An implementation of radiative transfer in the cosmological simulation code gadget

We present a novel numerical implementation of radiative transfer in the cosmological smoothed particle hydrodynamics (SPH) simulation code gadget . It is based on a fast, robust and photon-conserving integration scheme where the radiation transport problem is approximated in terms of moments of the transfer equation and by using a variable Eddington tensor as a closure relation, following the Optically Thin Variable Eddington Tensor suggestion of Gnedin & Abel. We derive a suitable anisotropic diffusion operator for use in the SPH discretization of the local photon transport, and we combine this with an implicit solver that guarantees robustness and photon conservation. This entails a matrix inversion problem of a huge, sparsely populated matrix that is distributed in memory in our parallel code. We solve this task iteratively with a conjugate gradient scheme. Finally, to model photon sink processes we consider ionization and recombination processes of hydrogen, which is represented with a chemical network that is evolved with an implicit time integration scheme. We present several tests of our implementation, including single and multiple sources in static uniform density fields with and without temperature evolution, shadowing by a dense clump and multiple sources in a static cosmological density field. All tests agree quite well with analytical computations or with predictions from other radiative transfer codes, except for shadowing. However, unlike most other radiative transfer codes presently in use for studying re-ionization, our new method can be used on-the-fly during dynamical cosmological simulation, allowing simultaneous treatments of galaxy formation and the re-ionization process of the Universe.     

Eddington ratio and accretion efficiency in active galactic nuclei evolution

The cosmological evolution of active galactic nuclei (AGN) is important for understanding the mechanism of accretion on to supermassive black holes and the related evolution of the host galaxy. In this work, we include objects with very low Eddington ratio  (10⁻³–10⁻²)  in an evolution scenario, and compare the results with the observed local distribution of black holes. We test several possibilities for the AGN population, considering obscuration and dependence with luminosity, and investigate the role of the Eddington ratio λ and radiative accretion efficiency ε on the shape of the evolved mass function. We find that three distinct populations of AGN can evolve with a wider parameter range than is usually considered, and still be consistent with the local mass function. In general, the black holes in our solutions are spinning rapidly. Taking fixed values for ε and λ neither provides a full knowledge of the evolution mechanism nor is consistent with the existence of low-Eddington-ratio objects.     

On the theory of magnetar QPOs

We examine the effect of inhomogeneous re-ionization on the galaxy power spectrum and the consequences for probing dark energy. To model feedback during re-ionization, we apply an ansatz setting the galaxy overdensity proportional to the underlying ionization field. Thus, inhomogeneous re-ionization may leave an imprint in the galaxy power spectrum. We evolve this imprint to low redshift and use the Fisher-matrix formalism to assess the effect on parameter estimation. We show that a combination of low-redshift  ( z = 0.3)  and high-redshift  ( z = 3)  galaxy surveys can constrain the size of cosmological H  ii regions during re-ionization. This imprint can also cause confusion when using baryon oscillations or other features of the galaxy power spectrum to probe the dark energy. We show that when bubbles are large, and hence detectable, our ability to constrain w can be degraded by up to 50 per cent. When bubbles are small, the imprint has little or no effect on measuring dark energy parameters.     

Gravitational wave background from neutrino-driven gamma-ray bursts

We study cosmic microwave background (CMB) secondary anisotropies produced by inhomogeneous reionization by means of cosmological simulations coupled with the radiative transfer code crash . The reionization history is consistent with the Wilkinson Microwave Anisotropy Probe Thomson optical depth determination. We find that the signal arising from this process dominates over the primary CMB component for   l ≳ 4000  and reaches a maximum amplitude of   l ( l + 1) C_l /2π≃ 1.6 × 10⁻¹³  on arcmin scales (i.e. l as large as several thousands). We then cross-correlate secondary CMB anisotropy maps with neutral hydrogen 21-cm line emission fluctuations obtained from the same simulations. The two signals are highly anticorrelated on angular scales corresponding to the typical size of H  ii regions (including overlapping) at the 21-cm map redshift. We show how the CMB/21-cm cross-correlation can be used: (i) to study the nature of the reionization sources; (ii) to reconstruct the cosmic reionization history; (iii) to infer the mean cosmic ionization level at any redshift. We discuss the feasibility of the proposed experiment with forthcoming facilities.     

