Cosmological simulations of black hole growth: AGN luminosities and downsizing

Abstract: In this study, we present a detailed, statistical analysis of black hole growth and the evolution of active galactic nuclei (AGN) using cosmological hydrodynamic simulations run down to $z=0$. The simulations self-consistently follow radiative cooling, star formation, metal enrichment, black hole growth and associated feedback processes from both supernovae typeII/Ia and AGN. We consider two simulation runs, one with a large co-moving volume of $(500\ \mathrm{Mpc})^3$ and one with a smaller volume of $(68\ \mathrm{Mpc})^3$ but with a by a factor of almost 20 higher mass resolution. Consistently with previous results, our simulations can widely match observed black hole properties of the local Universe. Furthermore, our simulations can successfully reproduce the evolution of the bolometric AGN luminosity function for both the low-luminosity and the high-luminosity end up to $z=3.0$. In addition, the smaller but higher resolution run is able to match the observational data of the low bolometric luminosity end at higher redshifts $z=3-4$. We also perform a direct comparison with the observed soft and hard X-ray luminosity functions of AGN, including an empirical correction for a torus-level obscuration, and find a similarly good agreement. These results nicely demonstrate that the observed "anti-hierarchical" trend in the AGN number density evolution (i.e. the number densities of luminous AGN peak at higher redshifts than those of faint AGN) is self-consistently predicted by our simulations. Implications of this downsizing behaviour on active black holes, their masses and Eddington-ratios are discussed. Overall, the downsizing behaviour in the AGN number density as a function of redshift can be mainly attributed to the evolution of the gas density in the resolved vicinity of a (massive) black hole. (shortened)

• The paper demonstrates that advanced cosmological simulations merge large-volume and high-resolution models to accurately reproduce AGN luminosity functions and naturally capture black hole downsizing.
• The paper employs detailed treatments of radiative cooling, star formation, and redshift-dependent dust obscuration to align soft and hard X-ray AGN observations with simulation outputs.
• The paper finds that declining Eddington ratios over cosmic time elucidate the gradual drop in luminous AGN, providing critical insights into black hole growth dynamics.

Cosmological Simulations of Black Hole Growth: A Comprehensive Analysis

The paper by Hirschmann et al. explores the complex landscape of black hole (BH) growth and active galactic nuclei (AGN) evolution as delineated through cosmological hydrodynamic simulations. It emphasizes understanding AGN luminosities and the anti-hierarchical growth or downsizing of black holes over cosmic time. The paper serves to corroborate the feasibility of the hierarchical structure formation model within the ΛCDM paradigm to naturally account for the observed downsizing.

The authors employ two distinct simulation sets from the Magneticum Pathfinder simulation suite: one with a large co-moving volume of (500 Mpc)$^3$ and another with a smaller, more finely resolved volume of (68 Mpc)$^3$. These simulations incorporate a sophisticated treatment of radiative cooling, star formation, and feedback processes from supernovae and AGN, implemented via the Gadget3 code—an enhanced version of the widely utilized SPH code GADGET-2.

Key Findings

1. Black Hole Growth and AGN Luminosity Function: The simulations align well with the observed bolometric AGN luminosity function, particularly up to $z=3$. Importantly, the study effectively merges the outputs from the large and small-volume simulations to address resolution constraints and successfully models moderately luminous AGN up to $z=4-5$. The work extends previous simulations by accurately predicting the high-luminosity end of the AGN luminosity function—a critical advancement for cosmological modeling.
2. Soft and Hard X-ray AGN Luminosity Functions: Including a redshift and luminosity-dependent dust obscuration model, the simulations reproduce the observed soft and hard X-ray AGN luminosity functions, aligning well with empirical data. This further cements the model's robustness in accounting for observational biases related to dust attenuation.
3. Radio Luminosity at $z=0$: Distinguishing between low and high excitation radio galaxies, the simulations achieve congruence with the radio luminosity function at $z=0$. This success highlights the versatility of the simulation framework in capturing diverse AGN modes and extends its relevance to radio AGN studies.
4. Downsizing Phenomenon: The simulations inherently predict the downsizing behavior, where number densities of luminous AGN peak at higher redshifts than those of their fainter counterparts. This provides a numerical substantiation for the natural occurrence of downsizing within a hierarchical assembly scenario, as major BH growth is observed to settle in place by $z=1$.
5. Eddington Ratios and Their Evolution: Eddington ratio distributions were found to peak at successively lower values with decreasing redshift, in agreement with various observational analyses. Massive BHs are shown to accrete at progressively lower Eddington ratios over time, elucidating the gradual decline of luminous AGN in the universe.

Practical and Theoretical Implications

The study's results demonstrate how cosmological simulations can reconcile hierarchical structure formation with the observed downsizing of supermassive black holes, refuting challenges posed by simplistic interpretations of AGN formation models. The alignment of simulation results with empirical AGN X-ray and radio luminosity functions strengthens the predictive power of such simulations and highlights their potential for probing the evolution of AGN and their host galaxies.

The simulation results also draw attention to the importance of parameter calibration in hydrodynamical simulations—specifically, ensuring that simulations capture the energetic interplay between feedback processes and accretion physics to prevent overestimation of the massive ends of BH and stellar mass functions.

Prospects for Future Research

Future avenues should include enhancing AGN feedback models, possibly incorporating mechanical momentum-driven winds to simulate more robust quasar feedback necessary for halting star formation in massive galaxies. Increasing the resolution of cosmological simulations will be essential to further mitigate discrepancies in low-mass AGN predictions at high redshifts. Additionally, further exploration of the relationship between AGN light curves and host galaxy properties will yield deeper insights into the mechanisms underpinning AGN triggering within merging systems.

In summary, the paper by Hirschmann et al. enriches our comprehension of BH growth dynamics and AGN evolution, cementing a critical link between these processes and the large-scale structure within a ΛCDM framework. Through meticulous simulation work and the consideration of various observational equivalences such as luminosity functions and dust obscuration, this study underscores both the triumphs and challenges facing the simulation of cosmic structure formation.