of binary evolution. In almost all of these special systems, one or both binary components are compact objects (white dwarfs, neutron stars or black holes).
 
Despite their importance, the evolution of such compact interacting binary stars remains poorly understood. For example, we still don't even know the dominant pathway(s) for producing Type Ia supernovae, even though we routinely use these objects as cosmological standard candles. The problem is that some of the most important physical processes for binary evolution - such as ''magnetic braking''' or the in-spiral associated with the common envelope phase - are extremely difficult to model accurately and self-consistently. Much of what we have learned about these systems has come from ''population synthesis''' studies, in which the properties of the detectable Galactic (or extragalactic) populations of these systems are predicted via numerical simulations.
 
In this project, we will draw on publicly available stellar evolution (MESA) and population synthesis (POSYDON) codes in order to develop a next-generation population synthesis data base and framework for compact binary systems. A unique feature of our work will be an emphasis on adopting and testing state-of-the-art physical and theoretical constraints on all key physical processes. For example, we will implement the latest magnetic braking laws suggested by observations of single and detached binary stars in order to determine whether they are consistent with the observed compact binary populations. We will then exploit this framework to predict the populations of white dwarf, neutron star and/or black hole binary populations that can be observed both electromagnetically and with gravitational wave detectors.
 
Depending on progress and the student's interest, we may also use modern machine learning techniques to develop a fast emulator for complex and expensive binary population synthesis calculations.

 
This PhD project explores how clouds have shaped Earth's climate stability through time, with a focus on their role in the planet's earliest atmospheres. Clouds strongly influence whether the Earth warms or cools, and their feedback remains one of the most significant uncertainties in climate science today. By investigating how these processes operated under very different atmospheric conditions in Earth's past, this project will provide valuable insights into both ancient climate transitions and the challenges of predicting future climate change.
 
The student will work at the frontier of climate modelling, developing innovative tools to represent cloud processes more realistically in numerical models. There will also be opportunities to apply machine learning to accelerate simulations in global climate models and to link high-resolution dynamics with larger-scale climate questions. This training will equip the student with highly transferable skills in numerical modelling, scientific computing, and data analysis, alongside expertise in atmospheric sciences and climate dynamics.
 
The project offers the chance to address big, open questions such as how Earth avoided global glaciation under a faint young Sun, and whether abrupt climate tipping points could emerge from cloud–atmosphere interactions. The successful candidate will join a vibrant, interdisciplinary research community, working closely with experts in climate modelling and palaeoclimate science (OASIS group, Palaeoclimate group).
 
The outcomes will advance understanding of Earth's climate resilience and instability, while providing the student with cutting-edge skills at the interface of mathematics, physics, and climate science.
 

 
The aurora (northern lights) is rich in structure and dynamics on multiple scales, but the mechanisms producing different features are not yet well understood. The University of Southampton operates a high- resolution imaging system on Svalbard, in the high Arctic, called ASK (Auroral Structure & Kinetics), designed for the study of rapid, fine scale < 10 km) aurora. Case studies of short individual events have advanced our understanding of fine scale aurora, but ASK has been operating throughout the dark winter seasons for several years and has produced hundreds of TB of data, the vast majority of which has not been studied or examined at all. We have therefore developed a citizen science project called Aurora Zoo, where volunteers classify and analyse ASK image sequences, enabling large statistical studies.
 
In this project you will use output from the Aurora Zoo to investigate the occurrence frequencies and conditions required for different auroral types, to constrain and advance theories for their associated generation mechanisms, and to better characterise and forecast the effects of the aurora on the atmosphere. Comparisons will be made with other data sets, in particular upstream solar wind data and other spacecraft measurements. Machine learning techniques may be applied to output from the Aurora Zoo to train a classifier capable of rapidly expanding the size of the statistical studies and classifying future data, possibly in near real time. The project may include the opportunity for fieldwork to the Arctic to calibrate the instrumentation and install an automated classification system.
 

 
“Psychedelic kaleidoscope” aurora is a type of highly dynamic fine scale (< 10 km) aurora, in which the shape remains approximately constant, but light flows through the structure in a pattern resembling phase fronts of a wave. The term was introduced in the Aurora Zoo citizen science project. So far little is known about this type of aurora, except that its dynamics have very small scale sizes which cannot be explained using existing theories, and the accelerated electrons producing the psychedelic kaleidoscope have high energies. Evidence has also been found that psychedelic kaleidoscope aurora can affect signals from Global Navigation Satellite System (e.g. GPS) spacecraft, potentially causing positioning errors or loss of lock. The few existing studies of psychedelic kaleidoscope have been done using Southampton’s Auroral Structure and Kinetics (ASK) high- resolution imager, on Svalbard in the high Arctic. We have also recently installed a new high-resolution imager in mainland northern Norway, underneath the main auroral oval, and saw psychedelic kaleidoscope in the first data.
 
