Diffuse galactic gamma-ray continuum emission

Some of the most important questions about the diffuse gamma-ray continuum emission from the Galaxy are reviewed, based on Compton Observatory (CGRO) results, especially COMPTEL and EGRET, and also earlier COS-B analyses. The key issues include the rôle of emission from cosmic-ray interactions with molecular hydrogen and its energy dependence, emissivity gradients and their interpretation, the cosmic-ray electron spectrum and the effect of discrete sources. The relative contribution of the various emission processes at low and high latitudes is estimated and a plausible synthesis of the observed spectrum over 5 decades of energy is presented. In the energy range above 30 MeV, models based either on explicit cosmic-ray gradients or cosmic-ray/gas coupling can give acceptable fits to the data, and a clear distinction has yet to be made. The quality of the EGRET data may make this possible in the future. The value of the CO-to-H₂ conversion factor from -rays is still uncertain and there is considerable evidence for cloud-to-cloud variations. The existence of a small emissivity gradient is well established, but is difficult to explain in a diffusive cosmic-ray propagation picture with sources distributed like SNR or pulsars unless there is a larger halo than suggested by cosmic-ray composition studies. In the energy range 1–30 MeV covered by COMPTEL the spectrum of the diffuse emission has been measured and is consistent with a combination of bremsstrahlung and inverse-Compton emission; spatial analysis shows strong evidence for a component with a wide latitude extent which is plausibly identified with the inverse-Compton component. The molecular hydrogen appears to be only a weak -ray emitter at low energies, which can be interpreted in terms of reduced MeV cosmic-ray electron density in molecular clouds. New data on the hard X-ray diffuse galactic emission is becoming available and indicates the need for a low-energy upturn in the electron spectrum or some other additional component. The contribution of unresolved sources to the diffuse emission is unknown but-probably lies in the range 10–20%. At high latitudes the galactic emission is intense enough to significantly complicate the identification of the extragalactic component; in particular the inverse-Compton emission from a halo a few kpc in extent can account for much of the high-latitude galactic emission. The detection of the Large Magellanic Cloud and the non-detection of the Small Magellanic Cloud provide constraints on extragalactic cosmic-rays, and provide an interesting comparison with the properties of the galactic system. On account of the large amount of data from CGRO now available, this is a subject in rapid development, and this paper provides a snapshot of the situation around mid-1995.     

Large-scale structure of the Galaxy and high-energy gamma-ray observations

Detailed information on the high-energy gamma-ray emission from our Galaxy has become available through the two dedicated satellite missions SAS-2 and COS-B. The consistency of the two datasets is discussed; while a satisfying general agreement is observed, a few distinct discrepancies point to possible time variations within the compact source component of the total galactic emission. The bulk of emission appears very well correlated to the column density of the total interstellar gas, as traced by radio observations of Hi and CO. The gamma-ray observations exclude the possibility that H₂ dominates in the inner Galaxy, its mass should not exceed the mass existing in the form of Hi. Neither a significant galactocentric gradient of the (high-energy) cosmic-ray flux density is suggested inside the solar circle (outside a decrease is needed), nor a linear coupling between the cosmic rays and the gas is indicated by the gamma-ray data. The systematic variation with longitude of the spectrum of the gamma-ray emission points to an increased flux of cosmic-ray electrons in the 100 MeV to 1 GeV energy range in regions where dense clouds are concentrated. The variation could as well be due to the largely unresolved population of compact gamma-ray objects.     

Galactic effects of the cosmic-ray gas

On an astronomical scale cosmic rays must be considered a tenuous and extremely hot (relativistic) gas. The pressure of the cosmic-ray gas is comparable to the other gas and field pressures in interstellar space, so that the cosmic-ray pressure must be taken into account in treating the dynamical properties of the gaseous disk of the galaxy. This review begins with a survey of present knowledge of the cosmic-ray gas. Then the kinetic properties of the gas are developed, followed by an exposition of the dynamical effects of the cosmic-ray gas on a large-scale magnetic field embedded in a thermal gas. The propagation of low-frequency hydromagnetic waves is worked out in the fluid approximation.The dynamical properties of the gaseous disk of the galaxy are next considered. The equations for the equilibrium distribution in the direction perpendicular to the disk are worked out. It is shown that a self-consistent equilibrium can be constructed within the range of the observational estimates of the gas density, scale height, turbulent velocity, field strength, cosmic-ray pressure, and galactic gravitational acceleration. Perturbation calculations then show that the equilibrium is unstable, on scales of a few hundred pc and in times of the order 2 × 10⁷ years. The instability is driven about equally by the magnetic field and the cosmic-ray gas and dominates self-gravitation. Hence the instability dominates the dynamics of the interstellar gas and is the major effect in forming interstellar gas clouds. Star formation is the end result of condensation of the interstellar gas into clouds, indicating, then, that cosmic rays play a major role in initiating star formation in the galaxy.The cosmic rays are trapped in the unstable gaseous disk and escape from the disk only in so far as their pressure is able to inflate the magnetic field of the disk. The observed scale height of the galactic disk, the short life (10⁶ years) of cosmic-ray particles in the disk of the galaxy, and their observed quiescent state in the disk, indicate that the galactic magnetic field acts as a safety valve on the cosmic ray pressure P so that PB ²/8. We infer from the observed life and quiescence of the cosmic rays that the mean field strength in the disk of the galaxy is 3–5 × 10^–6 gauss.     

