Search

The interplay between atomic gas, the star-formation history of a galaxy and its environment are intrinsically linked, and we need to decouple these dependencies to understand their role in galaxy formation and evolution. In this paper, we analyse the star formation histories (SFHs) of 187 galaxies from the MIGHTEE-HI Survey Early Science Release data, focusing on the relationships between HI properties and star formation. A strong correlation emerges between a galaxy's HI-to-stellar mass ratio and the time of formation, alongside an inverse correlation between stellar mass and time of formation, regardless of the inferred SFH. Additionally, galaxies with lower stellar masses and higher HI-to-stellar mass ratios exhibit longer gas depletion times compared to more massive galaxies, which appear to have depleted their gas and formed stars more efficiently. This suggests that smaller, gas-rich galaxies have higher depletion times due to shallower potential wells and less efficient star formation. Furthermore, we explore the connection between spin-filament alignment and HI content. We find no significant correlation between peak star formation activity and proximity to filaments. However, we do find that the two galaxies in our sample within 1 Mpc of a filament have very low gas-depletion timescales and have their spin axis misaligned with the filament, suggestive of a link between the galaxy properties and proximity to a filament., Comment: submitted to MNRAS; 14 pages, 14 figures

One of the surprising early findings with JWST has been the discovery of a strong "roll-over" or a softening of the absorption edge of Ly$\alpha$ in a large number of galaxies at ($z\gtrsim 6$), in addition to systematic offsets from photometric redshift estimates and fundamental galaxy scaling relations. This has been interpreted as damped Ly$\alpha$ absorption (DLA) wings from high column densities of neutral atomic hydrogen (HI), signifying major gas accretion events in the formation of these galaxies. To explore this new phenomenon systematically, we assemble the JWST/NIRSpec PRImordial gas Mass AssembLy (PRIMAL) legacy survey of 494 galaxies at $z=5.5-13.4$. We characterize this benchmark sample in full and spectroscopically derive the galaxy redshifts, metallicities, star-formation rates, and ultraviolet slopes. We define a new diagnostic, the Ly$\alpha$ damping parameter $D_{\rm Ly\alpha}$ to measure and quantify the Ly$\alpha$ emission strength, HI fraction in the IGM, or local HI column density for each source. The JWST-PRIMAL survey is based on the spectroscopic DAWN JWST Archive (DJA-Spec). All the software, reduced spectra, and spectroscopically derived quantities and catalogs are made publicly available in dedicated repositories. The fraction of strong galaxy DLAs are found to be in the range $65-95\%$ at $z>5.5$. The fraction of strong Ly$\alpha$ emitters (LAEs) is found to increase with decreasing redshift, in qualitative agreement with previous observational results, and are predominantly associated with low-metallicity and UV faint galaxies. By contrast, strong DLAs are observed in galaxies with a variety of intrinsic physical properties. Our results indicate that strong DLAs likely reflect a particular early assembly phase of reionization-era galaxies, at which point they are largely dominated by pristine HI gas accretion. [abridged], Comment: 20 pages, 18 figures, 3 tables. Submitted to A&A. Comments welcome! All data and catalogs are available through the DAWN JWST Archive (DJA): https://dawn-cph.github.io/dja/ and https://github.com/keheintz/jwst-primal

Galaxy cluster outskirts mark the transition region from the mildly non-linear cosmic web to the highly non-linear, virialised, cluster interior. It is in this transition region that the intra-cluster medium (ICM) begins to influence the properties of accreting galaxies and groups, as ram pressure impacts a galaxy's cold gas content and subsequent star formation rate. Conversely, the thermodynamical properties of the ICM in this transition region should also feel the influence of accreting substructure (i.e. galaxies and groups), whose passage can drive shocks. In this paper, we use a suite of cosmological hydrodynamical zoom simulations of a single galaxy cluster, drawn from the nIFTy comparison project, to study how the dynamics of substructure accreted from the cosmic web influences the thermodynamical properties of the ICM in the cluster's outskirts. We demonstrate how features evident in radial profiles of the ICM (e.g. gas density and temperature) can be linked to strong shocks, transient and short-lived in nature, driven by the passage of substructure. The range of astrophysical codes and galaxy formation models in our comparison are broadly consistent in their predictions (e.g. agreeing when and where shocks occur, but differing in how strong shocks will be); this is as we would expect of a process driven by large-scale gravitational dynamics and strong, inefficiently radiating, shocks. This suggests that mapping such shock structures in the ICM in a cluster's outskirts (via e.g. radio synchrotron emission) could provide a complementary measure of its recent merger and accretion history., Comment: 15 pages, 13 figures, version accepted for publication in MNRAS

