arXiv subjects

Rupert A. C. Croft

Publications and source records attributed to Rupert A. C. Croft.

At least 19 recordsLinked to original sources

arXiv2026

The Galaxy Luminosity Functions in ASTRID: Predictions for LSST

We present validated and forward-modelled galaxy luminosity functions and photometric predictions for the Vera C. Rubin Observatory Legacy Survey of Space and Time using the ASTRID cosmological hydrodynamical simulation. Galaxy magnitudes are computed by combining stellar population synthesis modeling with a physically motivated dust attenuation prescription in which the optical depth scales with metal surface density. The dust model is calibrated at z = 0 using SDSS luminosity functions and tested at intermediate redshifts (z = 0.5, 1.0, and 1.5) in rest-frame B, V , R, and I bands. We find that the attenuated luminosity functions reproduce observed galaxy statistics across multiple wavelengths and redshifts. Using this calibrated framework, we construct LSST-ready mock photometric catalogs over 0 <= z <= 2 in steps of Delta z = 0.1, containing ~378 million galaxies. We provide predicted apparent-magnitude luminosity functions in the LSST ugrizy bands, derive best-fit Schechter parameters as a compact analytic representation, and compute differential and cumulative galaxy number counts as a function of survey depth from Year 1 to Year 10.

astro-ph.GA↗

arXiv2026

The Properties of Little Red Dot Galaxies in the ASTRID Simulation

We present simulated counterparts of the ``Little Red Dot'' (LRD) galaxies observed with JWST, using the large cosmological hydrodynamic simulation, ASTRID. We create mock observations of the galaxies ($5 \leq z \leq 8$) in ASTRID, and find seventeen which fit the color and size criteria of LRDs. These LRDs are galaxies with high stellar masses ($\rm log(M_*/M_{\odot}) \geq 9.7$), and massive black holes ($\rm log(M_{BH}/M_{\odot}) \geq 6.8$). The host galaxies are dense, with stellar half mass radii ($\rm 325\,pc \leq r_{{\rm half},*} \leq 620\,pc$), and dust attenuation in the F444W band above 1.25. Their star formation has been recently quenched. They host relatively bright AGN that are dust-obscured and contribute significantly to the rest-frame optical red slope and have relatively low luminosity in the rest-frame ultraviolet, where the host galaxy's stars are more dominant. These LRDs are in an evolutionary phase of miniquenching that is the result of AGN feedback from their massive black holes. The LRDs in ASTRID are bright with F444W magnitudes of $23.5-25.5$. The less massive and fainter galaxies in ASTRID lack the dust concentration necessary to produce the red slope of an LRD, though this could be an effect of limited resolution. Most of the highest Eddington black holes are not LRDs due to insufficient dust attenuation from their host galaxies, which are also experiencing relatively high star formation rates. This results in their spectra being too flat, despite their highly accreting black holes.

astro-ph.CO↗

arXiv2025

From ASTRID to BRAHMA -- The role of overmassive black holes in little red dots in cosmological simulations

We leverage the overmassive black holes ($\rm M_{BH}/M_{\ast} \approx0.1$) present in a realization of the BRAHMA cosmological hydrodynamic simulation suite to investigate their role in the emission of the unique ``little red dot'' (LRD) objects identified by the James Webb Space Telescope (JWST). We find that these black holes can produce LRD-like observables when their emission is modeled with a dense gas cloud shrouding the active galactic nucleus (AGN). Between redshifts 5 and 8, we find the number density of LRDs in this simulation to be $\rm 2.04 \pm 0.32 \times 10^{-4} \space Mpc^{-3}$, which is broadly consistent with current estimates for the total LRD population from JWST. Their emission in the rest-frame visible spectrum is dominated by their AGN, which induces the red color indicative of LRDs via a very strong Balmer break. Additionally, the elevated mass of the black holes reduces the temperature of their accretion discs. This shifts the peak of the AGN emission towards longer wavelengths, and increases their brightness in the rest-frame visible spectrum relative to lower mass black holes accreting at the same rate. These simulated LRDs have very minimal dust attenuation ($\rm A_V = 0.21 \pm 0.12$), limiting the amount of dust re-emission that would occur in the infrared, making them very likely to fall below the observed detection limits from observatories like the Atacama Large Millimeter Array (ALMA). In contrast to the BRAHMA box, the ASTRID simulation produces systematically smaller black holes and predicts LRD number densities that are more than two orders of magnitude lower than current measurements. We therefore conclude that the presence of black holes that are overmassive relative to their host galaxy, and enshrouded in dense gas, is necessary for AGN-dominated LRD models to reproduce both the observed properties and abundances of JWST LRD populations.

