AGN IN DWARF GALAXIES

Recent astronomical observations have uncovered actively-accreting central black holes in dwarf galaxies, which has prompted a likely paradigm shift as black holes had previously been neglected in dwarf galaxy models. We investigated whether these active galactic nuclei (AGN) could shut down the metamorphosis of gas into stars in dwarf galaxies since a ‘feedback mechanism’ is needed to prevent excessive star formation and match observations. Previously, supernovae had been assumed to be the main feedback mechanism in (massive) dwarfs, though whether supernova feedback alone can be effective enough is still controversial.

We modelled the impact of AGN activity in dwarf galaxies using simulations of individual galaxies as well as cosmological simulations based on the FABLE simulation suite (encompassing tens of thousands of galaxies, see Figure 1). We find that AGN feedback has a crucial effect on the large-scale gas outflows, significantly increasing both outflow velocities and temperatures. These boosted outflows can suppress fresh gas inflows from the cosmological environment and shut down star formation by depleting the gas reservoir of the host galaxy.

However, note that the AGN efficiency in dwarfs crucially depends on the assumed AGN accretion model, supernova feedback strength and black hole seeding mechanism(s). Next-generation galaxy formation models combined with future deep electromagnetic surveys (e.g. JWST, RST, Athena, Lynx, SKA) and gravitational-wave observatories (e.g. LISA, PTAs, AION) will allow us to break these degeneracies and elucidate the role of AGN feedback in dwarfs.

DWARF GALAXY FORMATION WITH COSMIC RAYS, RADIATION AND MAGNETIC FIELDS

Understanding star formation in dwarf galaxies has proven a persistent challenge for numerical simulations. While supernovae feedback is believed to be the main regulator of their gas conversion into stars, various properties of simulated galaxies differ from observations, such as their star formation histories as well as morphologies and kinematics. In order to construct more realistic simulated dwarf galaxies, accounting for additional baryonic physics has been advocated frequently (see also the research highlight above). In their absence, methods such as `calibrating’ stellar feedback have allowed simulations to reproduce the expected stellar masses at the cost of glossing over the physical fidelity.

Our team is investigating the role of radiative transfer, cosmic rays and magneto-hydrodynamics in high resolution, cosmological simulations of dwarf galaxies. Accounting for each of these additional components has different effects on the evolution and properties of galaxies, as depicted in Figure 2. Specifically, we find it is crucial to incorporate self-consistently radiative transfer effects which lead to dwarf’s sizes and kinematics being in much better agreement with current observational constraints. Furthermore, the inclusion of cosmic rays results in very different outflow properties both in terms of their kinematics and thermodynamics, which provides some of the key predictions for the next generation observational facilities such as JWST.

SIMULATING AGN JETS

Jets launched by active galactic nuclei (AGN) are believed to play a significant role in shaping the properties of massive galaxies and provide an energetically viable mechanism through which star formation can become quenched. Studying AGN feedback is made difficult, however, by the fact that it is an inherently multi-scale problem with significant coupling between processes on disparate scales. We, therefore, use numerical simulations and novel algorithms aimed to bridge vast spatial scales to study AGN feedback of which there are numerous open questions regarding its nature.

On galaxy cluster scales, it is well accepted that the vast amount of energy injected into the jet lobes is sufficient to offset the cooling losses within the intracluster medium (ICM) and prevent the formation of a cooling catastrophe. However, the processes by which this jet energy is effectively and largely isotopically communicated to the ICM is not well understood. Simulations carried out by our team, with AGN jets of an unprecedented resolution injected in a self-consistent cosmological environment, showed that the ICM motions or “weather” in addition to weak shocks are crucial to solving this problem. Through the cluster weather action, the jet lobes can be significantly moved and deformed, stirring them around the cluster, which ultimately leads to jet lobe disruption and effective energy transfer to the ICM. X-ray mock observations of our simulated clusters clearly show the so-called “X-ray cavities” and “X-ray bright rims” generated by supermassive black hole-driven jets. Remarkably, many of the features of these mock observations resemble those found in observations of real galaxy clusters as depicted in Figure 4.

