arXiv subjects

J. Stuart B. Wyithe

Publications and source records attributed to J. Stuart B. Wyithe.

At least 37 records · Page 2Linked to original sources

arXiv2023

Thermal and Reionisation History within a Large-Volume Semi-Analytic Galaxy Formation Simulation

We predict the 21-cm global signal and power spectra during the Epoch of Reionisation using the MERAXES semi-analytic galaxy formation and reionisation model, updated to include X-ray heating and thermal evolution of the intergalactic medium. Studying the formation and evolution of galaxies together with the reionisation of cosmic hydrogen using semi-analytic models (such as MERAXES) requires N-body simulations within large volumes and high mass resolutions. For this, we use a simulation of side-length $210~h^{-1}$ Mpc with $4320^3$ particles resolving dark matter haloes to masses of $5\times10^8~h^{-1}~M_\odot$. To reach the mass resolution of atomically cooled galaxies, thought to be the dominant population contributing to reionisation, at $z=20$ of $\sim 2\times10^7~h^{-1}~M_\odot$, we augment this simulation using the DARKFOREST Monte-Carlo merger tree algorithm (achieving an effective particle count of $\sim10^{12}$). Using this augmented simulation we explore the impact of mass resolution on the predicted reionisation history as well as the impact of X-ray heating on the 21-cm global signal and the 21-cm power spectra. We also explore the cosmic variance of 21-cm statistics within $70^{3}$ $h^{-3}$ Mpc$^3$ sub-volumes. We find that the midpoint of reionisation varies by $Δz\sim0.8$ and that the cosmic variance on the power spectrum is underestimated by a factor of $2-4$ at $k\sim 0.1-0.4$ Mpc$^{-1}$ due to the non-Gaussian nature of the 21-cm signal. To our knowledge, this work represents the first model of both reionisation and galaxy formation which resolves low-mass atomically cooled galaxies while simultaneously sampling sufficiently large scales necessary for exploring the effects of X-rays in the early Universe.

astro-ph.CO↗

arXiv2023

The role of Pop III stars and early black holes in the 21cm signal from Cosmic Dawn

Modeling the 21cm global signal from the Cosmic Dawn is challenging due to the many poorly constrained physical processes that come into play. We address this problem using the semi-analytical code "Cosmic Archaeology Tool" (CAT). CAT follows the evolution of dark matter halos tracking their merger history and provides an ab initio description of their baryonic evolution, starting from the formation of the first (Pop III) stars and black holes (BHs) in mini-halos at z > 20. The model is anchored to observations of galaxies and AGN at z < 6 and predicts a reionization history consistent with constraints. In this work we compute the evolution of the mean global 21cm signal between $4\leq z \leq 40$ based on the rate of formation and emission properties of stars and accreting black holes. We obtain an absorption profile with a maximum depth $δ{\rm T_b} = -95$ mK at $z \sim 26.5$ (54 MHz). This feature is quickly suppressed turning into an emission signal at $z = 20$ due to the contribution of accreting BHs that efficiently heat the IGM at $z < 27$. The high-$z$ absorption feature is caused by the early coupling between the spin and kinetic temperature of the IGM induced by Pop III star formation episodes in mini-halos. Once we account for an additional radio background from early BHs, we are able to reproduce the timing and the depth of the EDGES signal only if we consider a smaller X-ray background from accreting BHs, but not the shape.

astro-ph.CO↗

arXiv2022

JWST's PEARLS: Prime Extragalactic Areas for Reionization and Lensing Science: Project Overview and First Results