The MIGHTEE large survey project will survey four of the most well-studied extragalactic deep fields, totalling 20 square degrees to $\mu$Jy sensitivity at Giga-Hertz frequencies, as well as an ultra-deep image of a single ~1 square degree MeerKAT pointing. The observations will provide radio continuum, spectral line and polarisation information. As such, MIGHTEE, along with the excellent multi-wavelength data already available in these deep fields, will allow a range of science to be achieved. Specifically, MIGHTEE is designed to significantly enhance our understanding of, (i) the evolution of AGN and star-formation activity over cosmic time, as a function of stellar mass and environment, free of dust obscuration; (ii) the evolution of neutral hydrogen in the Universe and how this neutral gas eventually turns into stars after moving through the molecular phase, and how efficiently this can fuel AGN activity; (iii) the properties of cosmic magnetic fields and how they evolve in clusters, filaments and galaxies. MIGHTEE will reach similar depth to the planned SKA all-sky survey, and thus will provide a pilot to the cosmology experiments that will be carried out by the SKA over a much larger survey volume., Comment: Proceedings of Science, "MeerKAT Science: On the Pathway to the SKA", Stellenbosch, 25-27 May 2016

The Hydrogen Intensity and Real-time Analysis eXperiment (HIRAX) is a new 400-800MHz radio interferometer under development for deployment in South Africa. HIRAX will comprise 1024 six meter parabolic dishes on a compact grid and will map most of the southern sky over the course of four years. HIRAX has two primary science goals: to constrain Dark Energy and measure structure at high redshift, and to study radio transients and pulsars. HIRAX will observe unresolved sources of neutral hydrogen via their redshifted 21-cm emission line (`hydrogen intensity mapping'). The resulting maps of large-scale structure at redshifts 0.8-2.5 will be used to measure Baryon Acoustic Oscillations (BAO). HIRAX will improve upon current BAO measurements from galaxy surveys by observing a larger cosmological volume (larger in both survey area and redshift range) and by measuring BAO at higher redshift when the expansion of the universe transitioned to Dark Energy domination. HIRAX will complement CHIME, a hydrogen intensity mapping experiment in the Northern Hemisphere, by completing the sky coverage in the same redshift range. HIRAX's location in the Southern Hemisphere also allows a variety of cross-correlation measurements with large-scale structure surveys at many wavelengths. Daily maps of a few thousand square degrees of the Southern Hemisphere, encompassing much of the Milky Way galaxy, will also open new opportunities for discovering and monitoring radio transients. The HIRAX correlator will have the ability to rapidly and eXperimentciently detect transient events. This new data will shed light on the poorly understood nature of fast radio bursts (FRBs), enable pulsar monitoring to enhance long-wavelength gravitational wave searches, and provide a rich data set for new radio transient phenomena searches. This paper discusses the HIRAX instrument, science goals, and current status., Comment: 11 pages, 5 figures

The sources that drove cosmological reionization left clues regarding their identity in the slope and inhomogeneity of the ultraviolet ionizing background (UVB): Bright quasars (QSOs) generate a hard UVB with predominantly large-scale fluctuations while Population II stars generate a softer one with smaller-scale fluctuations. Metal absorbers probe the UVB's slope because different ions are sensitive to different energies. Likewise, they probe spatial fluctuations because they originate in regions where a galaxy-driven UVB is harder and more intense. We take a first step towards studying the reionization-epoch UVB's slope and inhomogeneity by comparing observations of 12 metal absorbers at $z\sim6$ versus predictions from a cosmological hydrodynamic simulation using three different UVBs: a soft, spatially-inhomogeneous "galaxies+QSOs" UVB; a homogeneous "galaxies+QSOs" UVB (Haardt & Madau 2012); and a QSOs-only model. All UVBs reproduce the observed column density distributions of CII, SiIV, and CIV reasonably well although high-column, high-ionization absorbers are underproduced, reflecting numerical limitations. With upper limits treated as detections, only a soft, fluctuating UVB reproduces both the observed SiIV/CIV and CII/CIV distributions. The QSOs-only UVB overpredicts both CIV/CII and CIV/SiIV, indicating that it is too hard. The Haardt & Madau (2012) UVB underpredicts CIV/SiIV, suggesting that it lacks amplifications near galaxies. Hence current observations prefer a soft, fluctuating UVB as expected from a predominantly Population II background although they cannot rule out a harder one. Future observations probing a factor of two deeper in metal column density will distinguish between the soft, fluctuating and QSOs-only UVBs., Comment: 13 pages, 10 figures, re-submitted to MNRAS after incorporating referee's suggestions

