Search for: All records

Theoretical models of galaxy formation and evolution are primarily investigated through cosmological simulations and semi-analytical models. The former method consumes$\mathcal{O}\left({10}^6\right)$core-hours explicitly modeling the dynamics of the galaxies, whereas the latter method only requires$\mathcal{O}\left({10}^3\right)$core-hours foregoing directly simulating internal structure for computational efficiency. In this work, we present a proof-of-concept machine learning regression model, using a graph neural network architecture, to predict the stellar mass of high-redshift galaxies solely from their dark matter merger trees, trained from a radiation hydrodynamics cosmological simulation of the first galaxies.

The exact evolution of elements in the Universe, from primordial to heavier elements produced via the r-process, is still under scrutiny. The supernova deaths of the very first stars led to the enrichment of their local environments, and can leave behind neutron stars (NSs) as remnants. These remnants can end up in binary systems with other NSs, and eventually merge, allowing for the r-process to occur. We study the scenario where a single NS merger (NSM) enriches a halo early in its evolution to understand the impact on the second generation of stars and their metal abundances. We perform a suite of high-resolution cosmological zoom-in simulations using enzo where we have implemented a new NSM model varying the explosion energy and the delay time. In general, an NSM leads to significant r-process enhancement in the second generation of stars in a galaxy with a stellar mass of ∼105 M⊙ at redshift 10. A high explosion energy leads to a Population II (Pop II) mass fraction of 72 per cent being highly enhanced with r-process elements, while a lower explosion energy leads to 80 per cent being enhanced, but only 14 per cent being highly enhanced. When the NSM has a short delay time of 10 Myr, only 5 per cent of the mass fraction of Pop II stars is highly enhanced, while 64 per cent is highly enhanced for the longest delay time of 100 Myr. This work represents a stepping stone towards understanding how NSMs impact their environments and the metal abundances of descendant generations of stars.

Cosmic reionization is likely driven by UV starlight emanating from the first generations of galaxies. A galaxy’s UV escape fraction, or the fraction of photons escaping from the galaxy, is useful to quantify its contribution to reionization. However, the UV escape fraction is notoriously difficult to predict due to local environment dependency and variability over time. Using data from the Renaissance Simulations, we attempt to make predictions about the impact of the first stars and galaxies on their environments. We present a time-independent classification model using a general artificial neural network architecture to predict the UV escape fraction given other galaxy properties—namely halo mass, stellar mass, redshift, star formation rate, lookback time, and gas fraction. We find our validation accuracy to be approximately 50%–65%, depending on the data set size from each zoom-in region of the Renaissance Simulations.

We present a study of the co-evolution of a population of primordial star-forming minihaloes at Cosmic Dawn. In this study, we highlight the influence of individual Population III stars on the ability of nearby minihaloes to form sufficient molecular hydrogen to undergo star formation. In the absence of radiation, we find the minimum halo mass required to bring about collapse to be ∼105 M⊙, this increases to ∼106 M⊙ after two stars have formed. We find an inverse relationship between halo mass and the time required for it to recover its molecular gas after being disrupted by radiation from a nearby star. We also take advantage of the extremely high resolution to investigate the effects of major and minor mergers on the gas content of star-forming minihaloes. Contrary to previous claims of fallback of supernova ejecta, we find minihaloes evacuated after hosting Pop III stars primarily recover gas through mergers with undisturbed haloes. We identify an intriguing type of major merger between recently evacuated haloes and gas-rich ones, finding that these ‘mixed’ mergers accelerate star formation instead of suppressing it like their low-redshift counterparts. We attribute this to the gas-poor nature of one of the merging haloes resulting in no significant rise in temperature or turbulence and instead inducing a rapid increase in central density and hydrostatic pressure. This constitutes a novel formation pathway for Pop III stars and establishes major mergers as potentially the primary source of gas, thus redefining the role of major mergers at this epoch.

