Galaxies, Active Galactic Nuclei and Cosmology

Supervisory team: Martin Krause, Rob Yates

The buildup of the chemical elements in the chart of nuclei is a fascinating story that involves a number of different factors: Most of the elements heavier than helium have been formed in stars and during different kinds of stellar explosions. Direct evidence for this process comes from the measurements of radioactive decay lines shortly after an explosion and also from radioactive decay of elements with longer half lives that accumulate in the interstellar medium. 26Al, ejected from a variety of sources, and 60Fe, ejected from supernovae, only, both have half lives of the order of a million years and have been found in the Milky Way with gamma ray observations. The observations can, however, only be interpreted with the help of computer simulations that predict sky distributions for different assumptions about how the stars shed the radioactive nuclei and how the ejecta evolve in the dynamic interstellar medium. Our group is active in such simulations. The project consists in running new simulations with the existing code, and in particular, explore the differences between the expected sky distributions for 26Al and 60Fe. The project is particularly timely since NASA's COSI mission (recently confirmed for launch in ≈ 2027) is expected to measure for the first time the distribution of radioactive 60Fe in the Milky Way.

Knowledge of hydrodynamics and some prior experience with simulation codes are beneficial, but can also be acquired at the beginning of the project.

Supervisory team: Chiaki Kobayashi, Emma Curtis-Lake

The universe started from the Big Bang, stars form and die, producing heavier elements in the universe. Some of the heavy elements are in a solid state - dust. Therefore, elemental abundance of stars and galaxies, as well as the composition of dust in galaxies, can give a stringent constraint on the formation and evolution of galaxies. With the James Webb Space Telescope, elemental abundances are estimated even at very distant galaxies (redshift 8-10). The student will develop self-consistent hydrodynamical simulations including elements and dust, based on our own simulation code. Supercomputer facilities, local LINUX cluster and national supercomputer facility (DiRAC), will also be provided for producing world-class simulation data. The simulations will follow formation of stars, death of stars (supernovae), production of elements (nucleosynthesis), growth of black holes, and feedback from supermassive black holes. By comparing observational data, the student will investigate how galaxies form and evolve in a cosmological context.

Supervisory team: Jan Forbrich, Elias Brinks

Thermal centimetric bremsstrahlung continuum emission from individual HII regions is one of the most direct and extinction-free tracers of the star formation rate (SFR) in galaxies. The VLA upgrade now enables large-scale studies of such emission in nearby galaxies. After a targeted survey of the known Giant Molecular Associations in M31, we have conducted an unbiased C-band (4-8 GHz) survey of the entire M31 galaxy, using the Karl G. Jansky Very Large Array. With a synthesized beam size of 13pc, this experiment will probe all radio sources on sub-cloud  scales and help identify not only candidate HII regions but also supernova remnants (SNRs) throughout the galaxy  at unprecedented sensitivity. SNRs in particular may no longer be associated with GMAs while tracing previous  generations of star formation. These data will be analyzed in conjunction with our continuing SMA and optical spectroscopic surveys, targeting star-forming regions in M31 with the goal of studying star formation scaling  relations (and more!). In this project, you will learn about radio interferometry using a key facility, and you will explore the radio sources of an entire external galaxy (a famous one at that) in a rich multi-wavelength context.

Supervisory team: Sugata Kaviraj, Aaron Watkins

Our current understanding of the Universe is dominated by bright objects (e.g. massive galaxies like the Milky Way), because such systems are brighter than the detection thresholds of past large observational surveys (e.g. the SDSS). However, the majority of stars in the Universe actually reside in the faint or ‘low surface brightness’ regime, i.e. in objects and structures that are much fainter than the detection limits of past surveys. This regime contains all dwarf (low-mass) galaxies which dominate the galaxy number density, making them critical to our understanding of galaxy evolution. It also includes faint tidal debris created by galaxy mergers, which are key to understanding how gravity, the predominant force in the Universe, shapes galaxy evolution over cosmic time. Put simply, a complete understanding of how the Universe evolves is not possible without a detailed comprehension of the low surface brightness regime.