Continuum shielding and flow dynamics in active galactic nuclei

We study the ionization, thermal structure and dynamics of active galactic nuclei (AGN) flows that are partially shielded from the central continuum. We utilize a detailed non-local thermodynamic equilibrium photoionization and radiative transfer code using exact (non-Sobolev) calculations. We find that shielding has a pronounced effect on the ionization, thermal structure and the dynamics of such flows. Moderate shielding is especially efficient in accelerating flows to high velocities because it suppresses the ionization level of the gas. The ionization structure of shielded gas tends to be distributed uniformly over a wide range of ionization levels. In such gas, radiation pressure due to trapped line photons can dominate over the thermal gas pressure and have a significant effect on the thermal stability of the flow. Heavily shielded flows are driven mainly by line radiation pressure, and so line locking has a large effect on the flow dynamics. We show that the observed 'L_α ghost' is a natural outcome in highly ionized flows that are shielded beyond the Lyman limit. We suggest that high-velocity AGN flows occupy only a small fraction of the volume and that their density depends only weakly on the velocity field.     

A possible EGRET source with a two-sided radio jet structure on a parsec scale – 3EG J0808+5114

Recent observations have revealed that damped Lyα clouds (DLAs) host star formation activity. In order to examine if such star formation activity can be triggered by ionization fronts, we perform high-resolution hydrodynamics and radiative transfer simulations of the effect of radiative feedback from propagating ionization fronts on high-density clumps. We examine two sources of ultraviolet (UV) radiation field to which high-redshift ( z ∼ 3) galaxies could be exposed: one corresponding to the UV radiation originating from stars within the DLA, itself, and the other corresponding to the UV background radiation. We find that, for larger clouds, the propagating I-fronts created by local stellar sources can trigger cooling instability and collapse of significant part, up to 85 per cent, of the cloud, creating conditions for star formation in a time-scale of a few Myr. The passage of the I-front also triggers collapse of smaller clumps (with radii below ∼4 pc), but in these cases the resulting cold and dense gas does not reach conditions conducive to star formation. Assuming that 85 per cent of the gas initially in the clump is converted into stars, we obtain a star formation rate of  ∼0.25 M_⊙ yr⁻¹ kpc⁻²  . This is somewhat higher than the value derived from recent observations. On the other hand, the background UV radiation which has harder spectrum fails to trigger cooling and collapse. Instead, the hard photons which have long mean free-path heat the dense clumps, which as a result expand and essentially dissolve in the ambient medium. Therefore, the star formation activity in DLAs is strongly regulated by the radiative feedback, both from the external UV background and internal stellar sources and we predict quiescent evolution of DLAs (not starburst-like evolution).     

The escape of ionizing photons from supernova-dominated primordial galaxies

In order to assess the contribution of Lyman break galaxies (LBGs) and Lyman α emitters (LAEs) at redshifts  3 < z < 7  to the ionization of intergalactic medium (IGM), we investigate the escape fractions of ionizing photons from supernova-dominated primordial galaxies by solving the three-dimensional (3D) radiative transfer. The model galaxy is employed from an ultra-high-resolution chemodynamic simulation of a primordial galaxy by Mori & Umemura, which well reproduces the observed properties of LAEs and LBGs. The total mass of model galaxy is  10¹¹ M_⊙  . We solve not only photoionization but also collisional ionization by shocks. In addition, according to the chemical enrichment, we incorporate the effect of dust extinction, taking the size distributions of dust into account. As a result, we find that dust extinction reduces the escape fractions by a factor of 1.5–8.5 in the LAE phase and by a factor of 2.5–11 in the LBG phase, while the collisional ionization by shocks increases the escape fractions by a factor of  ≈2  . The resultant escape fractions are 0.07–0.47 in the LAE phase and 0.06–0.17 in the LBG phase. These results are well concordant with the recent estimations derived from the flux density ratio at 1500 to 900 Å of LAEs and LBGs. Combining the resultant escape fractions with the luminosity functions of LAEs and LBGs, we find that high- z LAEs and LBGs can ionize the IGM at   z = 3–5  . However, ionizing radiation from LAEs as well as LBGs falls short of ionizing the IGM at   z > 6  . That implies that additional ionization sources may be required at   z > 6  .     

