arXiv subjects

Mauro Bernardini

Publications and source records attributed to Mauro Bernardini.

11 recordsLinked to original sources

arXiv2026

The Nature of Post-Starburst Galaxies: Real Deal or Masquerading Impostors?

Post-starburst galaxies (PSBs) are a population of galaxies with spectral and photometric features indicative of rapid quenching following a recent starburst. The origin and nature of PSBs are currently debated. For example, a number of observed PSBs unexpectedly host substantial molecular gas despite their low inferred star-formation activity. Furthermore, the relative roles of galaxy interactions and quenching mechanisms in PSBs remain unclear. We study PSBs at $z=0.7$ and $z=1$ in the FIREbox cosmological simulation, selecting them primarily via their rest-frame optical photometric properties. The fraction of PSBs in FIREbox broadly agrees with observations, although some candidates are clear impostors with star-formation rates comparable to star-forming galaxies of similar mass. Impostors are rich in molecular gas and have a larger near-to-mid infrared flux ratios compared to quenched PSBs in the sample. The role of galaxy interactions of PSBs in FIREbox depends on their stellar mass. At low stellar masses ($\lesssim 10^{10}~\mathrm{M}_\odot$), PSBs have interaction fractions comparable to those of non-PSBs in the simulation, consistent with a scenario in which stellar feedback and gas consumption drive temporary quenching of star formation. At higher stellar masses ($\gtrsim 10^{10}~\mathrm{M}_\odot$), PSBs are preferentially interacting systems compared to non-PSBs, with major mergers providing the dominant contribution. We conclude that stellar feedback and galaxy interactions in FIREbox can produce galaxies with observational properties akin to those of observed PSBs, many of which are actively forming stars. Additional quenching channels, such as massive black hole feedback, are likely required to explain a long-lived, quenched population of PSBs.

astro-ph.GA↗

arXiv2026

Unveiling the dark Universe with HI and EMBER-2

Next-generation radio telescopes will provide unprecedented data volumes of the neutral hydrogen (HI) distribution across cosmic time. The spatial and kinematic distribution of HI is a biased tracer of the underlying matter field, and as such contains information on the distribution of dark matter over a wide range of scales. Extracting dark matter properties from HI, however, is non-trivial because baryonic processes linked to galaxy formation significantly modify the HI distribution. Additionally, methods that use empirical relations, often calibrated via numerical simulations, do not use the full field-level information to model the complex relation between HI and dark matter. We use the recently introduced EMBER-2 model to directly predict dark matter distributions from HI tracers over a wide redshift range, z=0-6. After training on cosmological galaxy formation simulations run with FIRE-2, our method accurately recovers key statistics, including dark matter mass fractions, surface density profiles and cross-correlations, where the latter are reconstructed at an accuracy of 20% down to scales of k = 100 h/cMpc constituting a significant improvement over traditional approaches. The presented method may become a key ingredient in future inference pipelines as it can be readily integrated into downstream analysis tasks of radio surveys.

astro-ph.CO↗

arXiv2025

EMBER-2: Emulating baryons from dark matter across cosmic time with deep modulation networks

Galaxy formation is a complex problem that connects large scale cosmology with small scale astrophysics over cosmic timescales. Hydrodynamical simulations are the most principled approach to model galaxy formation, but have large computational costs. Recently, emulation techniques based on Convolutional Neural Networks (CNNs) have been proposed to predict baryonic properties directly from dark matter simulations. The advantage of these emulators is their ability to capture relevant correlations, but at a fraction of the computational cost compared to simulations. However, training basic CNNs over large redshift ranges is challenging, due to the increasing non-linear interplay between dark matter and baryons paired with the memory inefficiency of CNNs. This work introduces EMBER-2, an improved version of the EMBER (EMulating Baryonic EnRichment) framework, to simultaneously emulate multiple baryon channels including gas density, velocity, temperature and HI density over a large redshift range, from z=6 to z=0. EMBER-2 incorporates a context-based styling network paired with Modulated Convolutions for fast, accurate and memory efficient emulation capable of interpolating the entire redshift range with a single CNN. Although EMBER-2 uses fewer than 1/6 the number of trainable parameters than the previous version, the model improves in every tested summary metric including gas mass conservation and cross-correlation coefficients. The EMBER-2 framework builds the foundation to produce mock catalogues of field level data and derived summary statistics that can directly be incorporated in future analysis pipelines. We release the source code at the official website https://maurbe.github.io/ember2/.

