arXiv subjects
Publications and source records attributed to Martin J. Rees.
Fred Hoyle famously drew attention to the significance of apparent coincidences in the energy levels of the carbon and oxygen nucleus. This paper addresses the possible implications of other coincidences in cosmology.
Collapsed objects have definitely been observed: some are stellar-mass objects, the endpoint of massive stars; others, millions of times more massive, have been discovered in the cores of most galaxies. Their formation poses some still-unanswered questions. But for relativists the key question is whether observations can probe the metric in the strong-field domain, and test whether it indeed agrees with the Kerr geometry predicted by general relativity.
Motivated by the recent detection by WMAP of a large optical depth to Thomson scattering -- implying a very early reionization epoch -- we assess a scenario where the universe was reionized by `miniquasars' powered by intermediate-mass black holes (IMBHs), the remnants of the first generation of massive stars. Pregalactic IMBHs form within minihalos above the cosmological Jeans mass collapsing at z=24, get incorporated through mergers into larger and larger systems, sink to the center owing to dynamical friction, and accrete cold material. The merger history of dark halos and associated IMBHs is followed by Monte Carlo realizations of the merger hierarchy in a LCDM cosmology. While seed IMBHs that are as rare as the 3.5-sigma peaks of the primordial density field evolve largely in isolation, a significant number of black hole binary systems will form if IMBHs populate the more numerous 3-sigma peaks instead. In the case of rapid binary coalescence a fraction of IMBHs will be displaced from galaxy centers and ejected into the IGM by the `gravitational rocket' effect, rather than accrete and shine as miniquasars. We show that, under a number of plausible assumptions for the amount of gas accreted onto IMBHs and their emission spectrum, miniquasars powered by IMBHs may be responsible for cosmological reionization at z~15. Reionization by miniquasars with a hard spectrum may be more `economical' than stellar reionization, as soft X-rays escape more easily from the dense sites of star formation and travel further than EUV radiation. Energetic photons will make the diffuse IGM warm and weakly ionized prior to the epoch of reionization breakthrough, set an entropy floor, and reduce gas clumping. Future 21 cm observations may detect a preheated, weakly-ionized IGM in emission against the CMB. (abridged)
The recent claim by Coburn & Boggs to have detected a very high degree of linear polarization in the prompt emission of GRB 021206 has stimulated interest in how much polarization could arise in gamma-ray bursts from synchrotron emission. Alternatively, as Shaviv & Dar have shown, GRB polarization could be produced by inverse Compton scattering in the point-source limit. We discuss polarization from a fireball that upscatters a soft radiation field. We show that, after the proper angular integration, the residual polarization can be large, in some cases approaching the point-source limit. We discuss the probability of realizing the geometrical conditions in which a large polarization is obtained showing that, for a particularly bright burst as GRB 021206, the detection of polarization at the first attempt in the Compton drag scenario is not unlikely.
The cosmic ultraviolet (UV) ionizing background impacts the formation of dwarf galaxies in the low-redshift universe (z=3) by suppressing gas infall into galactic halos with circular velocities up to v(circ)=75 km/s. Using a one-dimensional, spherically symmetric hydrodynamics code (Thoul & Weinberg 1995), we examine the effect of an ionizing background on low-mass galaxies forming at high redshifts (z>10). We find that the importance of photoionization feedback is greatly reduced, because (1) at high redshift, dwarf-galaxy sized objects can self-shield against the ionizing background, (2) collisional cooling processes at high redshift are more efficient, (3) the amplitude of the ionizing background at high redshift is lower, and (4) the ionizing radiation turns on when the perturbation that will become the dwarf galaxy has already grown to a substantial overdensity. We find that because of these reasons, gas can collect inside halos with circular velocities as low as v(circ)=10 km/s. This result has important implications for the reionization history of the universe.
In models for gamma ray burst afterglows, it is normally assumed that the external shock strongly amplifies the magnetic field and that this field maintains a steady value throughout the shocked region. We discuss the effects of modifying this (probably simplistic) assumption by allowing for a short decay time. The observations are incompatible with a post-shock field that decays too rapidly. However if the field pervades only a few percent of the total thickness of the shocked shell (and the electrons undergo only inverse Compton losses in the remainder) the model could be compatible with data. This would suggest a strong dependence of model parameters on the uncertainties of shock physics, therefore calling for independent external density estimates. We claim that afterglow emission should be instead seen as a laboratory where we can understand relativistic shock physics. The model we propose here together with afterglow observations in all wavebands could help to pin down the field structure.