We give an overview and describe the rationale, methods, and first results from NIRCam images of the JWST "Prime Extragalactic Areas for Reionization and Lensing Science" ("PEARLS") project. PEARLS uses up to eight NIRCam filters to survey several prime extragalactic survey areas: two fields at the North Ecliptic Pole (NEP); seven gravitationally lensing clusters; two high redshift proto-clusters; and the iconic backlit VV 191 galaxy system to map its dust attenuation. PEARLS also includes NIRISS spectra for one of the NEP fields and NIRSpec spectra of two high-redshift quasars. The main goal of PEARLS is to study the epoch of galaxy assembly, AGN growth, and First Light. Five fields, the JWST NEP Time-Domain Field (TDF), IRAC Dark Field (IDF), and three lensing clusters, will be observed in up to four epochs over a year. The cadence and sensitivity of the imaging data are ideally suited to find faint variable objects such as weak AGN, high-redshift supernovae, and cluster caustic transits. Both NEP fields have sightlines through our Galaxy, providing significant numbers of very faint brown dwarfs whose proper motions can be studied. Observations from the first spoke in the NEP TDF are public. This paper presents our first PEARLS observations, their NIRCam data reduction and analysis, our first object catalogs, the 0.9-4.5 $μ$m galaxy counts and Integrated Galaxy Light. We assess the JWST sky brightness in 13 NIRCam filters, yielding our first constraints to diffuse light at 0.9-4.5 μm. PEARLS is designed to be of lasting benefit to the community.

astro-ph.CO↗

arXiv2022

Generating extremely large-volume reionisation simulations

Preparing for the first detection of the cosmic 21-cm signal from large-scale interferometer experiments requires rigorous testing of the data analysis and reduction pipelines. To validate that these pipelines do not erroneously remove or add features that can mimic the cosmic signal (e.g. from side-lobes or large-scale power leakage), we require reionisation simulations larger than the experiments primary field of view. For an experiment such as the MWA, with a field of view of $\sim25^{2}$ deg.$^{2}$, this would require a simulation of several Gpcs, which is currently infeasible. To overcome this, we developed a simplified version of the semi-numerical reionisation simulation code 21CMFAST preferencing large volumes over some physical accuracy by assuming linear theory for structure formation. With this, we constructed a 7.5 Gpc comoving volume with voxel resolution of $\sim1.17$ cMpc tailored specifically to the binned spectral resolution of the MWA. This simulation was used for validating the pipelines for the 2020 MWA 21-cm power spectrum (PS) upper limits (Trott et al.). We then use this large-volume simulation to explore: (i) whether smaller volume simulations are biased by the missing large-scale modes, (ii) non-Gaussianity in estimates of the cosmic variance, (iii) biases in the recovered 21-cm PS following foreground wedge removal and (iv) the impact of tiling smaller volume simulations to achieve extremely large volumes. In summary, we find: (i) no biases from missing large-scale power, (ii) significant contribution from non-Gaussianity in the cosmic variance as expected following Mondal et al. (iii) an over-estimate of the 21-cm PS of 10-20 per cent following wedge mode excision for our particular model and (iv) tiling smaller volume simulations under-estimates the large-scale power and also the estimated cosmic variance.

astro-ph.CO↗

arXiv2021

Dark-ages Reionization and Galaxy Formation Simulation XX. The Ly$α$ IGM transmission properties and environment of bright galaxies during the Epoch of Reionization

The highly neutral inter-galactic medium (IGM) during the Epoch of Reionization (EoR) is expected to suppress Ly$α$ emission with damping-wing absorption, causing nearly no Ly$α$ detection from star-forming galaxies at $z{\sim}8$. However, spectroscopic observations of the 4 brightest galaxies (${\rm H}_{160}{\sim}25$ mag) at these redshifts do reveal prominent Ly$α$ line, suggesting locally ionised IGM. In this paper, we explore the Ly$α$ IGM transmission and environment of bright galaxies during the EoR using the Meraxes semi-analytic model. We find brighter galaxies to be less affected by damping-wing absorption as they are effective at ionizing surrounding neutral hydrogen. Specifically, the brightest sources (${\rm H}_{160}{\lesssim}25.5$ mag) lie in the largest ionized regions in our simulation, and have low attenuation of their Ly$α$ from the IGM (optical depth ${<}1$). Fainter galaxies (25.5 mag${<}{\rm H}_{160}{<}27.5$ mag) have transmission that depends on UV luminosity, leading to a lower incidence of Ly$α$ detection at fainter magnitudes. This luminosity-dependent attenuation explains why Ly$α$ has only been observed in the brightest galaxies at $z{\sim}8$. Follow-up observations have revealed counterparts in the vicinity of these confirmed $z{\sim}8$ Ly$α$ emitters. The environments of our modelled analogues agree with these observations in the number of nearby galaxies, which is a good indicator of whether Ly$α$ can be detected among fainter galaxies. At the current observational limit, galaxies with ${\ge}2$--5 neighbours within $2'{\times}2'$ are ${\sim}2$--3 times more likely to show Ly$α$ emission. JWST will discover an order of magnitude more neighbours, revealing ${\gtrsim}50$ galaxies in the largest ionizing bubbles and facilitating direct study of reionization morphology.