Observations suggest that CII was more abundant than CIV in the intergalactic medium towards the end of the hydrogen reionization epoch. This transition provides a unique opportunity to study the enrichment history of intergalactic gas and the growth of the ionizing background (UVB) at early times. We study how carbon absorption evolves from z=10-5 using a cosmological hydrodynamic simulation that includes a self-consistent multifrequency UVB as well as a well-constrained model for galactic outflows to disperse metals. Our predicted UVB is within 2-4 times that of Haardt & Madau (2012), which is fair agreement given the uncertainties. Nonetheless, we use a calibration in post-processing to account for Lyman-alpha forest measurements while preserving the predicted spectral slope and inhomogeneity. The UVB fluctuates spatially in such a way that it always exceeds the volume average in regions where metals are found. This implies both that a spatially-uniform UVB is a poor approximation and that metal absorption is not sensitive to the epoch when HII regions overlap globally even at column densites of 10^{12} cm^{-2}. We find, consistent with observations, that the CII mass fraction drops to low redshift while CIV rises owing the combined effects of a growing UVB and continued addition of carbon in low-density regions. This is mimicked in absorption statistics, which broadly agree with observations at z=6-3 while predicting that the absorber column density distributions rise steeply to the lowest observable columns. Our model reproduces the large observed scatter in the number of low-ionization absorbers per sightline, implying that the scatter does not indicate a partially-neutral Universe at z=6., Comment: 16 pages, 14 figures, accepted to MNRAS

We present evidence for a skewed distribution of UV FeII emission in quasars within candidate overdense regions spanning spatial scales of ~ 50 Mpc at 1.11 < z < 1.67, compared to quasars in field environments at comparable redshifts. The overdense regions have an excess of high equivalent width sources (W2400 > 42 \AA), and a dearth of low equivalent width sources. There are various possible explanations for this effect, including dust, Ly\alpha fluorescence, microturbulence, and iron abundance. We find that the most plausible of these is enhanced iron abundance in the overdense regions, consistent with an enhanced star formation rate in the overdense regions compared to the field., Comment: 9 pages, 5 figures, accepted for publication in MNRAS

We use a radiation hydrodynamic simulation of the hydrogen reionization epoch to study OI absorbers at z~6. The intergalactic medium (IGM) is reionized before it is enriched, hence OI absorption originates within dark matter halos. The predicted abundance of OI absorbers is in reasonable agreement with observations. At z=10, roughly 70% of sightlines through atomically-cooled halos encounter a visible (N_OI > 10^14 cm^-2) column. Reionization ionizes and removes gas from halos less massive than 10^8.4 M_0, but 20% of sightlines through more massive halos encounter visible columns even at z=5. The mass scale of absorber host halos is 10-100 times smaller than the halos of Lyman break galaxies and Lyman-alpha emitters, hence absorption probes the dominant ionizing sources more directly. OI absorbers have neutral hydrogen columns of 10^19-10^21 cm^-2, suggesting a close resemblance between objects selected in OI and HI absorption. Finally, the absorption in the foreground of the z=7.085 quasar ULASJ1120+0641 cannot originate in a dark matter halo because halo gas at the observed HI column density is enriched enough to violate the upper limits on the OI column. By contrast, gas at less than one third the cosmic mean density satisfies the constraints. Hence the foreground absorption likely originates in the IGM., Comment: 21 pages, 15 figures, re-submitted to MNRAS after many improvements suggested by referee, results unchanged