Detection of OH and wide H i absorption toward B0218+357

Large-scale polarization of the cosmic microwave background measured by the WMAP satellite requires a mean optical depth to Thomson scattering,  τ_e∼ 0.17  . The reionization of the Universe must therefore have begun at relatively high redshift. We have studied the reionization process using supercomputer simulations of a large and representative region of a universe which has cosmological parameters consistent with the WMAP results (  Ω_m= 0.3, Ω_Λ= 0.7, h = 0.7, Ω_b= 0.04, n = 1  and  σ₈= 0.9  ). Our simulations follow both the radiative transfer of ionizing photons and the formation and evolution of the galaxy population which produces them. A previously published model with ionizing photon production as expected for zero-metallicity stars distributed according to a standard stellar initial mass function (IMF) (10⁶¹ photons per unit solar mass of formed stars) and with a moderate photon escape fraction from galaxies (5 per cent), produces  τ_e= 0.104  , which is within 1.0 to  1.5σ  of the 'best' WMAP value. Values of up to 0.16 can be produced by taking larger escape fractions or a top-heavy IMF. The data do not require a separate populations of 'miniquasars' or of stars forming in objects with total masses below  10⁹ M_⊙  . Reconciling such early reionization with the observed Gunn–Peterson troughs in   z > 6  quasars may be challenging. Possible resolutions of this problem are discussed.     

Radiation from early black holes – I. Effects on the neutral intergalactic medium

In the pre-reionization Universe, the regions of the intergalactic medium (IGM) which are far from luminous sources are the last to undergo reionization. Until then, they should be scarcely affected by stellar radiation; instead, the X-ray emission from an early black hole (BH) population can have much larger influence. We investigate the effects of such emission, looking at a number of BH model populations (differing for the cosmological density evolution of BHs, the BH properties, and the spectral energy distribution of the BH emission). We find that BH radiation can easily heat the IGM to  10³–10⁴ K  , while achieving partial ionization. The most interesting consequence of this heating is that BHs are expected to induce a 21-cm signal (  δ T _b∼ 20–30 mK  at   z ≲ 12  ) which should be observable with forthcoming experiments (e.g. LOFAR). We also find that at   z ≲ 10  BH emission strongly increases the critical mass separating star-forming and non-star-forming haloes.     

Introducing a hybrid radiative transfer method for smoothed particle hydrodynamics

A new means of incorporating radiative transfer into smoothed particle hydrodynamics (SPH) is introduced, which builds on the success of two previous methods – the polytropic cooling approximation as devised by Stamatellos et al. and flux-limited diffusion. This hybrid method preserves the strengths of its individual components, while removing the need for atmosphere matching or other boundary conditions to marry optically thick and optically thin regions. The code uses a non-trivial equation of state to calculate temperatures and opacities of SPH particles, which captures the effects of H₂ dissociation, H⁰ ionization, He⁰ and He⁺ ionization, ice evaporation, dust sublimation, molecular absorption, bound-free and free–free transitions and electron scattering. The method is tested in several scenarios, including (i) the evolution of a  0.07 M_⊙  protoplanetary disc surrounding a  0.5 M_⊙  star; (ii) the collapse of a  1 M_⊙  protostellar cloud and (iii) the thermal relaxation of temperature fluctuations in a static homogeneous sphere.     

NLTE models of line-driven stellar winds – III. Influence of X-ray radiation on wind structure of O stars

We study the influence of X-rays on the wind structure of selected O stars. For this purpose we use our non-local thermodynamic equilibrium (NLTE) wind code with inclusion of additional artificial source of X-rays, assumed to originate in the wind shocks.
We show that the influence of shock X-ray emission on wind mass-loss rate is relatively small. Wind terminal velocity may be slightly influenced by the presence of strong X-ray sources, especially for stars cooler than   T _eff≲ 35 000 K  .
We discuss the origin of the   L _X/ L ∼ 10⁻⁷  relation. For stars with thick wind this relation can be explained assuming that the cooling time depends on wind density. Stars with optically thin winds exhibiting the 'weak wind problem' display enhanced X-ray emission which may be connected with large shock cooling length. We propose that this effect can explain the 'weak wind problem'.
Inclusion of X-rays leads to a better agreement of the model ionization structure with observations. However, we do not find any significant influence of X-rays on P  v ionization fraction implying that the presence of X-rays cannot explain the P  v problem.
We study the implications of modified ionization equilibrium due to shock emission on the line transfer in the X-ray region. We conclude that the X-ray line profiles of helium-like ions may be affected by the line absorption within the cool wind.     