Astrophysics is currently entering a revolutionary era of new surveys, which not only have large areas but are also incredibly deep. In particular, the Legacy Survey of Space and Time (LSST) and the Subaru Strategic Program from the Hyper Suprime-Cam telescope, are poised to transform our understanding of the Universe, by providing images that are more than 100 times deeper than those from previous surveys. These images will enable us to perform detailed studies of the low surface brightness Universe for the first time.

This project will combine state-of-the-art data from these surveys with in-house cosmological simulations (e.g. NewHorizon) and advanced machine-learning techniques we have developed (e.g. Martin et al. 2020), to perform the first statistical studies of the low-surface-brightness Universe. The project will map the properties of dwarf galaxies in unprecedented detail, over at least half the lifetime of the Universe and quantify the role of key processes like galaxy merging in driving star-formation, black-hole growth and morphological transformation in galaxies over cosmic time.

The student will collaborate closely (through visits and conference trips) with colleagues in Paris, Oxford and a worldwide network of scientists within the international LSST project (in which our team members hold several leadership roles). The project will give the student an excellent skillset in astronomical observation, theory and machine-learning that is well-aligned with this new era of Big Data astronomy.

Supervisory team: Rafael S. de Souza, Carolyn Devereux

Estimating physical parameters of galaxies is a crucial task in the realm of extragalactic surveys. However, as astronomical surveys continue to grow in scale and complexity, the need for computationally-efficient methods for inferring these galaxy properties becomes increasingly urgent. Surveys like the Euclid survey, and the Vera C. Rubin Observatory are expected to generate an unprecedented volume of observations, rendering current methods, often based on Markov-Chain Monte Carlo (MCMC) techniques and forward modelling, impractical.

In this PhD research project, we seek a motivated candidate to tackle this challenge by harnessing the power of simulation-based inference (SBI). Recent advancements in SBI have demonstrated the potential for orders of magnitude faster parameter inference compared to traditional MCMC approaches. The project aims to capitalise on this potential by integrating existing forward-modelling libraries and spectral energy distribution (SED) modelling tools into the SBI framework.

Project Objectives:

• Explore and adapt state-of-the-art SBI techniques: The candidate will delve into the latest developments in simulation-based inference, gaining expertise in methodologies that can efficiently sample parameter spaces without requiring explicit probability density evaluations.
• Integration of SED modelling tools: The candidate will work on seamlessly integrating existing SED modelling libraries into the SBI framework, creating an implicit statistical model that is compatible with large-scale galaxy property estimation.
• Develop an efficient galaxy parameter estimation pipeline: Leveraging the insights gained from SBI and SED modelling, the candidate will design and implement a computational pipeline for estimating physical parameters of galaxies from photometric data.

Supervisory team: Kristen Coppin, Jim Geach

Over half of the star formation energy generation in the Universe is extincted at optical wavelengths and enshrouded by dust which absorbs and re-radiates the starlight in the far-infrared/submm; and the sub-mm and mm atmospheric windows allow us to access the redshifted far-infrared emission from this obscured or ``hidden’’ side of galaxy formation and evolution.  The James Clerk Maxwell Telescope (JCMT) Cosmology Legacy Survey (S2CLS; Geach et al. 2017) and its extension via the SCUBA-2 COSMOS survey (S2COSMOS; Simpson et al. 2017) and now S2XLS (PI Geach) and STUDIES (Wang et al. 2017) are the largest and most sensitive and ambitious single-dish surveys at 850 and 450 micron (in the submm wavebands) ever conducted.  In addition, the 50-m Large Millimeter Telescope in Mexico will be conducting unique and transformative imaging of the sky at millimeter wavelengths through a series of public Legacy Surveys (Ultra-Deep and Large Scale Structure surveys in particular) starting in late 2022 using the new TolTEC camera.  These unprecedented legacy surveys have been yielding thousands of high-redshift galaxies selected in the sub-mm/mm wavebands - providing an order-of-magnitude improvement in the sample sizes of previous surveys at these wavelengths!