astro-ph.GA↗

arXiv2025

Large-scale surveys of the quasar proximity effect

The UV radiation from high redshift quasars causes a local deficit in the neutral hydrogen absorption (Lyman-alpha forest) in their spectra, known as the proximity effect. Measurements from small samples of tens to hundreds of quasars have been used to constrain the global intensity of the UV background radiation, but so far the power of large-scale surveys such as the Sloan Digital Sky Survey and the Dark Energy Spectroscopic Instrument (DESI) survey has not been used to investigate the UV background in more detail. We develop a CDM-based halo model of the quasar proximity effect, which accounts by construction for the fact that quasars reside in overdense regions. We test this model on quasar Lyman-alpha spectra from the ASTRID cosmological hydrodynamic simulation, which includes self-consistent formation of quasar black holes and the intergalactic medium surrounding them. Fitting the model to individual quasar spectra, we constrain two parameters, r_eq (the radius at which the local quasar radiation intensity equals the background), and the quasar bias b_q (related to host halo mass). We find that r_eq can be recovered in an unbiased fashion with a statistical uncertainty of 25-50% from a single quasar spectrum. Applying such fitting to samples of millions of spectra from e.g., DESI would allow measurement of the UVBG intensity and its evolution with redshift with high precision. We use another, larger-scale, lower resolution simulation (Uchuu) to test how such a large sample of proximity effect measurements could be used to probe the spatial fluctuations in the intergalactic radiation field. We find that the large-scale structure of the UV radiation intensity could be mapped and its power spectrum measured on 100-1000 Mpc/h scales. This could allow the large-scale radiation field to join the density field as a dataset for constraining cosmology and the sources of radiation.

astro-ph.CO↗

arXiv2025

An AI super-resolution field emulator for cosmological hydrodynamics: the Lyman-α forest

We extend our super-resolution and emulation framework for cosmological dark matter simulations to include hydrodynamics. We present a two-stage deep learning model to emulate high-resolution (HR-HydroSim) baryonic fields from low-resolution (LR-HydroSim) simulations at redshift $z = 3$. The method takes as inputs an LR-HydroSim and the high-resolution initial conditions (HR-HydroICs). First, the model stochastically generates high-resolution baryonic fields from the LR-HydroSim. Second, a deterministic emulator refines these fields using HR-HydroICs to reconstruct small-scale structures including displacement, velocity, internal energy, and gas/star classification. Trained on paired low- and high-resolution simulations produced with \texttt{MP-Gadget}, the model captures small-scale structures of the intergalactic medium and %Lyman-$α$ forest observables down to the 100 kpc pressure smoothing scale relevant to the Lyman-$α$ forest. The model achieves subpercent error for overdensity, temperature, velocity, and optical depth fields, a mean relative error of 1.07\% in the large-scale flux power spectrum ($k < 3 \times 10^{-2}\ \mathrm{s/km}$), and less than 10\% error in the flux probability distribution function. Notably, the two-stage model reduces the compute time by a factor of $\sim$450 compared to full smoothed particle hydrodynamics at the same resolution. This work demonstrates the potential of this framework as a powerful and efficient tool for generating high-resolution fields offering fast and accurate alternatives to traditional cosmological hydrodynamic simulations and enabling large-volume mock datasets for next-generation cosmological surveys.

astro-ph.CO↗

arXiv2025

Bridging Literature and the Universe Via A Multi-Agent Large Language Model System