astro-ph.GA↗

arXiv2024

The HI covering fraction of Lyman Limit Systems in FIRE haloes

Atomic hydrogen (HI) serves a crucial role in connecting galactic-scale properties such as star formation with the large-scale structure of the Universe. While recent numerical simulations have successfully matched the observed covering fraction of HI near Lyman Break Galaxies (LBGs) and in the foreground of luminous quasars at redshifts $z \lesssim 3$, the low-mass end remains as-of-yet unexplored in observational and computational surveys. We employ a cosmological, hydrodynamical simulation (FIREbox) supplemented with zoom-in simulations (MassiveFIRE) from the Feedback In Realistic Environments (FIRE) project to investigate the HI covering fraction of Lyman Limit Systems ($N_{\mathrm{HI}} \gtrsim 10^{17.2}$ cm$^{-2}$) across a wide range of redshifts ($z=0-6$) and halo masses ($10^8-10^{13} M_{\odot}$ at $z=0$, $10^8-10^{11} M_{\odot}$ at $z=6$) in the absence of feedback from active galactic nuclei. We find that the covering fraction inside haloes exhibits a strong increase with redshift, with only a weak dependence on halo mass for higher-mass haloes. For massive haloes ($M_{\mathrm{vir}} \sim 10^{11}-10^{12} M_{\odot}$), the radial profiles showcase scale-invariance and remain independent of mass. The radial dependence is well-captured by a fitting function. The covering fractions in our simulations are in good agreement with measurements of the covering fraction in LBGs. Our comprehensive analysis unveils a complex dependence with redshift and halo mass for haloes with $M_{\mathrm{vir}} \lesssim 10^{10} M_{\odot}$ that future observations aim to constrain, providing key insights into the physics of structure formation and gas assembly.

astro-ph.GA↗

arXiv2024

Inflow and outflow properties, not total gas fractions, drive the evolution of the mass-metallicity relation

Observations show a tight correlation between the stellar mass of galaxies and their gas-phase metallicity (MZR). This relation evolves with redshift, with higher-redshift galaxies being characterized by lower metallicities. Understanding the physical origin of the slope and redshift evolution of the MZR may provide important insight into the physical processes underpinning it: star formation, feedback, and cosmological inflows. While theoretical models ascribe the shape of the MZR to the lower efficiency of galactic outflows in more massive galaxies, what drives its evolution remains an open question. In this letter, we analyze how the MZR evolves over $z=0-3$, combining results from the FIREbox cosmological volume simulation with analytical models. Contrary to a frequent assertion in the literature, we find that the evolution of the gas fraction does not contribute significantly to the redshift evolution of the MZR. Instead, we show that the latter is driven by the redshift-dependence of the inflow metallicity, outflow metallicity, and mass loading factor, whose relative importance depends on stellar mass. These findings also suggest that the evolution of the MZR is not explained by galaxies moving along a fixed surface in the space spanned by stellar mass, gas phase metallicity, and star formation rate.