Clouds and Diffuse Baryonic Dark Matter

The nature of the dark matter in the halo of our galaxy is still largely unknown. The microlensing events found so far towards the Large Magellanic Cloud suggest that at most about 20% of the halo dark matter is in form of MACHOs (Massive Astrophysical Compact Halo Objects). The dark matter could also, at least partially, be made of cold molecular clouds (mainly H₂). We proposed a model for baryonic dark matter, according to which dark clusters of brown dwarfs and cold self-gravitating H₂ clouds populate the outer galactic halo. A signature would be a diffuse -ray emission from the galactic halo. Basically, cosmic-ray protons in the galactic halo scatter on the clouds clumped into dark clusters, giving rise to a -ray flux. An analysis of EGRET data has led to the discovery of a statistically significant diffuse -ray emission from the galactic halo, which turns out to be in remarkably good agreement with our prediction.     

Some problems of gamma-astronomy

The modern state of gamma-ray astronomy is reviewed, the paper being mainly devoted to the theoretical models that describe generation of gamma-ray emission under astrophysical conditions. Basic information on the processes of generation and absorption of gamma-rays, as well as the results of observations for various gamma-ray photon energies are reported.In the region of soft gamma-ray emission (i.e., for energies less than tens of MeV), where emission in gamma-ray lines dominates, we also discuss the nature of gamma-ray bursts, the origin of gamma-ray emission from the galactic centre, etc.Discrete sources and, in particular, the mysterious source Cyg X-3 are discussed in the region of very high (E > 10¹² eV) and ultra-high (E > 10¹⁵ eV) energy gamma-ray emission.A larger portion of the review is devoted to the analysis of cosmic-ray origin on the basis of the available gamma-ray data in the region from several tens of MeV to several GeV. The peculiarity of this energy range is, in particular, in the fact that the diffuse galactic emission was observed mainly there. We also discuss the problem of determination of the cosmic-ray density gradient from the gamma-ray data.The origin of high-latitude gamma-ray emission, the problem of galactic gamma-ray halo, etc., are discussed.The theoretical models explaining the nature of unidentified gamma-ray sources, as well as the results of measurements and theoretical estimations of a gamma-ray flux from SN1987A are analysed.List of Notations m electron mass, m = 9.108 × 10^–28 g, - M proton mass, M = 1.672 × 10^–24 g, - e electron charge, e = 4.803 × 10^–10 CGS - c velocity of light, c = 2.9979 × 10¹⁰ cm s^–1, - k Boltzmann constant, k = 1.380 × 10^–16 erg grad^–1, - e electron - p proton - gamma-ray photon - p antiproton - ^{0 0}-meson - -lepton - e ⁺ positron - r, , x radio-frequency, gamma-ray, and X-ray emission bands - E total energy of a particle - E _k kinetic energy - p particle momentum - spectral index for particles - spectral index for emission - n particle density (concentration) - H magnetic field strength - T temperature - _ph energy of low-energy photons - emission frequency - r _H Larmor radius of relativistic particles - k wave number - , z cylindric coordinates, in this case the coordinate (radius) along the galactic disk, z perpendicular to the galactic disk - M solar mass, M = 1.99 × 10³³ g.     