We present the first results from our pilot 500 ks Chandra-LETG Large Program observation of the soft X-ray brightest source in the z>=0.4 sky, the blazar 1ES 1553+113, aimed to secure the first uncontroversial detections of the missing baryons in the X-rays. We identify a total of 11 possible absorption lines, with single-line statistical significances between 2.2-4.1\sigma. Six of these lines are detected at high single-line statistical significance (3.6 <= sigma <= 4.1), while the remaining five are regarded as marginal detections in association with either other X-ray lines detected at higher significance and/or Far-Ultraviolet (FUV) signposts. In particular, five of these possible intervening absorption lines, are identified as CV and CVI K\alpha absorbers belonging to three WHIM systems at z_X = 0.312, z_X = 0.237 and = 0.133, which also produce broad HI (and OVI for the z_X = 0.312 system) absorption in the FUV. For two of these systems (z_X = 0.312 and 0.237), the Chandra X-ray data led the a-posteriori discovery of physically consistent broad HI associations in the FUV, so confirming the power of the X-ray-FUV synergy for WHIM studies. The true statistical significances of these three X-ray absorption systems, after properly accounting for the number of redshift trials, are 5.8 sigma (z_X = 0.312; 6.3 sigma if the low-significance OV and CV K-beta associations are considered), 3.9 sigma (z_X = 0.237), and 3.8 sigma (\langle z_X \rangle = 0.133), respectively., Comment: 27 pages, 9 figures, Accepted for publication by the ApJ

We use a suite of cosmological hydrodynamic simulations including a self-consistent treatment for inhomogeneous reionisation to study the impact of galactic outflows and photoionisation heating on the volume-averaged recombination rate of the intergalactic medium (IGM). By incorporating an evolving ionising escape fraction and a treatment for self-shielding within Lyman limit systems, we have run the first simulations of "photon-starved" reionisation scenarios that simultaneously reproduce observations of the abundance of galaxies, the optical depth to electron scattering of cosmic microwave background photons \tau, and the effective optical depth to Lyman\alpha absorption at z=5. We confirm that an ionising background reduces the clumping factor C by more than 50% by smoothing moderately-overdense (\Delta=1--100) regions. Meanwhile, outflows increase clumping only modestly. The clumping factor of ionised gas is much lower than the overall baryonic clumping factor because the most overdense gas is self-shielded. Photoionisation heating further suppresses recombinations if reionisation heats gas above the canonical 10,000 K. Accounting for both effects within our most realistic simulation, C rises from <1 at z>10 to 3.3 at z=6. We show that incorporating temperature- and ionisation-corrected clumping factors into an analytical reionisation model reproduces the numerical simulation's \tau to within 10%. Finally, we explore how many ionising photons are absorbed during the process of heating filaments by considering the overall photon cost of reionisation in analytical models that assume that the IGM is heated at different redshifts. For reionisation redshifts of 9--10, cold filaments boost the reionisation photon budget by ~1 photon per hydrogen atom., Comment: 19 pages, 8 figures; accepted to MNRAS

We have exploited the HST CANDELS WFC3/IR imaging to study the properties of (sub-)mm galaxies in GOODS-South. After using the deep radio and Spitzer imaging to identify galaxy counterparts for the (sub-)mm sources, we have used the new CANDELS data in two ways. First, we have derived improved photometric redshifts and stellar masses, confirming that the (sub-)mm galaxies are massive (=2.2x10^11 M_solar) galaxies at z=1-3. Second, we have exploited the depth and resolution of the WFC3/IR imaging to determine the sizes and morphologies of the galaxies at rest-frame optical wavelengths, fitting two-dimensional axi-symmetric Sersic models. Crucially, the WFC3/IR H-band imaging enables modelling of the mass-dominant galaxy, rather than the blue high-surface brightness features which often dominate optical (rest-frame UV) images of (sub-)mm galaxies, and can confuse visual morphological classification. As a result of this analysis we find that >95% of the rest-frame optical light in almost all of the (sub-)mm galaxies is well-described by either a single exponential disk, or a multiple-component system in which the dominant constituent is disk-like. We demonstrate that this conclusion is consistent with the results of high-quality ground-based K-band imaging, and explain why. The massive disk galaxies which host luminous (sub-)mm emission are reasonably extended (r_e=4 kpc), consistent with the sizes of other massive star-forming disks at z~2. In many cases we find evidence of blue clumps within the sources, with the mass-dominant disk becoming more significant at longer wavelengths. Finally, only a minority of the sources show evidence for a major galaxy-galaxy interaction. Taken together, these results support the view that most (sub-)mm galaxies at z~2 are simply the most extreme examples of normal star-forming galaxies at that era., Comment: 30 pages, 9 figures

We present two-dimensional, integral field spectroscopy covering the rest-frame wavelengths of strong optical emission lines in nine sub-mm-luminous galaxies (SMGs) at 2.0