astro-ph.GA↗

arXiv2023

Starburst-induced gas-stars kinematic misalignment

A kinematic misalignment of the stellar and gas components is a phenomenon observed in a significant fraction of galaxies. However, the underlying physical mechanisms are not well understood. A commonly proposed scenario for the formation of a misaligned component requires any pre-existing gas disc to be removed, via fly-bys or ejective feedback from an active galactic nucleus. In this Letter, we study the evolution of a Milky Way mass galaxy in the FIREbox cosmological volume that displays a thin, counter-rotating gas disc with respect to its stellar component at low redshift. In contrast to scenarios involving gas ejection, we find that pre-existing gas is mainly removed via the conversion into stars in a central starburst, triggered by a merging satellite galaxy. The newly-accreted, counter-rotating gas eventually settles into a kinematically misaligned disc. About 4.4 (8 out of 182) of FIREbox galaxies with stellar masses larger than 5e9 Msun at z=0 exhibit gas-star kinematic misalignment. In all cases, we identify central starburst-driven depletion as the main reason for the removal of the pre-existing co-rotating gas component, with no need for feedback from, e.g., a central active black hole. However, during the starburst, the gas is funneled towards the central regions, likely enhancing black hole activity. By comparing the fraction of misaligned discs between FIREbox and other simulations and observations, we conclude that this channel might have a non-negligible role in inducing kinematic misalignment in galaxies.

astro-ph.GA↗

arXiv2023

Starbursts driven by central gas compaction

Starburst (SB) galaxies are a rare population of galaxies with star formation rates (SFRs) greatly exceeding those of the majority of star-forming galaxies with similar stellar mass. It is unclear whether these bursts are the result of either especially large gas reservoirs or enhanced efficiencies in converting gas into stars. Tidal torques resulting from gas-rich galaxy mergers are known to enhance the SFR by funneling gas towards the centre. However, recent theoretical works show that mergers do not always trigger a SB and not all SB galaxies are interacting systems, raising the question of what drives a SB. We analyse a large sample of SB galaxies and a mass- and redshift-matched sample of control galaxies, drawn from the FIREbox cosmological volume at z=0-1. We find that SB galaxies have both larger molecular gas fractions and shorter molecular depletion times than control galaxies, but similar total gas masses. Control galaxies evolve towards the SB regime by gas compaction in their central regions, over timescales of about 70 Myr, accompanied by an increase in the fraction of ultra-dense and molecular gas. The driving mechanism behind the SB varies depending on the mass of the galaxy. Massive (Mstar > 1e10 Msun) galaxies undergoing intense, long-lasting SBs are mostly driven by galaxy interactions. Conversely, SBs in non-interacting galaxies are often triggered by a global gravitational instability, that can result in a breathing mode in low-mass galaxies.

astro-ph.GA↗

arXiv2023

FIREbox: Simulating galaxies at high dynamic range in a cosmological volume

We introduce a suite of cosmological volume simulations to study the evolution of galaxies as part of the Feedback in Realistic Environments project. FIREbox, the principal simulation of the present suite, provides a representative sample of galaxies (~1000 galaxies with Mstar > 10^8 Msun at z=0) at a resolution (~20 pc, m_b ~ 6x10^4 Msun) comparable to state-of-the-art galaxy zoom-in simulations. FIREbox captures the multiphase nature of the interstellar medium in a fully cosmological setting (L=22.1 Mpc) thanks to its exceptionally high dynamic range (~10^6) and the inclusion of multi-channel stellar feedback. Here, we focus on validating the simulation predictions by comparing to observational data. We find that simulated galaxies with Mstar < 10^{10.5-11} Msun have star formation rates, gas masses, and metallicities in broad agreement with observations. These galaxy scaling relations extend to low masses (Mstar ~ 10^7 Msun) and follow a (broken) power-law relationship. Also reproduced are the evolution of the cosmic HI density and the HI column density distribution at z~0-5. At low z, FIREbox predicts a peak in the stellar-mass--halo-mass relation, but also a higher abundance of massive galaxies and a higher cosmic star formation rate density than observed, showing that stellar feedback alone is insufficient to reproduce the properties of massive galaxies at late times. Given its high resolution and sample size, FIREbox offers a baseline prediction of galaxy formation theory in a $Λ$CDM Universe while also highlighting modeling challenges to be addressed in next-generation galaxy simulations.

