I work with observations of black holes made with state-of-the-art X-ray observatories, including XMM-Newton, Chandra and NuSTAR. I am developing novel data analysis techniques and computer simulations of how X-rays travel in the curved spacetime around black holes that enable theory to be directly related to observations to connect the extreme physics happening on small scales around black holes to fundamental questions in cosmology and the formation of the Universe, including:

• How is energy released from material spiralling into black holes to power the brightest objects in the Universe?

• What is the structure of the X-ray emitting corona and how is energy injected into it from the corona?

• How are jets launched by black holes and how are they related to the corona?

• What determines whether a jet can be launched?

• How does matter behave in the strong gravitational fields around black holes?

• Does Einstein's theory of General Relativity accurately describe the strong field regime around black holes?

• How can we detect signatures of General Relativity and strong light bending in X-ray observations of black holes?

• What controls the lifecycle of the AGN and governed the growth of black holes and their energy output over the history of the Universe?

X-ray Reflection & Reverberation

The X-rays emitted from the corona illuminate the accreting material in its final moments before it plunges into the black hole. My work aims to understand how we can use the reflection and reverberation of X-rays to map the structure of the corona and the accretion flow, right down to the innermost stable orbit and the event horizon. X-rays are 'reflected' from the accretion disc through the processes of Compton scattering, bremsstrahlung, photoelectric absorption and the emission of fluorescent lines, imprinting a number of characteristic features on the spectrum of the reflected X-rays. Particularly prominent is the K-alpha fluorescence line of iron, seen at 6.4keV.

The energy at which we measure the reflected X-rays is altered by a combination of Doppler shifts from the orbital motion of the reflecting material in the disc (material at the innermost stable orbit is travelling at half the speed of light!), relativistic beaming and redshifts due to the strong gravitational field in the proximity of the black hole. This means the emission lines we see from the accretion disc are broadened to a characteristic shape with a blueshifted peak and extended, redshifted 'wing' to low energies arising from the innermost parts of the accretion disc, closest to the black hole.

Most importatly, the relativistic effects shifting the energy of the reflected X-rays vary across the surface of the accretion disc. This means that the shift in energy of a reflected spectral feature contains information about exactly where on the disc that reflection arose. I have exploted this to devised a novel method to reconstruct the reflected emission from each part of the accretion disc from the profile of the relativistically broadened iron K-alpha line seen in the spectrum, to determine the emissivity profile of the accretion disc (the variation in reflected flux across the disc) directly from observations (Wilkins & Fabian 2011). This essentially allows us to resolve the reflection across the surface of the disc which is observed only as an unresolved point source on the sky.

The X-ray emission from the corona is extremely variable. Recent studies of this variability have added a further dimension to the study of accreting black holes. The reflected X-rays from the accretion disc will respond to changes in the luminosity of the corona, however a time delay is seen between variations in the continuum and in the reflection, corresponding to the additional light travel time between the corona and the disc. We see variations in the corona’s luminosity echo or reverberate off of the disc.

The energy shifts we measure in emission line photons reveal the locations at which they were emitted (or ‘reflected’) on the disc, while the reverberation time lag between the continuum and reflection at this energy reveals the distance of this part of the disc from the powerful corona. Combining these measurements lets us build up a 3D picture of the extreme environment just outside the event horizon of a black hole.

In 2014, I co-authored a review article on X-ray reverberation.

In 2021, observing the supermassive black hole in the galaxy I Zwicky 1 simultaneously using the X-ray telescopes NuSTAR and XMM-Newton, we caught a series of short, extremely bright X-ray flares. We were able to observe the echoes of these bright flares from the accretion disc. In particular, we were able to observe the echo of the flares off of material in the accretion disc that should be hidden behind the black hole from our point of view. We are able to observe these echoes because the light emitted from material behind the black hole is bent around the black hole in the strong gravitational field. These echoes were seen as additional short flashes of emission as the flares faded away, and we can identify their origin behind the black hole based how the energy of these photons is shifted and their time delays with respect to the original flare (Wilkins et al. 2021). Observing these echoes confirms the prediction of general relativity that light is bent around massive objects holds true even in the extreme gravitational field just outside the event horizon of a black hole, and lets us see general relativity and strong light bending in action.

Ray Tracing Simulations and Computational Techniques

In order to translate measurements of X-ray reflection and reverberation into an understanding of the physics of the extreme environment just outside the event horizon, I have developed a suite of general relativistic ray tracing simulations that model the emission of X-rays from the corona, the light-bending and relativistic effects they experience in the strong gravitational field around the black hole, their reflection from the accretion disc and their observation. These simulations are the key to translate measurements of energy shifts and time delays into a map of the inner regions, and precise measurements of the accretion disc and corona. These simulations are built from the ground-up, starting with the simplest possible models to discover which of the observed phenomena are inescapable consequences of observing X-ray emission from around a black hole. The complexity is gradually built up to let us constrain the structure and geometry of the X-ray emitting corona and the physical processes therein from real observations.

Tracing rays in the simulation from the accretion disc back to their site of emission, we are able to measure the location and spatial extent of the corona to better understand how it is formed and energised. Not only that, but we can use ray tracing simulations to predict signatures of strong gravity, as predicted by Einstein’s theory of General Relativity, that we can detect in X-ray observations of black holes. We can predict how X-rays reverberate off of material in the plunging region after it has passed the innermost stable circular orbit (Wilkins et al. 2020a). General Relativity tells us that this orbit is the closest to the black hole that material is able to remain stably circling the black hole. Beyond this point, the immense gravity of the black hole overwhelms the orbit and it must plunge inwards, through the event horizon. The strong gravity predicted by General Relativity around the black hole also causes some of the X-ray that have been reflected from the disc to be bent back to disc to be reflected a second time. This so-called returning radiation changes the spectrum and the time delays of the X-ray that reverberate off the innermost parts of the disc (Wilkins et al. 2020b). Detecting the plunging region and testing the presence of the innermost stable circular orbit, and finding those X-rays that have been returned to the disc to be reflected multiple times represent unique tests of General Relativity in the strong gravity regime just outside the event horizon of the black hole and give us a glimpse of the science we will be able to achieve with the next generation X-ray observatories.

I am particularly interested in high performance computing techniques. I have developed ray tracing codes and spectral/timing analysis routines for GPUs using the NVIDIA CUDA architecture, offering speed increases by factors of several hundred over conventional computing techniques.

Alongside ray tracing simulations, I am developing new data analysis techniques to relate theoretical predictions to measurements of X-ray reverberation. These echoes of the X-ray continuum are most commonly measured around relatively modest supermassive black holes (millions to tens of millions of Solar masses) in the local Universe. I have developed a technique to measure X-ray reverberation on longer timescales than can be accessed using conventional techniques. Using Gaussian processes to construct a statistical description of the X-ray variability, it is possible to measure time delays spanning multiple consecutive observations (with gaps). This enabled the detection of X-ray reverberation in 3C120, the first time reverberation was detected in a radio-loud AGN with a much more massive black hole (Wilkins 2019).