With so much data now in-hand there are several possible projects that could be carved out using a combination of these legacy (sub)mm surveys with existing ancillary multi-wavelength data to make progress on a key outstanding question in galaxy evolution: How are dust and metals built up in massive galaxies over cosmic time? Some key science that could be explored with these data sets by a keen student include (for example): 1) Measuring the morphologies and Spectra Energy Distributions of the dustiest (sub)mm galaxies identified in the surveys above with existing public JWST CEERS data; 2) locating and probing the high-redshift tail of the distribution of (sub)mm galaxies (via new mm observations with ToLTEC); and 3) exploring new parameter space on the dust content, obscuration fraction, and gas content in galaxies (via (sub)mm observations) as a function of mass out to much higher-redshift than has previously been explored.  The project can be tailored to some degree to match the student’s interests and skill set and available data.

We are also involved in ongoing efforts to perform detailed follow-up of these high-z submm-detected sources at higher resolution with the Atacama Large Millimetre Array (ALMA) situated at 5000m on the Chajnantor plateau in Chile.  It is envisaged that the findings of this work will feed naturally into new ALMA and other telescope proposals, such as the James Webb Space Telescope (JWST).

Supervisory team: Rafael S. de Souza, Ana Chies (external)

Globular cluster (GC) systems serve as pivotal proxies for tracing the assembly histories of their host galaxies. With mean ages surpassing 10 billion years, GCs are essentially the ancient fossils that chronicle the evolutionary pathways of galaxies and their surrounding environments. Numerous studies underscore the crucial role of environment in shaping the formation and subsequent evolution of GC systems. Consequently, GCs have been frequently employed in literature as dynamic markers for the galactic halo.

Despite ongoing efforts, a holistic understanding of how various populations of GCs formed within their host galaxies is yet to be fully realised. One significant hurdle  is acquiring adequate spectroscopic samples that can reliably inform us about stellar population  properties such as age and metallicity. Observations typically focus on the outskirts of the galaxies due to the  bright background of the galaxy's core. This often results in skewed and incomplete datasets. Recent breakthroughs in the field have pivoted towards multi-band photometry, which allows for the swift acquisition of substantial candidate samples, albeit with the drawback of increased contaminants.

The MUSE archive offers an extensive database of high-quality IFU cubes of nearby galaxies. However, the GC spectra are entangled together with the integrated light of the galaxy. By deriving stellar population and kinematical parameters, we can determine the assembly history of the GC systems and provide further complementary insight into the history of their host galaxies.

Project Objectives:

• Spectral extraction: Harness the potential of MUSE data cubes to systematically extract optimal spectra suitable for analyzing the stellar population and kinematics of GCs. While prior endeavors have sought to retrieve GC or nuclear star cluster spectra from MUSE data cubes, the spectra samples curated were limited. Only spectra with high signal-to-noise ratios were deemed usable. This project aspires to create an expansive database encompassing GC spectra across galaxies of varied shapes and environmental contexts.
• Analysis and Insights: Delve deep into the individual radial velocities of GCs and undertake a stellar population investigation to derive crucial parameters such as ages, metallicities, and abundances. Further, the project seeks to elucidate how the age-metallicity relationship within GC systems is influenced by environmental variables and determine if there exists an environmental effect affecting the colour-metallicity associations of GC systems.
• Data Dissemination: Produce a value-added catalogue (VAC) by processing the GC spectra using tools like starlight or Ppxf. This database, encompassing both the spectra and the VAC, will be made accessible to the wider research community.

Supervisory team: Gulay Gurkan, Dan Smith

Supermassive black holes (SMBHs) in the centre of massive galaxies can produce powerful radio jets and/or radiative energy (i.e. feedback). While observational studies suggest that feedback from powerful SMBHs have substantial effects on their host galaxies, such feedback mechanism is required in numerical simulations of galaxies in order to match the observations. Although the significance of SMBH feedback is well investigated in strong sources (statistically or a single source) we have only scratched the surface of the effects of this feedback in low-luminosity (i.e., radio-quiet active) galaxies.  The new-generation deep radio surveys from the Square Kilometer Array (SKA) pathfinder ASKAP on GAMA fields with excellent multi-wavelength coverage enabled us to select a sample of radio-quiet active galaxies at lower radio luminosities than previously probed. Preliminary results suggest that these sources present compact relativistic plasma released by SMBHs which may play a crucial role in the fate of their host galaxies.