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↗

arXiv2020

An efficient hybrid method to produce high resolution large volume dark matter simulations for semi-analytic models of reionisation

Resolving faint galaxies in large volumes is critical for accurate cosmic reionisation simulations. While less demanding than hydrodynamical simulations, semi-analytic reionisation models still require very large N-body simulations in order to resolve the atomic cooling limit across the whole reionisation history within box sizes $\gtrsim 100 \, h^{-1} {\rm Mpc}$. To facilitate this, we extend the mass resolution of N-body simulations using a Monte Carlo algorithm. We also propose a method to evolve positions of Monte Carlo halos, which can be an input for semi-analytic reionisation models. To illustrate, we present an extended halo catalogue that reaches a mass resolution of $M_\text{halo} = 3.2 \times 10^7 \, h^{-1} \text{M}_\odot$ in a $105 \, h^{-1} {\rm Mpc}$ box, equivalent to an N-body simulation with $\sim 6800^3$ particles. The resulting halo mass function agrees with smaller volume N-body simulations with higher resolution. Our results also produce consistent two-point correlation functions with analytic halo bias predictions. The extended halo catalogues are applied to the \textsc{meraxes} semi-analytic reionisation model, which improves the predictions on stellar mass functions, star formation rate densities and volume-weighted neutral fractions. Comparison of high resolution large volume simulations with both small volume or low resolution simulations confirms that both low resolution and small volume simulations lead to reionisation ending too rapidly. Lingering discrepancies between the star formation rate functions predicted with and without our extensions can be traced to the uncertain contribution of satellite galaxies.

astro-ph.GA↗

arXiv2020

The host galaxies of z=7 quasars: predictions from the BlueTides simulation

We examine the properties of the host galaxies of $z=7$ quasars using the large volume, cosmological hydrodynamical simulation BlueTides. We find that the 10 most massive black holes and the 191 quasars in the simulation (with $M_{\textrm{UV,AGN}} 10^{6.5}M_\odot$ black hole, indicating that minor mergers play at least some role in growing black holes in the early Universe.

astro-ph.GA↗

arXiv2020

Exploring reionisation and high-z galaxy observables with recent multi-redshift MWA upper limits on the 21-cm signal

We use the latest multi-redshift ($z=6.5-8.7$) upper limits on the 21-cm signal from the Murchison Widefield Array (MWA) to explore astrophysical models which are inconsistent with the data. These upper limits are achieved using 298 h of carefully excised data over four observing seasons. To explore these upper limits in the context of reionisation astrophysics, we use 21CMMC. We then connect the disfavoured regions of parameter space to existing observational constraints on reionisation such as high-$z$ galaxy ultra-violet (UV) luminosity functions, background UV photoionisation rate, intergalactic medium (IGM) neutral fraction, the electron scattering optical depth and the soft-band X-ray emissivity. We find the vast majority of models disfavoured by the MWA limits are already inconsistent with existing observational constraints. These inconsistent models arise from two classes of models: (i) `cold' reionisation and (ii) pure matter density fluctuations (i.e. no reionisation). However, a small subsample of models are consistent implying the existing MWA limits provide unique information in disfavouring models of reionisation, albeit extremely weakly. We also provide the first limits on the soft-band X-ray emissivity from galaxies at high redshifts, finding $1σ$ lower limits of $ε_{{\rm X},0.5-2~{\rm keV}}\gtrsim10^{34.5}$ erg s$^{-1}$ Mpc$^{-3}$. Finally, we recover 95 per cent disfavoured limits on the IGM spin temperature of $\bar{T}_{\rm S}\lesssim$ 1.3, 1.4, 1.5, 1.8, 2.1, 2.4 K at $z=6.5, 6.8, 7.1, 7.8, 8.2, 8.7$. With this we infer the IGM must have undergone, at the very least, a small amount of X-ray heating. Note, the limits on $ε_{{\rm X},0.5-2~{\rm keV}}$ and $\bar{T}_{\rm S}$ are conditional on the IGM neutral fraction.