In this project you will perform a detailed study of psychedelic kaleidoscope aurora, with the aim of understanding how it is formed and how it might affect the atmosphere and ionosphere. In particular, we will use data from spacecraft to determine properties of the associated auroral electron precipitation and electrodynamics, and we will use the ground-based optical data to form a dispersion relation for the wave associated with the auroral dynamics. A thorough comparison of aurora seen on Svalbard and mainland Norway may provide useful insights.
 

 
A complex system of electric currents flows through Earth's dynamic magnetosphere. A key part of this system is the resistive ionosphere, which acts as a load in the circuit. Although we have a large-scale picture of the current flow between the ionosphere and magnetosphere, observations show that the currents are much more structured and dynamic than this large-scale picture represents. At smaller scales our understanding becomes increasingly limited, but understanding the small-scale dynamics is key to understanding the physics of magnetosphere-ionosphere coupling and associated heating of the atmosphere. The aurora is a signature of charged particle precipitation into the polar atmosphere. These particles carry part of the magnetic field-aligned current between the magnetosphere and ionosphere, which closes through the ionosphere. This whole system varies rapidly in time and space.
 
This project will investigate the current flow and electric potential structure associated with auroral arcs of different types and widths. The project will use measurements of field-aligned currents made by spacecraft together with observations from Southampton's high- resolution auroral cameras. Ground-based radar and optical observations of the aurora can be used to measure temperatures and densities of the ionosphere and neutral atmosphere, which we can use to determine the ionospheric conductivity. One possible direction for the project is to use a 3D computational model to investigate how the ionospheric conductivity influences the structuring and dynamics of the current flow. The combination of state-of-the-art observations and computational modelling will be a powerful tool to advance our understanding of magnetosphere-ionosphere coupling and auroral dynamics.
 

 
Simulating the turbulent atmosphere of Jupiter is one of the great challenges in planetary science. The planet's rapid 10-hour rotation and unique conditions provide an exceptional laboratory for testing climate models under extreme regimes. Understanding the physics behind Jupiter's climate is key to explaining the striking diversity of atmospheric phenomena—multiple jets, giant storms, and complex turbulence—that shape its dynamic weather.
 
For decades, models of Jupiter's circulation have offered only qualitative comparisons to observations. These approaches often rely on assumptions tuned specifically to Jupiter and remain limited in scope. This PhD project aims to go beyond the current state of the art.
 
You will use OASIS, a cutting-edge computational model of planetary climates, to perform 3D simulations of Jupiter and Jupiter-like atmospheres at unprecedented levels of detail. The project will give you access to high-performance computing facilities, international collaborations, visits to partner research groups abroad, and advanced training in modern computational methods for planetary climate science.
 
Research directions include:
 
1) Investigating the role of cloud physics in driving large-scale storms and other dynamical features.
2) Analysing simulations at unprecedented spatial resolution.
3) Identifying the mechanisms behind the global distribution of ammonia.
4) Exploring how these mechanisms operate under higher levels of stellar irradiation.

 
You will join the new and growing Planetary Sciences group at the University of Southampton.
 

Supermassive black holes live in the centre of galaxies and grow by the accretion of gas from their surroundings. This process of black hole growth occurs throughout the evolution of the Universe and powers some of the most spectacular and energetic events we can observe: Active Galactic Nuclei.
 
In this project the student will investigate the physics behind how the gas is transported to the black hole. For this purpose, the student will analyse observations of gas in different phases (atomic, ionised and molecular) and apply physical models to study: 1) the mechanisms by which the gas is transported to the black hole, and 2) how much gas is transported in total, which controls the rate of growth of the black hole. The data will consist of observations from state-of-the-art telescopes, such as ALMA, and the student will have the opportunity to propose and plan for further observations.
 
The student will be part of a local research group at Southampton working on the multiwavelength properties of black holes and galaxies and will also be part of a wider international research team with opportunities for visits and training.
 

 
Type Ia supernovae, thermonuclear explosions of white dwarf stars, are astronomers' best tool to measure distances in the universe and trace its expansion across cosmic time. They were central to the discovery of cosmic acceleration and dark energy. New evidence now hints that dark energy may evolve, challenging Einstein's cosmological constant, the leading theory for its nature. Testing this possibility is a key goal of two next-generation surveys: the Rubin Observatory's Legacy Survey of Space and Time (LSST) and the Time Domain Extragalactic Survey (TiDES) on 4MOST. Both have recently achieved "first light" and will deliver tens of thousands of distant supernovae, samples 50 times larger than those available today.  
 
The University of Southampton plays a leading role in these collaborations. This project provides immediate access to the incoming data, placing you at the forefront as the first supernova discoveries are made. Working within an outstanding national and international team, you will investigate fundamental unknowns in supernova physics: their explosion mechanisms, progenitor systems, and links to the stars and dust in their host galaxies. By disentangling how these factors affect supernova luminosities, and thus inferred distances, the project will enable a new, state-of-the-art measurement of dark energy.  
 
As a PhD student, you will be among the first worldwide to explore these datasets. You will gain advanced skills in survey data analysis, statistical modelling, machine learning and AI, alongside experience in international teamwork and scientific communication - preparing you for careers in astrophysics, data science, and technology-driven industries.