This PhD project will make use of existing data sets to decipher the connection between SMBHs and their host galaxies focusing on radio-quiet active galaxies. There will be opportunities to support the available data sets with further follow-up observations from other telescopes and/or access to the Middle Ages Galaxy Properties with Integral field spectroscopy (MAGPI) survey data. Part of this PhD project will focus on addressing one of the prominent questions in astrophysics: Are the powerful SMBHs and their radio-quiet counterparts fundamentally different? For this the student will use the state-of-the-art radio data from LOw Frequency Array (LOFAR), optical spectra from Sloan Digital Sky Survey (SDSS) and machine learning (ML) methods. There will be opportunities to visit beautiful Bologna in Italy and collaborate with Dr Isabella Prandoni (IRA-INAF) who is one of the leaders in the field. More importantly, this project will allow the student to participate in important collaborations, and establish methods for other fields with radio data, in preparation for the biggest radio astronomy project the SKA. With this project the student will have the opportunity to develop a set of crucial skills (in data fitting, using ML methods, composing an observing proposal, and radio data analysis) easily transferable to other fields/subjects.

Supervisory team: Gulay Gurkan, Martin Hardcastle

The radiative and jet power in active galactic nuclei (AGN or active supermassive black holes) is generated by accretion of material on to supermassive galactic-centre black holes. For quasars, where the radiative power is by definition very high, objects with high radio luminosities form 10 per cent of the population, although it is not clear whether this is a stable phase. Traditionally, quasars with high radio luminosities have been thought to present jets with edge-brightened morphology (or FR II) due to the limitations of previous radio surveys. The LOw Frequency ARray (LOFAR) Two-metre Sky Survey (LoTSS; Shimwell et al. 2022) with its unprecedented sensitivity and resolution covering wide sky areas has enabled the first systematic selection and investigation of quasars with core-brightened morphology (or FR I). While these preliminary results suggest that jet power and/or galaxy environment may be the key to decipher why FRI-type quasars are not as common as FRII quasars, an in depth investigation of this phenomenon is essential. With this aim we have been granted state-of-the-art high resolution radio observations with LOFAR (angular resolution>=0.1arcsec at 144 MHz) and e-MERLIN (angular resolution >=0.1arcsec at 1.4 GHz). These observations were recently carried out and ready for reduction and analysis.

The student will analyse these key data sets to address fundamental questions about SMBHs such as; how does the FRI quasar population compare to FRI radio galaxies in terms of their jet characteristics?; what are the differences between FRI and FRII quasar jets; do these differ in different accretion mode systems? The aforementioned data sets will allow the student to investigate spectral properties of kpc scale FRI jets in quasars in a novel way. It is worth noting that such observations of these exotic objects have never been done before so the results of these observations will not only be important for AGN research but will also define cases for the future VLBI studies such as with the Square Kilometre Array. With this project the student will have the opportunity to develop a set of crucial skills (in understanding and analysis of radio data, using high-performance computing clusters and AGN jet models) easily transferable to other fields/subjects. The student will also have the opportunity to participate in LOFAR collaboration.

Interest and/or experience in high resolution radio interferometry is desirable.

Supervisory team: William Alston, Dominic Walton

Accreting black holes are unique sources of some of the most extreme physics in the universe. They are a testbed for understanding general relativity and high-energy physics processes, having implications for all areas of science and space exploration. Key to solving many unanswered questions in contemporary astrophysics are precise measurements of the black hole mass, spin and detailed knowledge of the processes involved in the accretion and ejection of matter around the black hole. The accretion of matter is non-static, so the application of time series methods to the data provides important insight into the geometry and physical processes involved – something which spectral information cannot achieve alone.

The recent discovery of short time delays – or reverberation - between the intrinsic X-ray emission and the accretion disc reprocessing component around the black hole has made it possible to spatially map the close regions around black holes. This spatial information is measured by modelling the response of the disc to a flash of X-rays coming from the corona. This allows us to decode the geometry, relativistic effects and light bending in the extremely curved spacetime. Recently, we have modelled the reverberation signal in a nearby supermassive black hole as the system evolves over several months (Alston et al 2020). This revealed for the first time a dynamic picture of material around the event horizon, turning our previously static picture of the inner accretion flow into a movie. This powerful method means we are able to resolve structures to an accuracy of ~1 gravitational radii (Rg) - equivalent to measuring structures the size of the sun at a distance of over 1 billion light years.