astro-ph.CO↗

arXiv2020

Limits to Rest-Frame Ultraviolet Emission From Far-Infrared-Luminous z~6 Quasar Hosts

We report on a Hubble Space Telescope search for rest-frame ultraviolet emission from the host galaxies of five far-infrared-luminous $z\simeq{}6$ quasars and the $z=5.85$ hot-dust free quasar SDSS J0005-0006. We perform 2D surface brightness modeling for each quasar using a Markov-Chain Monte-Carlo estimator, to simultaneously fit and subtract the quasar point source in order to constrain the underlying host galaxy emission. We measure upper limits for the quasar host galaxies of $m_J>22.7$ mag and $m_H>22.4$ mag, corresponding to stellar masses of $M_\ast<2\times10^{11}M_\odot$. These stellar mass limits are consistent with the local $M_{\textrm{BH}}$-$M_\ast$ relation. Our flux limits are consistent with those predicted for the UV stellar populations of $z\simeq6$ host galaxies, but likely in the presence of significant dust ($\langle A_{\mathrm{UV}}\rangle\simeq 2.6$ mag). We also detect a total of up to 9 potential $z\simeq6$ quasar companion galaxies surrounding five of the six quasars, separated from the quasars by 1.4''-3.2'', or 8.4-19.4 kpc, which may be interacting with the quasar hosts. These nearby companion galaxies have UV absolute magnitudes of -22.1 to -19.9 mag, and UV spectral slopes $β$ of -2.0 to -0.2, consistent with luminous star-forming galaxies at $z\simeq6$. These results suggest that the quasars are in dense environments typical of luminous $z\simeq6$ galaxies. However, we cannot rule out the possibility that some of these companions are foreground interlopers. Infrared observations with the James Webb Space Telescope will be needed to detect the $z\simeq6$ quasar host galaxies and better constrain their stellar mass and dust content.

astro-ph.GA↗

arXiv2020

A Strong-Lensing Model for the WMDF JWST/GTO Very Rich Cluster Abell 1489

We present a first strong-lensing model for the galaxy cluster RM J121218.5+273255.1 ($z=0.35$; hereafter RMJ1212; also known as Abell 1489). This cluster is amongst the top 0.1\% richest clusters in the redMaPPer catalog; it is significantly detected in X-ray and through the Sunyaev-Zel'dovich effect in ROSAT and \emph{Planck} data, respectively; and its optical luminosity distribution implies a very large lens, following mass-to-light scaling relations. Based on these properties it was chosen for the Webb Medium Deep Fields (WMDF) JWST/GTO program. In preparation for this program, RMJ1212 was recently imaged with GMOS on Gemini North and in seven optical and near-infrared bands with the \emph{Hubble Space Telescope}. We use these data to map the inner mass distribution of the cluster, uncovering various sets of multiple images. We also search for high-redshift candidates in the data, as well as for transient sources. We find over a dozen high-redshift ($z\gtrsim6$) candidates based on both photometric redshift and the dropout technique. No prominent ($\gtrsim5 σ$) transients were found in the data between the two HST visits. Our lensing analysis reveals a relatively large lens with an effective Einstein radius of $θ_{E}\simeq32\pm3''$ ($z_{s}=2$), in broad agreement with the scaling-relation expectations. RMJ1212 demonstrates that powerful lensing clusters can be selected in a robust and automated way following the light-traces-mass assumption.

astro-ph.GA↗

arXiv2020

Dark-ages reionization and galaxy formation simulation -- XVIII. The high-redshift evolution of black holes and their host galaxies

