Abstract

Supermassive black holes (SMBHs) are central to the evolution of massive, if not all, galaxies, driving a complex feedback cycle across an extraordinary range of scales, from the accretion disc to the cosmic web. Recent observations, particularly from the James Webb Space Telescope (JWST), have uncovered SMB Hs in lower-mass galaxies and at higher redshifts than predicted, challenging existing theoretical models. In this talk, I will present the latest advancements in modelling active galactic nuclei (AGN) feedback from SMB Hs across a range of galaxy masses and cosmic epochs. Using high-resolution cosmological zoom-in simulations, we have investigated the profound effects of AGN feedback on dwarf galaxy evolution, including outflows, quenching, and cusp-to-core transformations. Motivated by these results, we conducted large-volume cosmological simulations to explore whether variations in accretion and seeding models may explain JWST ’s overmassive early-universe SMB Hs. Despite these insights, significant uncertainties remain, particularly in accretion rates, as many models inherently suppress AGN activity in low-mass systems. To address these gaps, we have developed a novel, unified accretion disc model for SMB Hs, which self-consistently predicts their mass and spin evolution. This model integrates principles from the advection-dominated inflow-outflow solution (ADIOS) and state-of-the-art GR®MHD simulations. Validated through idealised simulations of single and binary SMB Hs, our findings demonstrate that the choice of accretion disc model critically affects predictions for electromagnetic counterparts and the spin properties of SMB Hs, which will be constrained by gravitational wave observatories like LISA and IPTA . This underscores the need for multi-scale SMBH models to fully leverage the potential of high-redshift and multi-messenger surveys.