This project involves the development and application of time series analysis methods, such as Fourier analysis, Gaussian Processes, and generative machine learning methods, to a wide variety of X-ray and optical data on accreting black holes. Together with the development and application of theoretical models, including ray-tracing in a GR framework, we will make the most accurate measurements of black hole mass and spin to date in a large sample of sources. The understanding we will gain from sources in the local universe will allow us to push this method to higher redshift and to test models of black hole growth over cosmic timescales.

Supervisory team: Carolyn Devereux, Rafael S. de Souza

Dark matter is distributed across the universe in a large scale structure known as the Cosmic Web; a map of voids, sheets, filaments and knots. It influences how the universe evolves and how and where galaxies form. A recent paper (Nguyen et al 2023) has shown that the growth of structure does not match the expansion of the universe as we would expect from the 𝛬CDM model. This result adds to the growing number of anomalies in our measurements of the universe.

Work by Buncher and Kind (2020) on using machine learning to classify a simulated version of the cosmic web has demonstrated the usefulness of this approach. This project uses machine learning classification and visualisation techniques on observational survey data to explore how large scale structure has grown. It is possible to compare the early dark matter structure, for example, using the Planck gravitational lensing of the CMB, to the late universe dark matter structure, for example, using galaxy weak lensing maps. The analysis of structure growth could lead to insights into recent cosmological tensions (H_o and S8) as well as how dark energy has changed over time.

With surveys from telescopes such as LOFAR, Euclid and the Vera Rubin Observatory covering large areas of the sky it is a good time to be exploring the large scale structure growth. Understanding the growth of large scale structure, and the formation of the cosmic web, is important to understanding the universe. This project sits within a department that has a large data science and machine learning activity and there is opportunity to develop worldwide links on the project.

Supervisory team: Carolyn Devereux, Alyssa Drake

JWST data has already shown us that galaxies in the early universe can behave differently than galaxies today. Melia (2023) recently studied JWST high redshift galaxies and found that early galaxies were brighter than expected. By assuming that brighter galaxies mean bigger mass, this result questions the accuracy of the 𝛬CDM model. This problem was first observed by Steinhardt et al (2016) and named it the ‘Impossibly Early Galaxy Problem’.

A recent paper (Sun et al 2023) has found that star forming bursts are more common in early galaxies and so could explain the early galaxy problem. More work is required to understand the properties of early galaxies. The aim of this project is to explore high redshift galaxies and find out about galaxy evolution in the early universe and whether it fits the 𝛬CDM model.

This project will use data from JWST, Weave LOFAR and other surveys to explore high redshift galaxies (of which there are more being detected and at higher redshifts) to determine how their properties and formation are different than galaxies today. The project will investigate whether the observed differences can be explained by the 𝛬CDM model and if not what could explain the discrepancies. This could lead to important insights into our understanding of the universe.

Supervisory team: Alyssa Drake, Dan Smith

Quasars are the most luminous class of active galactic nuclei (AGN) thanks to the energy released by their rapidly-accreting supermassive black holes. The extreme luminosity of these objects means they serve as unique probes of the early Universe, in particular allowing studies of unprecedented detail inside the transformative period of cosmic history known as the epoch of reionisation (EoR). This PhD project aims to unravel the mysteries of the EoR by studying the first quasars and their surrounding galaxies, as well as the evolution of the AGN population across cosmic time. In the first year, using MUSE observations at z = 6, the post-holder will search for Lyman alpha emission from galaxies in proximity to high-redshift quasars, shedding light on the physical properties of objects living close to these cosmic powerhouses. The post-holder will later have access to state-of-the-art spectroscopy from the WEAVE-LOFAR survey (led at UH), shifting focus to the evolution of the AGN population. Selection of sources with LOFAR via their radio emission offers a unique perspective on AGN activity, allowing for a homogeneous selection of objects regardless of their dust content, and in combination with complete optical spectroscopy from WEAVE serves as an immensely powerful toolbox to trace the growth of obscured supermassive black holes to the present day. This comprehensive approach will allow us to understand the evolution of the most powerful objects in the Universe, and gain insights into the birth of the first supermassive black holes.