Search

Cosmological galaxy formation simulations are powerful tools to understand the complex processes that govern the formation and evolution of galaxies. However, evaluating the realism of these simulations remains a challenge. The two common approaches for evaluating galaxy simulations is either through scaling relations based on a few key physical galaxy properties, or through a set of pre-defined morphological parameters based on galaxy images. This paper proposes a novel image-based method for evaluating the quality of galaxy simulations using unsupervised deep learning anomaly detection techniques. By comparing full galaxy images, our approach can identify and quantify discrepancies between simulated and observed galaxies. As a demonstration, we apply this method to SDSS imaging and NIHAO simulations with different physics models, parameters, and resolution. We further compare the metric of our method to scaling relations as well as morphological parameters. We show that anomaly detection is able to capture similarities and differences between real and simulated objects that scaling relations and morphological parameters are unable to cover, thus indeed providing a new point of view to validate and calibrate cosmological simulations against observed data., Comment: 13 pages, 16 figures, published in MNRAS

We append two additional black hole (BH) accretion models, namely viscous disc and gravitational torque-driven accretion, into the Numerical Investigation of a Hundred Astrophysical Objects (NIHAO) project of galaxy simulations. We show that these accretion models, characterized by a weaker dependence on the BH mass compared to the commonly used Bondi-Hoyle accretion, naturally create a common evolutionary track (co-existence) between the mass of the BH and the stellar mass of the galaxy, even without any direct coupling via feedback (FB). While FB is indeed required to control the final BH and stellar mass of the galaxies, our results suggest that FB might not be the leading driver of the cosmic co-evolution between these two quantities; in these models, co-evolution is simply determined by the shared central gas supply. Conversely, simulations using Bondi-Hoyle accretion show a two-step evolution, with an early growth of stellar mass followed by exponential growth of the central supermassive black hole (SMBH). Our results show that the modelling of BH accretion (sometimes overlooked) is an extremely important part of BH evolution and can improve our understanding of how scaling relations emerge and evolve, and whether SMBH and stellar mass co-exist or co-evolve through cosmic time., Comment: 13 pages, 10 figures

Although active galactic nuclei (AGN) feedback is required in simulations of galaxies to regulate star formation, further downstream effects on the dark matter distribution of the halo and stellar kinematics of the central galaxy can be expected. We combine simulations of galaxies with and without AGN physics from the Numerical Investigation of a Hundred Astrophysical Objects (NIHAO) to investigate the effect of AGN on the dark matter profile and central stellar rotation of the host galaxies. Specifically, we study how the concentration-halo mass ($c-M$) relation and the stellar spin parameter ($\lambda_R$) are affected by AGN feedback. We find that AGN physics is crucial to reduce the central density of simulated massive ($\gtrsim 10^{12}$ M$_\odot$) galaxies and bring their concentration to agreement with results from the Spitzer Photometry & Accurate Rotation Curves (SPARC) sample. Similarly, AGN feedback has a key role in reproducing the dichotomy between slow and fast rotators as observed by the ATLAS$^{3\text{D}}$ survey. Without star formation suppression due to AGN feedback, the number of fast rotators strongly exceeds the observational constraints. Our study shows that there are several collateral effects that support the importance of AGN feedback in galaxy formation, and these effects can be used to constrain its implementation in numerical simulations., Comment: Accepted for publication in MNRAS. 15 pages, 10 figures

The transition of high-mass galaxies from being blue and star forming to being red and dead is a crucial step in galaxy evolution, yet not fully understood. In this work, we use the NIHAO suite of galaxy simulations to investigate the relation between the transition time through the green valley and other galaxy properties. The typical green valley crossing time of our galaxies is approximately 400 Myr, somewhat shorter than observational estimates. The crossing of the green valley is triggered by the onset of AGN feedback and the subsequent shut down of star formation. Interestingly the time spent in the green valley is not related to any other galaxy properties, such as stellar age or metallicity, or the time at which the star formation quenching takes place. The crossing time is set by two main contributions: the ageing of the current stellar population and the residual star formation in the green valley. These effects are of comparable magnitude, while major and minor mergers have a negligible contribution. Most interestingly, we find the time that a galaxy spends to travel through the green valley is twice the $e$-folding time of the star formation quenching. This result is stable against galaxy properties and the exact numerical implementation of AGN feedback in the simulation. Assuming a typical crossing time of about one Gyr inferred from observations, our results imply that any mechanism or process aiming to quench star formation, must do it on a typical timescale of 500 Myr., Comment: Accepted by MNRAS

An extensive catalog of spatially-resolved galaxy rotation curves and multi-band optical light profiles for 1752 observed spiral galaxies is assembled to explore the drivers of diversity in galaxy structural parameters, rotation curve shapes, and stellar mass profiles. Similar data were extracted from the NIHAO galaxy simulations to identify any differences between observations and simulations. Several parameters, including the inner slope "S" of a rotation curve (RC), were tested for diversity. Two distinct populations are found in observed and simulated galaxies; (i) blue, low mass spirals with stellar mass M* < 10^9.3 Msol and roughly constant "S", and (ii) redder, more massive and more diverse spirals with rapidly increasing "S". In all cases, the value of "S" seems equally contributed by the baryonic and non-baryonic (dark) matter. Diversity is shown to increase mildly with mass. Numerical simulations reproduce well most baryon-dominated galaxy parameter distributions, such as the inner stellar mass profile slope and baryonic scaling relations, but they struggle to match the full diversity of observed galaxy rotation curves (through "S") and most dark-matter-dominated parameters. To reproduce observations, the error broadening of the simulation's intrinsic spread of RC metrics would have to be tripled. The differences in various projections of observed and simulated scaling relations may reflect limitations of current sub-grid physics models to fully capture the complex nature of galaxies. For instance, AGNs are shown to have a significant effect on the shapes of simulated RCs. The inclusion of AGN feedback brings simulated and observed inner RC shapes into closer agreement., Comment: 22 pages, 14 figures, submitted to MNRAS, comments welcome