According to the collapsar model, gamma-ray bursts are thought to be produced in shocks that occur after the relativistic jet has broken free from the stellar envelope. If the mass density of the collimated outflow is less than that of the stellar envelope, the jet will then be surrounded by a cocoon of relativistic plasma. This material would itself be able to escape along the direction of least resistance, which is likely to be the rotation axis of the stellar progenitor, and accelerate in approximately the same way as an impulsive fireball. We discuss how the properties of the stellar envelope have a decisive effect on the appearance of a cocoon propagating through it. The relativistic material that accumulated in the cocoon would have enough kinetic energy to substantially alter the structure of the relativistic outflow, if not in fact provide much of the observed explosive power. Shock waves within this plasma can produce gamma-ray and X-ray transients, in addition to the standard afterglow emission that would arise from the deceleration shock of the cocoon fireball.
We estimate the rate at which stars are captured by supermassive black holes (BHs) in the centres of bulges and elliptical galaxies assuming that these initially had an isothermal cusp (rho ~ r^{-2} with velocity dispersion sigma_*). If radial orbits can be adequately replenished a BH can grow significantly by capturing stars and particles of the dark halo; the latter could be up to 20--40 percent of the total mass captured. A tight relation M_bh 10^8 (sigma_*/200\kms)^5 between BH hole mass and stellar velocity dispersion arises at the high mass end M_bh >= 10^8 M_sun if these giant BHs grow primarily by swallowing stars. For smaller masses a gradual flattening of the relation and an increase of the scatter are expected due to tidal disruption of stars. At moderate redshifts when bright galaxies may still possess isothermal cusps tidal disruption flares may be as frequent as 10^{-4} -- 10^{-2} yr^{-1} per galaxy, a few percent of the supernovae rate. Efficient randomization of the orbits can occur when the dense cores of infalling satellites sink to the sphere of influence of the BH by dynamical friction. Efficient replenishment of the loss cone of radial orbits can also explain why the supermassive binary BHs expected in hierarchically merging galaxies do generally coalesce, as suggested by the observed relation between BH mass and the inferred mass of stars ejected from an isothermal cusp.
We compute the luminosity of K_alpha emission lines produced by astrophysically abundant elements in the soft X-ray spectra of the early afterglow of gamma-ray bursts. We find that the detection of these lines can be a diagnostic for the geometrical set-up of the reprocessing material. In particular we can distinguish between a ``geometry dominated'' model, in which the line emission is coming from an extended region and its duration arises from light-travel-time effects and an ``engine dominated'' model, where the line emitting gas is in a smaller region, irradiated for a longer period. These lines therefore offer clues to the dynamics and time-scale of the explosion leading to a gamma-ray burst.
Absorption or emission against the cosmic microwave background radiation (CMB) may be observed in the redshifted 21cm line if the spin temperature of the neutral intergalactic medium prior to reionization differs from the CMB temperature. This so-called 21cm tomography should reveal important information on the physical state of the intergalactic medium at high redshifts. The fluctuations in the redshifted 21 cm, due to gas density inhomogeneities at early times, should be observed at meter wavelengths by the next generation radio telescopes such as the proposed {\it Square Kilometer Array (SKA)}. Here we show that the extra-galactic radio sources provide a serious contamination to the brightness temperature fluctuations expected in the redshifted 21 cm emission from the IGM at high redshifts. Unless the radio source population cuts off at flux levels above the planned sensitivity of SKA, its clustering noise component will dominate the angular fluctuations in the 21 cm signal. The integrated foreground signal is smooth in frequency space and it should nonetheless be possible to identify the sharp spectral feature arising from the non-uniformities in the neutral hydrogen density during the epoch when the first UV sources reionize the intergalactic medium.
Estimating the age of the Universe is an old problem. Rapid progress in observational cosmology in recent years has led to more accurate values of the fundamental parameters. The current most popular model is a flat Universe, with about 30% of the critical density in the form of matter (baryonic and non-baryonic) and 70% in the form of `dark energy'. These two densities, together with the Hubble constant (estimated to be about 70 km/s/Mpc), constrain the age of the Universe to be approximately 13 Gyr. This expansion age is uncomfortably close to the age of the oldest globular clusters (approximately 12.5 Gyr), in particular if they formed relatively recently. We review here proposed models of globular cluster formation and point out possible conflicts with cosmology.
The hardness of the ionizing continuum from the first sources of UV radiation plays a crucial role in the reionization of the intergalactic medium (IGM). While usual stellar populations have soft spectra, mini-quasars or metal-free stars with high effective temperatures may emit hard photons, capable of doubly ionizing helium and increasing the IGM temperature. Absorption within the source and in the intervening IGM will render the ionizing continuum of high-redshift sources inaccessible to direct observation. Here we show that HeII recombination lines from the first luminous objects are potentially detectable by the Next Generation Space Telescope. Together with measurements of the hydrogen Balmer alpha emission line, this detection can be used to infer the ratio of HeII to HI ionizing photons. A measurement of this ratio would shed light on the nature and emission mechanism of the first luminous sources, with important astrophysical consequences for the reheating and reionization of the IGM.