As cosmological simulations and their associated software become increasingly complex, physicists face the challenge of searching through vast amounts of literature and user manuals to extract simulation parameters from dense academic papers, each using different models and formats. Translating these parameters into executable scripts remains a time-consuming and error-prone process. To improve efficiency in physics research and accelerate the cosmological simulation process, we introduce SimAgents, a multi-agent system designed to automate both parameter configuration from the literature and preliminary analysis for cosmology research. SimAgents is powered by specialized LLM agents capable of physics reasoning, simulation software validation, and tool execution. These agents collaborate through structured communication, ensuring that extracted parameters are physically meaningful, internally consistent, and software-compliant. We also construct a cosmological parameter extraction evaluation dataset by collecting over 40 simulations in published papers from Arxiv and leading journals that cover diverse simulation types. Experiments on the dataset demonstrate a strong performance of SimAgents, highlighting its effectiveness and potential to accelerate scientific research for physicists. Our demonstration video is available at: https://youtu.be/w1zLpm_CaWA. The complete system and dataset are publicly available at https://github.com/xwzhang98/SimAgents.

astro-ph.IM↗

arXiv2025

AI-assisted super-resolution cosmological simulations IV: An emulator for deterministic realizations

Super-resolution (SR) models in cosmological simulations use deep learning (DL) to rapidly enhance low-resolution (LR) runs with statistically correct fine details. These models preserves large-scale structures by conditioning on an LR version of the simulation. On smaller scales, the generative process is inherently stochastic, producing multiple possible SR realizations with distinct small-scale structures. Validation of reconstructed SR runs from LR simulations requires ensuring that specific statistics of interest are accurately reproduced by comparing SR outputs with target high resolution (HR) runs. In this study, we develop an emulator designed to reproduce the small-scale structures of target HR simulation with high fidelity. By processing an SR realization alongside the high-resolution initial condition (HRIC), we transform the SR output to emulate the result of a full simulation with that HRIC. By comparing various metrics, from visualization to individual halo measurements, we demonstrate that the emulated SR runs closely align with the target HR simulation, even at length scales an order of magnitude smaller than the corresponding LR run. These results show the potential of this method for efficiently generating accurate simulations and mock observations for large galaxy surveys.

astro-ph.CO↗

arXiv2025

Weak Lensing the non-Linear Ly-alpha Forest

We evaluate the performance of the Lyman-$α$ forest weak gravitational lensing estimator of Metcalf et al. on forest data from hydrodynamic simulations and ray-traced simulated lensing potentials. We compare the results to those obtained from the Gaussian random field simulated Ly$α$ forest data and lensing potentials used in previous work. We find that the estimator is able to reconstruct the lensing potentials from the more realistic data, and investigate dependence on spectrum signal to noise. The non-linearity and non-Gaussianity in this forest data arising from gravitational instability and hydrodynamics causes a reduction in signal to noise by a factor of $\sim2.7$ for noise free data and a factor of $\sim 1.5$ for spectra with signal to noise of order unity (comparable to current observational data). Compared to Gaussian field lensing potentials, using ray-traced potentials from N-body simulations incurs a further signal to noise reduction of a factor of $\sim1.3$ at all noise levels. The non-linearity in the forest data is also observed to increase bias in the reconstructed potentials by $5-25\%$, and the ray-traced lensing potential further increases the bias by $20-30\%$. We demonstrate methods for mitigating these issues including Gaussianization and bias correction which could be used in real observations.

astro-ph.CO↗

arXiv2025

Forecasting the Detection of Lyman-Alpha Forest Weak Lensing from the Dark Energy Spectroscopic Instrument and Other Future Surveys

