Every supermassive black hole at the centre of a galaxy can blow a wind from its accretion disk, but no one knows exactly how that wind works or how strong it really is. This matters because those winds are thought to be the main way a black hole talks back to its host galaxy—pushing gas away, shutting down star formation, and shaping how galaxies grow over cosmic time. Current computer simulations of galaxy evolution treat this feedback with crude approximations, ignoring how it changes with black hole mass, how fast it is feeding, or the chemical makeup of the gas. The team has already shown that earlier calculations for similar winds from white dwarfs overestimated the wind’s power by a factor of roughly 2,000, because they simplified the radiation physics. If this project succeeds, it will replace those guesses with a physically grounded recipe for galaxy simulations. That will let astronomers model how galaxies and black holes co-evolve with far greater accuracy. The work is fundamental science—it will not change a smartphone or a power grid tomorrow—but understanding how galaxies form is the bedrock of modern cosmology, and past fundamental advances in radiation hydrodynamics have fed into everything from fusion research to medical imaging.
View original technical description
All massive galaxies harbour a central supermassive black hole (SMBH). If the SMBH is fed at a sufficiently high rate, it can become extremely luminous. Many aspects of the accretion process that powers such active galactic nuclei (AGN) remain uncertain. Importantly, even though accretion (usually via a disc) causes SMBHs to gain mass, the process is inevitably accompanied by mass loss, often in the form of disk winds. However, the physical properties of these outflows – and even their basic driving mechanism – remain extremely poorly understood. This is a problem for two reasons. First, these outflows can significantly affect observations, impeding the construction of a reliable physical picture of AGN. Second, these disk winds are critical to our understanding of galaxy formation and evolution, since they provide a key “feedback” mechanism by which an SMBH can affect its host galaxy. In the absence of a sound physical understanding of these outflows, cosmological simulations currently incorporate this feedback via simplistic recipes, without accounting for variations with SMBH mass, accretion rate and metallicity. The fundamental obstacle to understanding AGN disk winds is that radiation pressure on bound electrons must be important: simulating such line-driven outflows requires complex radiation-hydrodynamics (RHD), in a regime where none of the standard approximations are valid. In particular, the treatment of ionization and radiation is critical to correctly estimate the wind driving forces. Over the last few years, we have developed a unique, state-of-the-art RHD code that implements the crucial physics. Our first benchmark simulations – for line-driven disk winds from accreting white dwarfs – have already revealed a huge problem with earlier, more approximate calculations: due to their simplified treatment of radiation, the kinetic wind power was overestimated by a factor of ? 2000. We now propose to tackle line-driving in AGN. Specifically, we will determine the parameter space, physical properties and observational signatures of line- driven disk winds by carrying out RHD simulations accounting correctly for ionization and radiative transfer for a wide range of AGN properties; implement additional physical processes that may affect mass loss in AGN, such as clumping, X-ray irradiation and large-scale magnetic fields; construct a new, physically motivated feedback recipe for cosmological simulations, accounting for the dependence on SMBH mass, accretion rate and metallicity. The proposed work will significantly improve our current physical picture of AGN. It will also drive progress in understanding the co-evolution of SMBHs and their host galaxies.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know