In this investigation we quantify the metallicities of low mass galaxies by constructing the most comprehensive census to date. We use galaxies from the SDSS and DEEP2 survey and estimate metallicities from their optical emission lines. We also use two smaller samples from the literature which have metallicities determined by the direct method using the temperature sensitive [OIII]4363 line. We examine the scatter in the local mass-metallicity (MZ) relation determined from ~20,000 star-forming galaxies in the SDSS and show that it is larger at lower stellar masses, consistent with the theoretical scatter in the MZ relation determined from hydrodynamical simulations. We determine a lower limit for the scatter in metallicities of galaxies down to stellar masses of ~10^7 M_solar that is only slightly smaller than the expected scatter inferred from the SDSS MZ relation and significantly larger than what is previously established in the literature. The average metallicity of star-forming galaxies increases with stellar mass. By examining the scatter in the SDSS MZ relation, we show that this is mostly due to the lowest metallicity galaxies. The population of low mass, metal-rich galaxies have properties which are consistent with previously identified galaxies that may be transitional objects between gas-rich dwarf irregulars and gas-poor dwarf spheroidals and ellipticals., Comment: Accepted to ApJ. 17 pages, 17 figures

We examine the cosmic growth of the red sequence in a cosmological hydrodynamic simulation that includes a heuristic prescription for quenching star formation that yields a realistic passive galaxy population today. In this prescription, halos dominated by hot gas are continually heated to prevent their coronae from fueling new star formation. Hot coronae primarily form in halos above \sim10^12 M\odot, so that galaxies with stellar masses \sim10^10.5 M\odot are the first to be quenched and move onto the red sequence at z > 2. The red sequence is concurrently populated at low masses by satellite galaxies in large halos that are starved of new fuel, resulting in a dip in passive galaxy number densities around \sim10^10 M\odot. Stellar mass growth continues for galaxies even after joining the red sequence, primarily through minor mergers with a typical mass ratio \sim1:5. For the most massive systems, the size growth implied by the distribution of merger mass ratios is typically \sim2\times the corresponding mass growth, consistent with observations. This model reproduces mass-density and colour-density trends in the local universe, with essentially no evolution to z = 1, with the hint that such relations may be washed out by z \sim 2. Simulated galaxies are increasingly likely to be red at high masses or high local overdensities. In our model, the presence of surrounding hot gas drives the trends with both mass and environment., Comment: 15 pages, 8 figures. MNRAS accepted

We carry out a new suite of cosmological radiation hydrodynamic simulations and explore the relative impacts on reionization-epoch star formation of galactic outflows and photoionization heating. By itself, an extragalactic ultraviolet background (EUVB) suppresses the luminosity function by less than 50% at z=6, overproducing the observed galaxy abundance by a factor of 3-5. Galactic outflows restore agreement with observations without preventing Population II star formation from reionizing the Universe by z=6. The resulting EUVB suppresses star formation in halos with virial temperatures below 10^5K but has a weaker impact in more massive halos. Nonetheless, the low-mass halos contribute up to 50% of all ionizing photons owing to the EUVB's inhomogeneity. Overall, star formation rate scales as halo mass M_h to the 1.3-1.4 in halos with $M_h=10^{8.2--10.2}\msun$. This is a steeper dependence than is often assumed in reionization models, boosting the expected power spectrum of 21 centimeter fluctuations on large scales. The luminosity function rises steeply to at least M_1600=-13, indicating that reionization was driven by faint galaxies (M_1600 >= -15) that have not yet been observed. Our models cannot simultaneously explain observations of galaxies, the cosmic microwave background, and the intergalactic medium. Increased dynamic range will alleviate the existing discrepancies, but observations may still require additional physics such as a variable ionizing escape fraction (abridged)., Comment: 23 pages, 15 Figures, accepted to ApJ

Massive galaxies today typically are not forming stars despite being surrounded by hot gaseous halos with short central cooling times. This likely owes to some form of "quenching feedback" such as merger-driven quasar activity or radio jets emerging from central black holes. Here we implement heuristic prescriptions for these phenomena on-the-fly within cosmological hydrodynamic simulations. We constrain them by comparing to observed luminosity functions and color-magnitude diagrams from SDSS. We find that quenching from mergers alone does not produce a realistic red sequence, because 1 - 2 Gyr after a merger the remnant accretes new fuel and star formation reignites. In contrast, quenching by continuously adding thermal energy to hot gaseous halos quantitatively matches the red galaxy luminosity function and produces a reasonable red sequence. Small discrepancies remain - a shallow red sequence slope suggests that our models underestimate metal production or retention in massive red galaxies, while a deficit of massive blue galaxies may reflect the fact that observed heating is intermittent rather than continuous. Overall, injection of energy into hot halo gas appears to be a necessary and sufficient condition to broadly produce red and dead massive galaxies as observed., Comment: 23 pages, 14 figures. MNRAS accepted. Added Sec. 4.4 and significantly modified the Discussion at the suggestion of the referee