The apparent angular positions of quasars are deflected on the sky by the gravitational field sourced by foreground matter. This weak lensing effect is measurable through the distortions it introduces in the lensed quasar spectra. Discrepancies in the statistics of the Lyman-$α$ forest spectral absorption features can be used to reconstruct the foreground lensing potential. We extend the study of this method of Lyman-$α$ forest weak gravitational lensing to lower angular forest spectrum source densities than previous work. We evaluate the performance of the Lyman-$α$ lensing estimator of Metcalf et al. (2020) on mock data based on the angular forest source density ($50$ per square degree) and volume ($\sim$700,000 spectra total) of the DESI survey. We simulate the foreground galaxy distribution and lensing potentials with redshift evolution approximated by N-body simulation and simulate Gaussian-random Lyman-$α$ forests to produce mock data for the entire DESI footprint. By correlating the foreground galaxy distribution with the potential reconstructed by the estimator, we find that a weak lensing detection with signal to noise of $\sim4$ will be possible with the full DESI data. We show that spectral surveys with low density and high volume are promising candidates for forest weak lensing in addition to the high resolution data that have been considered in previous work. We present forecasts for future spectral surveys and show that with larger datasets a detection with signal to noise $>10$ will be possible.

astro-ph.CO↗

arXiv2023

Modeling Quasar Proximity Zones in a Realistic Cosmological Environment with a Self-consistent Light Curve

We study quasar proximity zones in a simulation that includes a self-consistent quasar formation model and realistic IGM environments. The quasar host halo is $10^{13}\ M_{\mathrm{\odot}}$ at $z=6$, more massive than typical halos studied in previous work. Between $6<z<7.5$, the quasar luminosity varies rapidly, with a mean magnitude of $M_{UV,mean}=-24.8$ and the fluctuation reaching up to two orders of magnitude. Using this light curve to post-process the dense environment around the quasar, we find that the proximity zone size ($R_{p}$) ranges between $0.5-5$ pMpc. We show that the light curve variability causes a similar degree of scatter in $R_{p}$ as does the density fluctuation, both of which result in a standard deviation of $\sim 0.3$ pMpc). The $R_{p}$ traces the light curve fluctuations closely but with a time delay of $\sim 10^4\ \mathrm{yr}$, breaking the correspondence between the $R_{p}$ and the contemporaneous $M_{UV}$. This also indicates that we can only infer quasar activity within the past $\sim 10^4$ years instead of the integrated lifetime from $R_{p}$ in the later part of cosmic reionization. Compared with the variable light curve, a constant light curve underestimates the $R_{p}$ by 13% at the dim end ($M_{UV}\sim -23.5$), and overestimates the $R_{p}$ by 30% at the bright end ($M_{UV}\sim -26$). By calculating the $R_{p}$ generated by a number of quasars, we show that variable light curves predict a wider $R_{p}$ distribution than lightbulb models, and readily explain the extremely small $R_{p}$ values that have been observed.

astro-ph.CO↗

arXiv2023

AI-assisted super-resolution cosmological simulations III: Time evolution

In this work, we extend our recently developed super-resolution (SR) model for cosmological simulations to produce fully time consistent evolving representations of the particle phase-space distribution. We employ a style-based constrained generative adversarial network (Style-GAN) where the changing cosmic time is an input style parameter to the network. The matter power spectrum and halo mass function agree well with results from high-resolution N-body simulations over the full trained redshift range ($10 \le z \le 0$). Furthermore, we assess the temporal consistency of our SR model by constructing halo merger trees. We examine progenitors, descendants and mass growth along the tree branches. All statistical indicators demonstrate the ability of our SR model to generate satisfactory high-resolution simulations based on low-resolution inputs.

astro-ph.CO↗

arXiv2022

The BlueTides Mock Image Catalogue: Simulated observations of high-redshift galaxies and predictions for JWST imaging surveys