Recent results from COMPTEL

The COMPTEL telescope aboard the Compton Gamma Ray Observatory has put MeV -rays into the midst of astronomy. Among recent highlights are the discovery of intense MeV emission from blazar-type active galactic nuclei, the surprising non-detection of any Seyfert galaxy at MeV energies, the first image of the Milky Way in the light of the²⁶Al line at 1.809 MeV (possibly including a detection of the Vela supernova remnant), the discovery of 3–7 MeV emission from the Orion complex, which can be identified with nuclear interaction lines of¹²C and¹⁶O at 4.44 and 6.13 MeV, the detection of the⁴⁴Ti line at 1.15 MeV from the supernova remnant Cas A, and the first results on the spectrum and propagation of low-energy (1–100 MeV) cosmic-ray electrons.also Leiden Observatory     

Cosmic-ray lifetime in the galaxy: Experimental results and models

The containment lifetime of the cosmic radiation is a crucial parameter in the investigation of the cosmic-ray origin and plays an important role in the dynamics of the Galaxy. The separation of the cosmic-ray Be isotopes achieved by two satellite experiments is considered in this paper, and from the measured isotopic ratio between the radioactive ¹⁰Be (half-life = 1.5 × 10⁶ yr) and the stable ⁹Be, it is deduced that the cosmic rays propagate through matter with an average density of 0.24 ± 0.07 atoms cm^-3, lower than the traditionally quoted average density in the galactic disk of 1 atom cm^-3. This paper reviews the implications of this result for the cosmic-ray age mainly in the context of two models of confinement and propagation: the homogeneous model, normally identified with confinement to the galactic gaseous disk, and a diffusion model in which the cosmic rays extend into a galactic halo. The propagation calculations use:

1. a newly deduced cosmic-ray pathlength distribution.
2. a self-consistent model of solar modulation.
3. an up-to-date set of fragmentation cross sections.

The satellite results and their implications are compared with the information on the cosmic-ray age derived from other cosmic-ray radioactive nuclei and the measured differential energy spectrum of high-energy electrons. It is a major conclusion of this paper that in a homogeneous model the cosmic-ray age is 15(+7, -4) million years, i.e., about a factor 4 longer than early estimates based on the abundances of the light nuclei Li, Be, and B and a nominal interstellar density of 1 atom cm ^-3. The lifetime is even longer when the satellite results are applied to a diffusion halo model. The deduced traversed matter density, together with other astrophysical considerations, suggest the population of a galactic halo by the cosmic rays.     

The relevance of molecular observations to gamma-ray astronomy

The interaction of cosmic rays with interstellar clouds may produce some of the observed gamma-ray sources. The use of molecular observations to estimate the cloud masses, which are used to derive cosmic-ray fluxes, is reviewed. Molecular diagnostics of high cosmic-ray ionization rates are discussed, and a detailed application of those diagnostics is summarised and presented as evidence that second-order Fermi acceleration is important in old supernova remnants and can produce cosmic rays of too low energy to induce gamma-ray emission.Proceedings of the XVIII General Assembly of the IAU: Galactic Astrophysics and Gamma-Ray Astronomy, held at Patras, Greece, 19 August 1982.Royal Society Jaffé Donation Fellow.     

New Experimental Data and What It Tells Us About the Sources and Acceleration of Cosmic Rays

This paper summarizes new data in several fields of astronomy that relate to the origin and acceleration of cosmic rays in our galaxy and similar nearby galaxies. Data from radio astronomy shows that supernova remnants, both in our galaxy and neighboring galaxies, appear to be the sources of most of the accelerated electrons observed in these galaxies. -ray measurements also reveal several strong sources associated with supernova remnants in our galaxy. These sources have -ray spectra that are suggestive of the acceleration of cosmic-ray nuclei. Cosmic-ray observations from the Voyager and Ulysses spacecraft suggest a source composition that is very similar to the solar composition but with distinctive differences in the ⁴He, ¹²C,¹⁴ N and ²²Ne abundances that are the imprint of giant W-R star nucleosynthesis. Injection effects which depend on the first ionization potential (FIP) of the elements involved are also observed, in a manner similar to the fractionization observed between the solar photosphere and corona and also analogous to the preferential acceleration observed for high FIP elements at the heliospheric solar wind termination shock. Most of the ⁵⁹Ni produced in the nucleosynthesis of Fe peak nuclei just prior to a SN explosion appears to have decayed to ⁵⁹Co before the cosmic rays have been accelerated, suggesting that the⁵⁹ Ni is accelerated at least 10⁵ yr after it is produced. The decay of certain K capture isotopes produced during cosmic-ray propagation has also been observed for the first time. These measurements suggest that re-acceleration after an initial principal acceleration cannot be large. The high energy spectral indices of cosmic-ray nuclei show a significant charge dependent trend with the index of hydrogen being -2.76 and that of Fe -2.61. The escape length dependence of cosmic rays from our galaxy can now be measured up to ~300 GeV nucl^-1 using the Fe sec/Fe ratio. This escape length is P ^-0.05 above 10 GeV nucl^-1 leading to a typical source spectral index of (2.70±0.10) -0.50 = -2.20 for nuclei. This is similar to the source index of -2.3 inferred for electrons within the errors of ±0.1 in the index for both components. Spacecraft measurements in the outer heliosphere suggest that the local cosmic-ray energy density is ~2eV cm^-3 – larger than previously assumed. Gamma-ray measurements of electron bremsstrahlung below 50 MeV from the Comptel experiment on CGRO show that fully 20–30% of this energy is in electrons, several times that previously assumed. New estimates of the amount of matter traversed by cosmic rays using measurements of the B/C ratio are also higher than earlier estimates – this value is now ~10 g cm^-2 at 1 GeV nucl^-1. Thus altogether cosmic rays are energetically a more important component of our galaxy than previously assumed. This has implications both for the types of sources that are capable of accelerating cosmic rays and also for the role that cosmic rays may play in ionizing the diffuse interstellar medium.     