Correlations between black holes and their host galaxies provide insight into what drives black hole-host co-evolution. We use the Meraxes semi-analytic model to investigate the growth of black holes and their host galaxies from high redshift to the present day. Our modelling finds no significant evolution in the black hole-bulge and black hole-total stellar mass relations out to a redshift of 8. The black hole-total stellar mass relation has similar but slightly larger scatter than the black hole-bulge relation, with the scatter in both decreasing with increasing redshift. In our modelling the growth of galaxies, bulges and black holes are all tightly related, even at the highest redshifts. We find that black hole growth is dominated by instability-driven or secular quasar-mode growth and not by merger-driven growth at all redshifts. Our model also predicts that disc-dominated galaxies lie on the black hole-total stellar mass relation, but lie offset from the black hole-bulge mass relation, in agreement with recent observations and hydrodynamical simulations.

astro-ph.GA↗

arXiv2019

Dark-Ages Reionisation & Galaxy Formation Simulation XVI: The Thermal Memory of Reionisation

Intergalactic medium temperature is a powerful probe of the epoch of reionisation, as information is retained long after reionisation itself. However, mean temperatures are highly degenerate with the timing of reionisation, with the amount heat injected during the epoch, and with the subsequent cooling rates. We post-process a suite of semi-analytic galaxy formation models to characterise how different thermal statistics of the intergalactic medium can be used to constrain reionisation. Temperature is highly correlated with redshift of reionisation for a period of time after the gas is heated. However as the gas cools, thermal memory of reionisation is lost, and a power-law temperature-density relation is formed, $T = T_0(1+δ)^{1-γ}$ with $γ\approx 1.5$. Constraining our model against observations of electron optical depth and temperature at mean density, we find that reionisation likely finished at $z_{\rm{reion}} = 6.8 ^{+ 0.5} _{-0.8}$ with a soft spectral slope of $α= 2.8 ^{+ 1.2} _{-1.0}$. By restricting spectral slope to the range $[0.5,2.5]$ motivated by population II synthesis models, reionisation timing is further constrained to $z_{\rm{reion}} = 6.9 ^{+ 0.4} _{-0.5}$. We find that, in the future, the degeneracies between reionisation timing and background spectrum can be broken using the scatter in temperatures and integrated thermal history.

astro-ph.CO↗

arXiv2019

Dark-ages Reionization and Galaxy Formation Simulation -- XIX: Predictions of infrared excess and cosmic star formation rate density from UV observations

We present a new analysis of high-redshift UV observations using a semi-analytic galaxy formation model, and provide self-consistent predictions of the infrared excess (IRX) -- $β$ relations and cosmic star formation rate density. We combine the Charlot & Fall dust attenuation model with the Meraxes semi-analytic model, and explore three different parametrisations for the dust optical depths, linked to star formation rate, dust-to-gas ratio and gas column density respectively. A Bayesian approach is employed to statistically calibrate model free parameters including star formation efficiency, mass loading factor, dust optical depths and reddening slope directly against UV luminosity functions and colour-magnitude relations at z ~ 4-7. The best-fit models show excellent agreement with the observations. We calculate IRX using energy balance arguments, and find that the large intrinsic scatter in the IRX -$β$ plane is driven by the specific star formation rate. Additionally, the difference among the three dust models suggests a factor of two systematic uncertainty in the dust-corrected star formation rate when using the Meurer IRX - $β$ relation at z > 4.

astro-ph.GA↗

arXiv2019

Dark-ages Reionization and Galaxy Formation Simulation -- XVII. Sizes, angular momenta and morphologies of high redshift galaxies

We study the sizes, angular momenta and morphologies of high-redshift galaxies using an update of the Meraxes semi-analytic galaxy evolution model. Our model successfully reproduces a range of observations from redshifts $z=0$-$10$. We find that the effective radius of a galaxy disc scales with UV luminosity as $R_e\propto L_{\textrm{UV}}^{0.33}$ at $z=5$-$10$, and with stellar mass as $R_e\propto M_\ast^{0.24}$ at $z=5$ but with a slope that increases at higher redshifts. Our model predicts that the median galaxy size scales with redshift as $R_e \propto (1+z)^{-m}$, where $m=1.98\pm0.07$ for galaxies with $(0.3$-$1)L^\ast_{z=3}$ and $m=2.15\pm0.05$ for galaxies with $(0.12$-$0.3)L^\ast_{z=3}$. We find that the ratio between stellar and halo specific angular momentum is typically less than one and decreases with halo and stellar mass. This relation shows no redshift dependence, while the relation between specific angular momentum and stellar mass decreases by $\sim0.5$ dex from $z=7$ to $z=2$. Our model reproduces the distribution of local galaxy morphologies, with bulges formed predominantly through galaxy mergers for low-mass galaxies, disc-instabilities for galaxies with $M_\ast\simeq10^{10}$-$10^{11.5}M_\odot$, and major mergers for the most massive galaxies. At high redshifts, we find galaxy morphologies that are predominantly bulge-dominated.