Early photometric results from JWST have revealed a number of galaxy candidates above redshift 10. The initial estimates of inferred stellar masses and the associated cosmic star formation rates are above most theoretical model predictions up to a factor of 20 in the most extreme cases, while this has been moderated after the recalibration of NIRCam and subsequent spectroscopic detections. Using these recent JWST observations, we use galaxy scaling relations from cosmological simulations to model the star formation history to very high redshifts, back to a starting halo mass of 107 M⊙, to infer the intrinsic properties of the JWST galaxies. Here, we explore the contribution of supermassive black holes, stellar binaries, and an excess of massive stars to the overall luminosity of high-redshift galaxies. Despite the addition of alternative components to the spectral energy distribution, we find stellar masses equal to or slightly higher than previous stellar mass estimates. Most galaxy spectra are dominated by the stellar component, and the exact choice for the stellar population model does not appear to make a major difference. We find that four of the 12 high-redshift galaxy candidates are best fit with a non-negligible active galactic nuclei component, but the evidence from the continuum alone is insufficient to confirm their existence. Upcoming spectroscopic observations of z > 10 galaxies will confirm the presence and nature of high-energy sources in the early Universe and will constrain their exact redshifts.

Forward-modeling observables from galaxy simulations enables direct comparisons between theory and observations. To generate synthetic spectral energy distributions (SEDs) that include dust absorption, re-emission, and scattering, Monte Carlo radiative transfer is often used in post-processing on a galaxy-by-galaxy basis. However, this is computationally expensive, especially if one wants to make predictions for suites of many cosmological simulations. To alleviate this computational burden, we have developed a radiative transfer emulator using an artificial neural network (ANN), ANNgelina, that can reliably predict SEDs of simulated galaxies using a small number of integrated properties of the simulated galaxies: star formation rate, stellar and dust masses, and mass-weighted metallicities of all star particles and of only star particles with age <10 Myr. Here, we present the methodology and quantify the accuracy of the predictions. We train the ANN on SEDs computed for galaxies from the IllustrisTNG project’s TNG50 cosmological magnetohydrodynamical simulation. ANNgelina is able to predict the SEDs of TNG50 galaxies in the ultraviolet (UV) to millimetre regime with a typical median absolute error of ∼7 per cent. The prediction error is the greatest in the UV, possibly due to the viewing-angle dependence being greatest in this wavelength regime. Our results demonstrate that our ANN-based emulator is a promising computationally inexpensive alternative for forward-modeling galaxy SEDs from cosmological simulations.

We present morphologies of galaxies atz≳ 9 resolved by JWST/NIRCam 2–5μm imaging. Our sample consists of 22 galaxy candidates identified by stringent dropout and photo-zcriteria in GLASS, CEERS, SMACS J0723, and Stephan’s Quintet flanking fields, one of which has been spectroscopically identified atz= 11.44. We perform surface brightness (SB) profile fitting with GALFIT for six bright galaxies with a signal-to-noise ratio = 10–40 on an individual basis and for stacked faint galaxies with secure point-spread functions (PSFs) of the NIRCam real data, carefully evaluating systematics by Monte Carlo simulations. We compare our results with those of previous JWST studies, and confirm that the effective radiir_eof our measurements are consistent with those of previous measurements atz∼ 9. We obtainr_e≃ 200–300 pc with the exponential-like profiles, Sérsic indexes ofn≃ 1–1.5, for galaxies atz∼ 12–16, indicating that the relation ofr_e∝ (1 +z)^sfor$s=-{1.22}_{-0.16}^{+0.17}$explains cosmic evolution overz∼ 0–16 for$\sim L_{z=3}^{*}$galaxies. One bright (M_UV= −21 mag) galaxy atz∼ 12, GL-z12-1, has an extremely compact profile withr_e= 39 ± 11 pc that is surely extended over the PSF. Even in the case that the GL-z12-1 SB is fit by active galactic nuclei + galaxy composite profiles, the best-fit galaxy component is again compact,$r_{\mathrm{e}}={48}_{-15}^{+38}$pc, which is significantly (>5σ) smaller than the typicalr_evalue atz∼ 12. Compared with numerical simulations, we find that such a compact galaxy naturally forms atz≳ 10, and that frequent mergers at the early epoch produce more extended galaxies following ther_e∝ (1 +z)^srelation.