astro-ph.GA↗

arXiv2022

From EMBER to FIRE: predicting high resolution baryon fields from dark matter simulations with Deep Learning

Hydrodynamic simulations provide a powerful, but computationally expensive, approach to study the interplay of dark matter and baryons in cosmological structure formation. Here we introduce the EMulating Baryonic EnRichment (EMBER) Deep Learning framework to predict baryon fields based on dark-matter-only simulations thereby reducing computational cost. EMBER comprises two network architectures, U-Net and Wasserstein Generative Adversarial Networks (WGANs), to predict two-dimensional gas and HI densities from dark matter fields. We design the conditional WGANs as stochastic emulators, such that multiple target fields can be sampled from the same dark matter input. For training we combine cosmological volume and zoom-in hydrodynamical simulations from the Feedback in Realistic Environments (FIRE) project to represent a large range of scales. Our fiducial WGAN model reproduces the gas and HI power spectra within 10% accuracy down to ~10 kpc scales. Furthermore, we investigate the capability of EMBER to predict high resolution baryon fields from low resolution dark matter inputs through upsampling techniques. As a practical application, we use this methodology to emulate high-resolution HI maps for a dark matter simulation of a L=100 Mpc/h comoving cosmological box. The gas content of dark matter haloes and the HI column density distributions predicted by EMBER agree well with results of large volume cosmological simulations and abundance matching models. Our method provides a computationally efficient, stochastic emulator for augmenting dark matter only simulations with physically consistent maps of baryon fields.

astro-ph.GA↗

arXiv2020

Towards a Polarisation Prediction for LISA via Intensity Interferometry

Compact Galactic binary systems with orbital periods of a few hours are expected to be detected in gravitational waves (GW) by LISA or a similar mission. At present, these so-called verification binaries provide predictions for GW frequency and amplitude. A full polarisation prediction would provide a new method to calibrate LISA and other GW observatories, but requires resolving the orientation of the binary on the sky, which is not currently possible. We suggest a method to determine the elusive binary orientation and hence predict the GW polarisation, using km-scale optical intensity interferometry. The most promising candidate is CD-30$^{\circ}$ 11223, consisting of a hot helium subdwarf with $m_B = 12$ and a much fainter white dwarf companion, in a nearly edge-on orbit with period 70.5 min. We estimate that the brighter star is tidally stretched by 6%. Resolving the tidal stretching would provide the binary orientation. The resolution needed is far beyond any current instrument, but not beyond current technology. We consider scenarios where an array of telescopes with km-scale baselines and/or the Very Large Telescope (VLT) and Extremely Large Telescope (ELT) are equipped with recently-developed kilo-pixel sub-ns single-photon counters and used for intensity interferometry. We estimate that a team-up of the VLT and ELT could measure the orientation to $\pm 1^{\circ}$ at 2$σ$ confidence in 24 hours of observation.

astro-ph.IM↗

arXiv2019

Predicting dark matter halo formation in N-body simulations with deep regression networks

Dark matter haloes play a fundamental role in cosmological structure formation. The most common approach to model their assembly mechanisms is through N-body simulations. In this work we present an innovative pathway to predict dark matter halo formation from the initial density field using a Deep Learning algorithm. We implement and train a Deep Convolutional Neural Network (DCNN) to solve the task of retrieving Lagrangian patches from which dark matter halos will condense. The volumetric multi-label classification task is turned into a regression problem by means of the euclidean distance transformation. The network is complemented by an adaptive version of the watershed algorithm to form the entire protohalo identification pipeline. We show that splitting the segmentation problem into two distinct sub-tasks allows for training smaller and faster networks, while the predictive power of the pipeline remains the same. The model is trained on synthetic data derived from a single full N-body simulation and achieves deviations of ~10% when reconstructing the dark matter halo mass function at z=0. This approach represents a promising framework for learning highly non-linear relations in the primordial density field. As a practical application, our method can be used to produce mock dark matter halo catalogues directly from the initial conditions of N-body simulations.

astro-ph.CO↗