astro-ph.GA↗

arXiv2018

Dependence of galaxy clustering on UV-luminosity and stellar mass at $z \sim 4 - 7$

We investigate the dependence of galaxy clustering at $z \sim 4 - 7$ on UV-luminosity and stellar mass. Our sample consists of $\sim$ 10,000 Lyman-break galaxies (LBGs) in the XDF and CANDELS fields. As part of our analysis, the $M_\star - M_{\rm UV}$ relation is estimated for the sample, which is found to have a nearly linear slope of $d\log_{10} M_\star / d M_{\rm UV} \sim 0.44$. We subsequently measure the angular correlation function and bias in different stellar mass and luminosity bins. We focus on comparing the clustering dependence on these two properties. While UV-luminosity is only related to recent starbursts of a galaxy, stellar mass reflects the integrated build-up of the whole star formation history, which should make it more tightly correlated with halo mass. Hence, the clustering segregation with stellar mass is expected to be larger than with luminosity. However, our measurements suggest that the segregation with luminosity is larger with $\simeq 90\%$ confidence (neglecting contributions from systematic errors). We compare this unexpected result with predictions from the \textsc{Meraxes} semi-analytic galaxy formation model. Interestingly, the model reproduces the observed angular correlation functions, and also suggests stronger clustering segregation with luminosity. The comparison between our observations and the model provides evidence of multiple halo occupation in the small scale clustering.

astro-ph.GA↗

arXiv2018

Dark-ages Reionization and Galaxy Formation Simulation - XV. Stellar evolution and feedback in dwarf galaxies at high redshift

We directly compare predictions of dwarf galaxy properties in a semi-analytic model (SAM) with those extracted from a high-resolution hydrodynamic simulation. We focus on galaxies with halo masses of 1e9 10$), with the relevant time-scale becoming significantly longer towards lower redshifts. This indicates efficient accretion in cold mode in these low-mass objects at high redshift. Note that the improved SAM, which has been calibrated against hydrodynamic simulations, can provide more accurate predictions of high-redshift dwarf galaxy properties that are essential for reionization study.

astro-ph.GA↗

arXiv2018

Dark-ages Reionization and Galaxy Formation Simulation - XIV. Gas accretion, cooling and star formation in dwarf galaxies at high redshift

We study dwarf galaxy formation at high redshift ($z\ge5$) using a suite of high- resolution, cosmological hydrodynamic simulations and a semi-analytic model (SAM). We focus on gas accretion, cooling and star formation in this work by isolating the relevant process from reionization and supernova feedback, which will be further discussed in a companion paper. We apply the SAM to halo merger trees constructed from a collisionless N-body simulation sharing identical initial conditions to the hydrodynamic suite, and calibrate the free parameters against the stellar mass function predicted by the hydrodynamic simulations at z = 5. By making comparisons of the star formation history and gas components calculated by the two modelling techniques, we find that semi-analytic prescriptions that are commonly adopted in the literature of low-redshift galaxy formation do not accurately represent dwarf galaxy properties in the hydrodynamic simulation at earlier times. We propose 3 modifications to SAMs that will provide more accurate high-redshift simulations. These include 1) the halo mass and baryon fraction which are overestimated by collisionless N-body simulations; 2) the star formation efficiency which follows a different cosmic evolutionary path from the hydrodynamic simulation; and 3) the cooling rate which is not well defined for dwarf galaxies at high redshift. Accurate semi-analytic modelling of dwarf galaxy formation informed by detailed hydrodynamical modelling will facilitate reliable semi-analytic predictions over the large volumes needed for the study of reionization.

astro-ph.CO↗