The multiplicity of metal-free (Population III) stars may influence their feedback efficiency within their host dark matter haloes, affecting subsequent metal enrichment and the transition to galaxy formation. Radiative feedback from massive stars can trigger nearby star formation in dense self-shielded clouds. In model radiation self-shielding, the H2 column density must be accurately computed. In this study, we compare two local approximations based on the density gradient and Jeans length with a direct integration of column density along rays. After the primary massive star forms, we find that no secondary stars form for both the direct integration and density gradient approaches. The approximate method reduces the computation time by a factor of 2. The Jeans length approximation overestimates the H2 column density by a factor of 10, leading to five numerically enhanced self-shielded, star-forming clumps. We conclude that the density gradient approximation is sufficiently accurate for larger volume galaxy simulations, although one must still caution that the approximation cannot fully reproduce the result of direct integration.

The existence of 109 M⊙ supermassive black holes (SMBHs) within the first billion years of the Universe remains a puzzle in our conventional understanding of black hole formation and growth. Several suggested formation pathways for these SMBHs lead to a heavy seed, with an initial black hole mass of 104–106 M⊙. This can lead to an overly massive BH galaxy (OMBG), whose nuclear black hole’s mass is comparable to or even greater than the surrounding stellar mass: the black hole to stellar mass ratio is Mbh/M* ≫ 10−3, well in excess of the typical values at lower redshift. We investigate how long these newborn BHs remain outliers in the Mbh − M* relation, by exploring the subsequent evolution of two OMBGs previously identified in the Renaissance simulations. We find that both OMBGs have Mbh/M* > 1 during their entire life, from their birth at z ≈ 15 until they merge with much more massive haloes at z ≈ 8. We find that the OMBGs are spatially resolvable from their more massive, 1011 M⊙, neighbouring haloes until their mergers are complete at z ≈ 8. This affords a window for future observations with JWST and sensitive X-ray telescopes to diagnose the heavy-seed scenario, by detecting similar OMBGs and establishing their uniquely high black hole-to-stellar mass ratio.

We explore the effect of dust on the growth of seed black holes (BHs) in the early universe. Previous 1D radiation-hydrodynamic (RHD) simulations show that increased radiation pressure on dust further suppresses the accretion rate than the case for the chemically pristine gas. Using the Enzo+Moray code, we perform a suite of 3D RHD simulations of accreting BHs in a dusty interstellar medium (ISM). We use the modified Grackle cooling library to consider dust physics in its nonequilibrium chemistry. The BH goes through an early evolutionary phase, where ionizing BH radiation creates an oscillating Hiiregion as it cycles between accretion and feedback. As the simulations proceed, dense cold gas accumulates outside the ionized region where inflow from the neutral medium meets the outflow driven by radiation pressure. In the late phase, high-density gas streams develop and break the quasi-spherical symmetry of the ionized region, rapidly boosting the accretion rate. The late phase is characterized by the coexistence of strong ionized outflows and fueling high-density gas inflows. The mean accretion rate increases with metallicity reaching a peak atZ∼ 0.01–0.1Z_☉, one order of magnitude higher than the one for pristine gas. However, as the metallicity approaches the solar abundance, the mean accretion rate drops as the radiation pressure becomes strong enough to drive out the high-density gas. Our results indicate that a dusty metal-poor ISM can accelerate the growth rate of BHs in the early universe, but can also stun its growth as the ISM is further enriched toward the solar abundance.