This article (which appeared with illustrations) was the astrophysics/cosmology contribution to the historical series ``Pathways of Discovery'' which appeared in ``Science'' during last year.
Recent numerical simulations of the fragmentation of primordial molecular clouds in hierarchical cosmogonies have suggested that the very first stars (the so-called Population III) may have been rather massive. Here we point out that a numerous population of massive black holes (MBHs) -- with masses intermediate between those of stellar and supermassive holes -- may be the endproduct of such an episode of pregalactic star formation. If only one MBH with m > 150 msun formed in each of the `minihalos' collapsing at z=20 from 3-sigma fluctuations, then the mass density of Pop III MBHs would be comparable to that of the supermassive variety observed in the nuclei of galaxies. Since they form in high-sigma rare density peaks, relic MBHs are predicted to cluster in the bulges of present-day galaxies as they become incorporated through a series of mergers into larger and larger systems. Dynamical friction would cause more than 50 (m/150 msun)^{1/2} such objects to sink towards the center. The presence of a small cluster of MBHs in galaxy nuclei may have several interesting consequences associated with tidal captures of ordinary stars (likely followed by disruption), MBH capture by the central supermassive black hole, gravitational wave radiation from such coalescences. Accreting pregalactic MBHs may be detectable as ultra-luminous, off-nuclear X-ray sources.
The early stage in the formation of a galaxy inevitably involves a spatially extended distribution of infalling, cold gas. If a central luminous quasar turned on during this phase, it would result in significant extended Lyman alpha emission, possibly accompanied by other lines. For halos condensing at redshifts between 3 < z < 8 and having virial temperatures between 2 x 10^5 K and 2 x 10^6 K, this emission results in a ``fuzz'' of characteristic angular diameter of a few arcseconds, and surface brightness between 10^-18 and 10^-16 erg/s/cm^2/asec^2. The fuzz around bright, high redshift quasars could be detected in deep narrow band imaging with current telescopes, providing a direct constraint on galaxy formation models. The absence of detectable fuzz might suggest that most that most of the protogalaxy's gas settles to a self-gravitating disk before a quasar turns on. However, continued gas infall from large radii, or an on-going merger spreading cold gas over a large solid angle, during the luminous quasar phase could also result in extended Ly$α$ emission, and can be constrained by deep narrow band imaging.
We assess supernova (SN)-driven pregalactic outflows as a mechanism for distributing the product of stellar nucleosynthesis over large cosmological volumes prior to the reionization epoch. SN ejecta will escape the grasp of halos with virial temperatures T_vir>10^{4.3} K (corresponding to masses M>10^8 h^{-1} M_sun at redshift z=9 when they collapse from 2-sigma fluctuations) if rapid cooling can take place, and a significant fraction of their baryonic mass is converted into stars over a dynamical timescale. We study the evolution of SN-driven bubbles as they blow out from subgalactic halos and propagate into the intergalactic medium (IGM), and show that to lift the halo gas out of the potential well the energy injection must continue at least until blow-away occurs. The collective explosive output of about ten thousands SNe per halo at these early epochs could pollute vast regions of intergalactic space to a mean metallicity =Omega_Z/Omega_b>0.003 (comparable to the levels observed in the Lyman-alpha forest at z=3) without much perturbing the IGM hydrodynamically, i.e. producing large variations of the baryons relative to the dark matter. The volume filling factor of the ejecta is higher than 20% if the star formation efficiency is of order 10%. When the filling factor of the ejecta becomes significant, enriched material will typically be at a higher adiabat than expected from photoionization.
Within the framework of the internal-external shocks model for gamma-ray bursts, we study the various mechanisms that can give rise to quiescent times in the observed gamma-ray light-curves. In particular, we look for the signatures that can provide us with evidence as to whether or not the central engine goes dormant for a period of time comparable to the duration of the gaps. We show that the properties of the prompt gamma-ray and X-ray emission can in principle determine whether the quiescent episodes are due to a modulated relativistic wind or a switching off of the central engine. We suggest that detailed observations of the prompt afterglow emission from the reverse shock will strongly constrain the possible mechanisms for the production of quiescent times in gamma-ray bursts.
The need for high angular resolution is emphasised, especially in the context of programs to understand massive black holes and the processes in their environment.