We present an analysis of the Ly-a forest toward 3C 273 from the Space Telescope Imaging Spectrograph at ~7 km/s resolution, along with re-processed data from the Far Ultraviolet Spectroscopic Explorer. The high UV flux of 3C 273 allows us to probe the weak, low z absorbers. The main sample consists of 21 HI absorbers that we could discriminate to a sensitivity of log NHI~ 12.5. The redshift density for absorbers with 13.1= 12.5, it is consistent with numerical model predictions. The Doppler parameter distribution is consistent with other low z samples. We find no evidence for a break in the column density power-law distribution to log NHI=12.3. A broad Ly-a absorber (BLA) is within Delta v =< 50 km/s and 1.3 local frame Mpc of two ~0.5L* galaxies, with an OVI absorber ~700 km/s away, similarly close to three galaxies and indicating overdense environments. We detect clustering on the Delta v<1000 km/s scale at 3.4 sigma significance for log NHI >= 12.6, consistent with the level predicted from hydrodynamical simulations, and indication for a Ly-a forest void at 0.09

The bimodality in observed present-day galaxy colours has long been a challenge for hierarchical galaxy formation models, as it requires some physical process to quench (and keep quenched) star formation in massive galaxies. Here we examine phenomenological models of quenching by post-processing the star formation histories of galaxies from cosmological hydrodynamic simulations that reproduce observations of star-forming galaxies reasonably well. We consider recipes for quenching based on major mergers, halo mass thresholds, gas temperature thresholds, and variants thereof. We compare the resulting simulated star formation histories to observed g-r colour-magnitude diagrams and red and blue luminosity functions from SDSS. The merger and halo mass quenching scenarios each yield a distinct red sequence and blue cloud of galaxies that are in broad agreement with data, albeit only under rather extreme assumptions. In detail, however, the simulated red sequence slope and amplitude in each scenario is somewhat discrepant, perhaps traceable to low metallicities in simulated galaxies. Merger quenching produces more massive blue galaxies, earlier quenching, and more frosting of young stars; comparing to relevant data tends to favor merger over halo mass quenching. Although physically-motivated quenching models can produce a red sequence, interesting generic discrepancies remain that indicate that additional physics is required to reproduce the star formation and enrichment histories of red and dead galaxies., Comment: 26 pages, 17 figures (7 color). MNRAS accepted. Minor revisions in sec. 1.3, 5.5, 6.2, 6.4, and some figures. Full resolution figures may be found at http://orion.as.arizona.edu/~jgabor/Quenching1/FIGS/

We study the topology of reionization using accurate three-dimensional radiative transfer calculations post-processed on outputs from cosmological hydrodynamic simulations. In our simulations, reionization begins in overdense regions and then "leaks" directly into voids, with filaments reionizing last owing to their combination of high recombination rate and low emissivity. This result depends on the uniquely-biased emissivity field predicted by our prescriptions for star formation and feedback, which have previously been shown to account for a wide array of measurements of the post-reionization Universe. It is qualitatively robust to our choice of simulation volume, ionizing escape fraction, and spatial resolution (in fact it grows stronger at higher spatial resolution) even though the exact overlap redshift is sensitive to each of these. However, it weakens slightly as the escape fraction is increased owing to the reduced density contrast at higher redshift. We also explore whether our results are sensitive to commonly-employed approximations such as using optically-thin Eddington tensors or substantially altering the speed of light. Such approximations do not qualitatively change the topology of reionization. However, they can systematically shift the overlap redshift by up to $\Delta z\sim 0.5$, indicating that accurate radiative transfer is essential for computing reionization. Our model cannot simultaneously reproduce the observed optical depth to Thomson scattering and ionization rate per hydrogen atom at $z=6$, which could owe to numerical effects and/or missing early sources of ionization., Comment: 16 pages, 9 figures, accepted to MNRAS