Hot accretion discs with thermal Comptonization and advection in luminous black hole sources

We solve for the structure of a hot accretion disc with unsaturated thermal Comptonization of soft photons and with advection, generalizing the classical model of Shapiro et al. The upper limit on the accretion rate due to advection constrains the luminosity to ≲ 0.15 y^3/5 α^7/5 of the Eddington limit, where y and α are the Compton and viscosity parameters, respectively. The characteristic electron temperature and Thomson optical depth of the inner flow at accretion rates within an order of magnitude of that upper limit are ∼ 10⁹ K and ∼ 1, respectively. The resulting spectra are then in close agreement with the X-ray and soft γ-ray spectra from black hole binaries in the hard state and Seyferts. At low accretion rates, bremsstrahlung becomes the dominant radiative process.     

Newtonian hydrodynamics of the coalescence of black holes with neutron stars – IV. Irrotational binaries with a soft equation of state

We present the results of three-dimensional hydrodynamical simulations of the final stages of in-spiral in a black hole–neutron star binary, when the separation is comparable to the stellar radius. We use a Newtonian smooth particle hydrodynamics (SPH) code to model the evolution of the system, and take the neutron star to be a polytrope with a soft (adiabatic indices     and     equation of state and the black hole to be a Newtonian point mass. The only non-Newtonian effect we include is a gravitational radiation back reaction force, computed in the quadrupole approximation for point masses. We use irrotational binaries as initial conditions for our dynamical simulations, which are begun when the system is on the verge of initiating mass transfer and followed for approximately 23 ms. For all the cases studied we find that the star is disrupted on a dynamical time-scale, and forms a massive     accretion torus around the spinning (Kerr) black hole. The rotation axis is clear of baryons (less than 10⁻⁵ M_⊙ within 10°) to an extent that would not preclude the formation of a relativistic fireball capable of powering a cosmological gamma-ray burst. Some mass (the specific amount is sensitive to the stiffness of the equation of state) may be dynamically ejected from the system during the coalescence and could undergo r-process nucleosynthesis. We calculate the waveforms, luminosities and energy spectra of the gravitational radiation signal, and show how they reflect the global outcome of the coalescence process.     

Cosmological reionization around the first stars: Monte Carlo radiative transfer

We study the evolution of ionization fronts around the first protogalaxies by using high-resolution numerical cosmological (Λ+ cold dark matter, CDM, model) simulations and Monte Carlo radiative transfer methods. We present the numerical scheme in detail and show the results of test runs from which we conclude that the scheme is both fast and accurate. As an example of interesting cosmological application, we study the reionization produced by a stellar source of total mass M =2×10⁸ M_⊙ turning on at z ≈12, located at a node of the cosmic web. The study includes a spectral energy distribution of a zero-metallicity stellar population, and two initial mass functions (IMFs; Salpeter/Larson). The expansion of the ionization front (I-front) is followed as it breaks out from the galaxy and is channelled by the filaments into the voids, assuming (in a 2D representation) a characteristic butterfly shape. The ionization evolution is very well tracked by our scheme, as realized by the correct treatment of the channelling and shadowing effects resulting from overdensities. We confirm previous claims that both the shape of the IMF and the ionizing power metallicity dependence are important to correctly determine the reionization of the Universe.     

On the influence of resonant scattering on cosmic microwave background polarization anisotropies

We implement the theory of resonant scattering in the context of cosmic microwave background (CMB) polarization anisotropies. We compute the changes in the E-mode polarization (EE) and temperature E-mode (TE) CMB power spectra introduced by the scattering on a resonant transition with a given optical depth τ_X and polarization coefficient E ₁. The latter parameter, accounting for how anisotropic the scattering is, depends on the exchange of angular momentum in the transition, enabling observational discrimination between different resonances. We use this formalism in two different scenarios: cosmological recombination and cosmological re-ionization. In the context of cosmological recombination, we compute predictions in frequency and multipole space for the change in the TE and EE power spectra introduced by scattering on the Hα and Pα lines of hydrogen. This constitutes a fundamental test of the standard model of recombination, and the sensitivity it requires is comparable to that needed in measuring the primordial CMB B-mode polarization component. In the context of re-ionization, we study the scattering off metals and ions produced by the first stars, and find that polarization anisotropies, apart from providing a consistency test for intensity measurements, give some insight on how re-ionization evolved. Since polarization anisotropies have memory of how anisotropic the line scattering is, they should be able to discern the O  i 63.2-μm transition from other possible transitions associated to O  iii , N  ii , N  iii , etc. The amplitude of these signals are, however, between 10 and 100 times below the (already challenging) level of CMB B-mode polarization anisotropies.     