We present a mock image catalogue of ~100,000 MUV=-22.5 to -19.6 mag galaxies at z=7-12 from the BlueTides cosmological simulation. We create mock images of each galaxy with the James Webb (JWST), Hubble, Roman, and Euclid Space Telescopes, as well as Subaru, and VISTA, with a range of near- and mid-infrared filters. We perform photometry on the mock images to estimate the success of these instruments for detecting high-z galaxies. We predict that JWST will have unprecedented power in detecting high-z galaxies, with a 95% completeness limit at least 2.5 magnitudes fainter than VISTA and Subaru, 1.1 magnitudes fainter than Hubble, and 0.9 magnitudes fainter than Roman, for the same wavelength and exposure time. Focusing on JWST, we consider a range of exposure times and filters, and find that the NIRCam F356W and F277W filters will detect the faintest galaxies, with 95% completeness at m=27.4 mag in 10ks exposures. We also predict the number of high-z galaxies that will be discovered by upcoming JWST imaging surveys. We predict that the COSMOS-Web survey will detect ~1000 MUV<-20.1 mag galaxies at 6.5<z<7.5, by virtue of its large survey area. JADES-Medium will detect almost 100% of MUV<-20 mag galaxies at z<8.5 due to its significant depth, however with its smaller survey area it will detect only ~100 of these galaxies at 6.5<z<7.5. Cosmic variance results in a large range in the number of predicted galaxies each survey will detect, which is more evident in smaller surveys such as CEERS and the PEARLS NEP and GOODS-S fields.

astro-ph.GA↗

arXiv2022

Deep Learning nearby galaxy peculiar velocities

We explore how information in images of nearby galaxies can be used to estimate their distance. We train a convolutional Neural Network (NN) to do this, using galaxy images from the Illustris simulation. We show that if the NN is trained on data with random errors added to the true distance (representing training using spectroscopic redshift instead of actual distance), then the NN can predict distances in a test dataset with greater accuracy than it was given in the training set. This is not unusual, as often NNs are trained on data with added noise, in order to increase robustness. In this case, however, it offers a route to estimating peculiar velocities of nearby galaxies. Given a galaxy with a known spectroscopic redshift one can use the NN-predicted distance to make an estimate of the peculiar velocity. Trying this using relatively low resolution (1.4 arcsec per pixel) simulated galaxy images we find fractional RMS distance errors of 7.7% for galaxies at a mean distance of 75 Mpc from the observer, leading to RMS peculiar velocity errors of 440 km/s. In a companion paper we apply the technique to 145,115 nearby galaxies from the NASA Sloan Atlas.

astro-ph.CO↗

arXiv2022

The Impact of Dust on the Sizes of Galaxies in the Epoch of Reionization

We study the sizes of galaxies in the Epoch of Reionization using a sample of ~100,000 galaxies from the BlueTides cosmological hydrodynamical simulation from z=7 to 11. We measure the galaxy sizes from stellar mass and luminosity maps, defining the effective radius as the minimum radius which could enclose the pixels containing 50% of the total mass/light in the image. We find an inverse relationship between stellar mass and effective half-mass radius, suggesting that the most massive galaxies are more compact and dense than lower mass galaxies, which have flatter mass distributions. We find a mildly negative relation between intrinsic far-ultraviolet luminosity and size, while we find a positive size-luminosity relation when measured from dust-attenuated images. This suggests that dust is the predominant cause of the observed positive size-luminosity relation, with dust preferentially attenuating bright sight lines resulting in a flatter emission profile and thus larger measured effective radii. We study the size-luminosity relation across the rest-frame ultraviolet and optical, and find that the slope decreases at longer wavelengths; this is a consequence of the relation being caused by dust, which produces less attenuation at longer wavelengths. We find that the far-ultraviolet size-luminosity relation shows mild evolution from z=7 to 11, and galaxy size evolves with redshift as $R\propto(1+z)^{-m}$, where $m=0.662\pm0.009$. Finally, we investigate the sizes of z=7 quasar host galaxies, and find that while the intrinsic sizes of quasar hosts are small relative to the overall galaxy sample, they have comparable sizes when measured from dust-attenuated images.

astro-ph.GA↗

arXiv2021

Deep Forest: Neural Network reconstruction of intergalactic medium temperature

We explore the use of Deep Learning to infer the temperature of the intergalactic medium from the transmitted flux in the high redshift Lyman-alpha forest. We train Neural Networks on sets of simulated spectra from redshift z=2-3 outputs of cosmological hydrodynamic simulations, including high temperature regions added in post-processing to approximate bubbles heated by Helium-II reionization. We evaluate how well the trained networks are able to reconstruct the temperature from the effect of Doppler broadening in the simulated input Lyman-alpha forest absorption spectra. We find that for spectra with high resolution (10 km/s pixel) and moderate signal to noise (20-50), the neural network is able to reconstruct the IGM temperature smoothed on scales of 6 Mpc/h quite well. Concentrating on discontinuities we find that high temperature regions of width 25 Mpc/h and temperature 20,000 K can be fairly easily detected and characterized. We show an example where multiple sightlines are combined to yield tomographic images of hot bubbles. Deep Learning techniques may be useful in this way to help us understand the complex temperature structure of the intergalactic medium around the time of Helium reionization.