One of the most promising processes by which AGN jets are launched is that of the Blandford-Znajek mechanism. Here, magnetic fields extract spin-energy from a Kerr black hole which is then used to launch jets parallel to the spin axis of the black hole. Motivated by this, we have developed a self-consistent sub-grid model for AGN accretion and feedback in the form of a Blandford-Znajek jet. Applying our model to simulations of the central regions of a typical Seyfert galaxy, we found that the outflow morphologies are highly dependent on the black hole spin magnitude and direction (see Figure 5) with the jets launched by misaligned black hole spins driving turbulent, multi-phase, quasi-bipolar outflows. This jet model also provides a useful tool to investigate the impact of AGN feedback in galaxy mergers. We will be able to investigate the link between electromagnetic observations and gravitational wave signals that would result if the black holes coalesce. This area of research is particularly important with the advent of multi-messenger astronomy and observatories such as Athena and LISA on the horizon.

MORPHOLOGIES OF SUPERMASSIVE BLACK HOLE MERGER HOSTS AND MULTIMESSENGER SIGNATURES

One of the key predictions of the hierarchical structure formation paradigm is that galaxies merge together to form more massive systems, with spectacular observational examples across a range of redshifts corroborating this picture. If the majority of massive galaxies host a supermassive black hole in their centre, it is expected that during galaxy mergers black holes will sink towards the merger remnant centre due to dynamical friction, and form a binary. The hardening rate and the ultimate fate of such a binary is currently unclear, but if embedded within certain types of stellar or gaseous environments it is possible that the black holes will merge themselves on a comparatively short timescale. Scarcity of supermassive black hole binary detections so far, together with low mass black hole mergers detections by ALIGO and Virgo, lends support that gravitational wave (GW) projects such as Laser Interferometer Space Antenna (LISA) and Pulsar Timing Arrays (PTAs) may, for the very first time, detect GWs from supermassive black hole mergers in the near future. Excitingly, in the case of supermassive black hole merger we may expect that there will be an electromagnetic counterpart detectable by e.g. JWST, Athena, MeerKAT and SKA, with the multi-messenger field expanding rapidly in the last couple of years.

Given their importance both for our understanding of the theory of galaxy formation as well as their upcoming detections from GW projects, we have investigated the mergers of supermassive black holes and their host galaxies, with a particular focus on the correlation between the black hole merger and the host galaxy morphology. Using the Illustris and IllustrisTNG simulations, we look at the galaxies which host black hole mergers, and find that the typical host galaxies show a range of disturbed morphologies, consistent with what we expect following a recent galaxy merger (see Figure 6).

Host galaxies remain morphologically disturbed for ~300-500 million years after the black holes form a binary, with the strongest effect found in mergers between high-mass black holes with comparable masses. While initially promising for multi-messenger astronomy, this timescale of several hundred million years is comparable to the expected time between black hole binary formation and the final coalescence and emission of GWs, suggesting that electromagnetic follow-ups to GW detections may not be capable of observing merging morphologies of the host galaxies. However, both this timescale and the expected black hole growth during the merger itself need to be much better constrained by future cosmological simulations to make robust predictions for LISA, PTA and multi-messenger science.

FABLE: Feedback Acting on Baryons in Large-scale Environments

The Feedback Acting on Baryons in Large-scale Environments (FABLE) simulations follow the formation and evolution of galaxies across a diverse range of environments; from the sparsely-populated field environment to small groups of just a handful of galaxies to rich clusters containing hundreds of member galaxies (see Figure 7). Each simulation in the FABLE suite evolves a patch of a mock universe from shortly after the Big Bang to the present-day using the state-of-the-art numerical code AREPO, which incorporates a comprehensive set of models for the physical processes that drive galaxy formation. Our diverse array of high-resolution simulations have been performed on DiRAC high-performance computing facilities and required several million CPU hours to complete. The FABLE simulations simultaneously produce realistic galaxies, groups and clusters, which allows us to explore in detail the astrophysical processes at play on this vast range of scales with the ultimate goal of constraining the cosmology of our Universe to a high precision from the wealth of upcoming observations of galaxy groups and clusters.