Future instrumental capabilities in the energy range of nuclear transitions

Gamma-ray lines are the fingerprints of nuclear transitions, carrying the memory of high energy processes in the universe. Setting out from what is presently known about line emission in gamma-ray astronomy, requirements for future telescopes are outlined. The inventory of observed line features shows that sources with a wide range of angular and spectral extent have to be handled: the scientific objectives for gamma-ray spectroscopy are spanning from compact objects as broad class annihilators, over longlived galactic radioisotopes with hotspots in the degree-range to the extremely extended galactic disk and bulge emission of the narrow e^-e⁺ line.The instrumental categories which can be identified in the energy range of nuclear astrophysics have their origins in the different concepts of light itself: geometrical optics is the base of coded aperture systems — these methods will continue to yield adequate performances in the near future. Beyond this, focusing telescopes and Compton telescopes, based on wave- and quantum- optics respectively, may be capable to further push the limits of resolution and sensitivity.     

Spectral observation of the composite supernova remnant G 29.7-0.3

The X-ray properties of the supernova remnant G 29.7-0.3 are discussed based on spectral data from the EXOSAT satellite. In the 2 to 10 keV range a featureless power-law spectrum is obtained, the best-fit parameters being: energy spectral index =-0.77, hydrogen column density on the line of sight N_H=2.3.10²² cm^–2. The incident X-ray flux from the source is (3.6±0.1) 10¹¹ erg cm^–2 s^–1 in the 2 to 10 keV range corresponding to an intrinsic luminosity of about 2. 10³⁶ erg s^–1 for a distance of 19 kpc. The source was not seen with the imaging instrument thus constraining the hydrogen column density to be N_H=(3.3 ±0.3) 10²² cm^–2 and the energy spectral index =1.0±0.15. This new observation is consistent with emission by a synchroton nebula presumably fed by an active pulsar. An upper limit of 1.5% for the pulsed fraction in the range of periods 32ms to 10⁴ s has been obtained.     

Probing the era of galaxy formation via TeV gamma ray absorption by the near infrared extragalactic background

We present models of the extragalactic background light (EBL) based on several scenarios of galaxy formation and evolution. We have treated galaxy formation with the Press-Schecter approximation for both cold dark matter (CDM) and cold+hot dark matter (CHDM) models, representing a moderate (z _f 3) and a late (z _f 1) era of galaxy formation respectively. Galaxy evolution has been treated by considering a variety of stellar types, different initial mass functions and star formation histories, and with an accounting of dust absorption and emission. We find that the dominant factor influencing the EBL is the epoch of galaxy formation. A recently proposed method for observing the EBL utilizing the absorption of 0.1 to 10 TeV gamma-rays from active galactic nuclei (AGN) is shown to be capable of discriminating between different galaxy formation epochs. The one AGN viewed in TeV light, Mrk 421, does show some evidence for a cutoff above 3 TeV; based on the EBL models presented here, we suggest that this is due to extinction in the source. The large absorption predicted at energies > 200 GeV for sources at z > 0.5 indicates that observations of TeV gamma-ray bursts (GRB) would constrain or eliminate models in which the GRB sources lie at cosmological distances.Now at University of Chicago, Dept. of Astronomy & Astrophysics.     