Soft X-ray spectroscopy of Compton-thick Seyfert 2 galaxies with BeppoSAX

We present an X-ray spectroscopic study of the bright Compton-thick Seyfert 2 galaxies NGC 1068 and the Circinus Galaxy, performed with BeppoSAX . Matt et al. interpreted the spectrum above 4 keV as the superposition of Compton reflection and warm plasma scattering of the nuclear radiation. When this continuum is extrapolated downwards to 0.1 keV, further components arise. The NGC 1068 spectrum is rich in emission lines, mainly owing to K _α transitions of He-like elements from oxygen to iron, plus a K _α fluorescent line from neutral iron. If the ionized lines originate in the warm scatterer, its thermal and ionization structure must be complex. From the continuum and line properties, we estimate a column density, N _warm, of the warm scatterer less than a few×10²¹ cm⁻². In the Circinus Galaxy, the absence of highly ionized iron is consistent with a scattering medium with U _X≲5 and N _warm∼ a few×10²² cm⁻². In both cases the neutral iron line is most naturally explained as fluorescence in the medium responsible for the Compton reflection continuum. In NGC 1068 an optically thin plasma emission with kT ≃500 eV and strongly sub-solar metallicity is required, while such a component is only marginal in the Circinus Galaxy. We tentatively identify this component as emission of diffuse hot gas in the nuclear starbursts. Possible causes for the metal depletion are discussed.     

Quasi-stars: accreting black holes inside massive envelopes

We study the structure and evolution of 'quasi-stars', accreting black holes embedded within massive hydrostatic gaseous envelopes. These configurations may model the early growth of supermassive black hole seeds. The accretion rate on to the black hole adjusts so that the luminosity carried by the convective envelope equals the Eddington limit for the total mass,   M _*+ M _BH≈ M _*  . This greatly exceeds the Eddington limit for the black hole mass alone, leading to rapid growth of the black hole. We use analytic models and numerical stellar structure calculations to study the structure and evolution of quasi-stars. We show that the photospheric temperature of the envelope scales as   T _ph∝ M ^−2/5_BH M ^7/20_*  , and decreases with time while the black hole mass increases. Once   T _ph < 10⁴ K  , the photospheric opacity drops precipitously and T _ph hits a limiting value, analogous to the Hayashi track for red giants and protostars, below which no hydrostatic solution for the convective envelope exists. For metal-free (Population III) opacities, this limiting temperature is approximately 4000 K. After a quasi-star reaches this limiting temperature, it is rapidly dispersed by radiation pressure. We find that black hole seeds with masses between 10³ and  10⁴ M_⊙  could form via this mechanism in less than a few Myr.     

Sakurai's Object: characterizing the near-infrared CO ejecta between 2003 and 2007

We present observations of Sakurai's Object obtained at 1–5 μm between 2003 and 2007. By fitting a radiative transfer model to an echelle spectrum of CO fundamental absorption features around  4.7 μm  , we determine the excitation conditions in the line-forming region. We find  ¹²C/¹³C = 3.5^+2.0_−1.5  , consistent with CO originating in ejecta processed by the very late thermal pulse, rather than in the pre-existing planetary nebula. We demonstrate the existence of  2.2 × 10⁻⁶≤ M _CO≤ 2.7 × 10⁻⁶ M_⊙  of CO ejecta outside the dust, forming a high-velocity wind of  500 ± 80 km s⁻¹  . We find evidence for significant weakening of the CO band and cooling of the dust around the central star between 2003 and 2005. The gas and dust temperatures are implausibly high for stellar radiation to be the sole contributor.     

Inefficient star formation: the combined effects of magnetic fields and radiative feedback

We investigate the effects of magnetic fields and radiative protostellar feedback on the star formation process using self-gravitating radiation magnetohydrodynamical calculations. We present results from a series of calculations of the collapse of  50 M_⊙  molecular clouds with various magnetic field strengths and with and without radiative transfer. We find that both magnetic fields and radiation have a dramatic impact on star formation, though the two effects are in many ways complementary. Magnetic fields primarily provide support on large scales to low-density gas, whereas radiation is found to strongly suppress small-scale fragmentation by increasing the temperature in the high-density material near the protostars. With strong magnetic fields and radiative feedback, the net result is an inefficient star formation process with a star formation rate of  ≲10  per cent per free-fall time that approaches the observed rate, although we have only been able to follow the calculations for 1/3 of a free-fall time beyond the onset of star formation.