astro-ph.CO↗

arXiv2021

AI-assisted super-resolution cosmological simulations II: Halo substructures, velocities and higher order statistics

In this work, we expand and test the capabilities of our recently developed super-resolution (SR) model to generate high-resolution (HR) realizations of the full phase-space matter distribution, including both displacement and velocity, from computationally cheap low-resolution (LR) cosmological N-body simulations. The SR model enhances the simulation resolution by generating 512 times more tracer particles, extending into the deeply non-linear regime where complex structure formation processes take place. We validate the SR model by deploying the model in 10 test simulations of box size 100 Mpc/h, and examine the matter power spectra, bispectra and 2D power spectra in redshift space. We find the generated SR field matches the true HR result at percent level down to scales of k ~ 10 h/Mpc. We also identify and inspect dark matter halos and their substructures. Our SR model generate visually authentic small-scale structures, that cannot be resolved by the LR input, and are in good statistical agreement with the real HR results. The SR model performs satisfactorily on the halo occupation distribution, halo correlations in both real and redshift space, and the pairwise velocity distribution, matching the HR results with comparable scatter, thus demonstrating its potential in making mock halo catalogs. The SR technique can be a powerful and promising tool for modelling small-scale galaxy formation physics in large cosmological volumes.

astro-ph.CO↗

arXiv2021

Deep Forest: Neural Network reconstruction of the Lyman-alpha forest

We explore the use of Deep Learning to infer physical quantities from the observable transmitted flux in the Lyman-alpha forest. We train a Neural Network using redshift z=3 outputs from cosmological hydrodynamic simulations and mock datasets constructed from them. We evaluate how well the trained network is able to reconstruct the optical depth for Lyman-alpha forest absorption from noisy and often saturated transmitted flux data. The Neural Network outperforms an alternative reconstruction method involving log inversion and spline interpolation by approximately a factor of 2 in the optical depth root mean square error. We find no significant dependence in the improvement on input data signal to noise, although the gain is greatest in high optical depth regions. The Lyman-alpha forest optical depth studied here serves as a simple, one dimensional, example but the use of Deep Learning and simulations to approach the inverse problem in cosmology could be extended to other physical quantities and higher dimensional data.

astro-ph.CO↗

arXiv2021

Observing the host galaxies of high-redshift quasars with JWST: predictions from the BlueTides simulation

The bright emission from high-redshift quasars completely conceals their host galaxies in the rest-frame ultraviolet/optical, with detection of the hosts in these wavelengths eluding even the Hubble Space Telescope (HST) using detailed point spread function (PSF) modelling techniques. In this study we produce mock images of a sample of z=7 quasars extracted from the BlueTides simulation, and apply Markov Chain Monte Carlo-based PSF modelling to determine the detectability of their host galaxies with the James Webb Space Telescope (JWST). While no statistically significant detections are made with HST, we predict that at the same wavelengths and exposure times JWST NIRCam imaging will detect ~50% of quasar host galaxies. We investigate various observational strategies, and find that NIRCam wide-band imaging in the long-wavelength filters results in the highest fraction of successful quasar host detections, detecting >80% of the hosts of bright quasars in exposure times of 5 ks. Exposure times of ~5 ks are required to detect the majority of host galaxies in the NIRCam wide-band filters, however even 10 ks exposures with MIRI result in <30% successful host detections. We find no significant trends between galaxy properties and their detectability. The PSF modelling can accurately recover the host magnitudes, radii, and spatial distribution of the larger-scale emission, when accounting for the central core being contaminated by residual quasar flux. Care should be made when interpreting the host properties measured using PSF modelling.

astro-ph.GA↗