Cosmic-ray acceleration and transport,and diffuse galactic γ-ray emission

Cosmic-ray acceleration and transport is considered from the point of view of application to diffuse galactic -ray sources. As an introduction we review several source models, in particular supernovae exploding inside or near large interstellar clouds. The complex problem of cosmic ray transport in random electromagnetic fields is reduced to three cases which should be sufficient for practical purposes. As far as diffusive acceleration is concerned, apart from reviewing the basic physical principles, we point out the relation between shock acceleration and 2nd order Fermi acceleration, and the relative importance of the two processes around interstellar shock waves. For -ray source models the interaction of cosmic rays with dense clouds assumes great importance. Past discussions had been confined to static interactions of clouds with the ambient medium in the sense that no large scale mass motions in the ambient interstellar medium were considered. The well-known result then is that down to some tens of MeV or less, cosmic-ray nucleons should freely penetrate molecular clouds of typical masses and sizes. The self-exclusion of very low energy nucleons however may affect electron transport with consequences for the Bremsstrahlung -luminosity of such clouds.In this paper we consider also the dynamical interaction of dense clouds with a surrounding hot interstellar medium. Through cloud evaporation and accretion there exist mass flows in the cloud surroundings. We argue that in the case of (small) cloud evaporation the galactic cosmic rays will be essentially excluded from the clouds. The dynamic effects of cosmic rays on the flow should be minor in this case. For the opposite case of gas accretion onto (large) clouds, cosmic-ray effects on the flow will in general be large, limiting the cosmic-ray compression inside the cloud to dynamic pressure equilibrium. This should have a number of interesting and new consequences for -ray astronomy. A first, qualitative discussion is given in the last section.Proceedings of the XVIII General Assembly of the IAU: Galactic Astrophysics and Gamma-Ray Astronomy, held at Patras, Greece, 19 August 1982.     

Simultaneous EXOSAT-IUE observations of NGC 4151

NGC 4151 was observed four times in Nov. 83. The results indicate that: a) there exists a correlation between the X-ray and UV fluxes on the long term; b) the soft X-ray excess between 0.1 and 1 keV is probably steeper than expected from the leaky absorber model by Holt et al (1980); c)the spectral fit to the ME data, after correction for a soft component, yields =1,73±0.27, N_H=(15.2±2.2)×10²² cm^–2, E.W.(Fe line)=0.208±0.084 keV, and does not require a strong overabundance of Fe in the absorber. The relationship between N_H and the strength of the broad emission lines is commented.     

O-star winds in the Magellanic Clouds and the Milky Way

UV P-Cygni profiles of OB-stars in the Magellanic Clouds (observed with HST), and the galaxy (observed with IUE) are analyzed empirically using a line formation procedure similar to the one described by Lamers et al. (1987). The assumption of a constant microturbulencev _turb throughout the wind is dropped and replaced by a radially increasing turbulence parameterv _turb(v), thus improving the fit for the emission peaks substantially, and at the same time avoiding the need for a justification of extremely supersonic turbulence in the vicinity of the wind's sonic point. The Sobolev optical depth is determined interatively at fixed velocities in the wind, which removes the bias introduced by the choice of a specific parameterization function. Where it was possible and necessary a full photospheric spectrum was used to illuminate the wind line. The terminal velocitiesv are are found to be largest in the Galaxy, smallest in the Small Magellanic Cloud, and intermediate or similar to galactic in the Large Cloud.     

Gamma Rays From Molecular Clouds

High energy -rays from individual giant molecular clouds contain unique information about the hidden sites of acceleration of galactic cosmic rays, and provide a feasible method for study of propagation of cosmic rays in the galactic disk on scales 100 pc. I discuss the spectral features of ⁰-decay -radiation from clouds/targets located in proximity of relatively young proton accelerators, and speculate that such `accelerator+target systems in our Galaxy can be responsible for a subset of unidentified EGRET sources. Also, I argue that the recent observations of high energy -rays from the Orion complex contain evidence that the level of the `sea of galactic cosmic rays may differ significantly from the flux and the spectrum of local (directly detected) particles.     

Gamma rays and neutrinos as complementary probes in astrophysics

Electrons are more susceptible to energy losses in magnetic fields and photon fields than protons. Hence, photons at various wavelengths, including gamma rays, bring more readily information on high-energy electrons than on protons. Neutrinos provide a unique tracer for protons. Furthermore, at high energies the neutrino flux can considerably exceed the gamma-ray flux, as gamma rays above ～1 MeV are degraded by γ-γ interactions in compact high-intensity sources. Active galactic nuclei (AGN) with outputs >10⁴⁵ ergs s^?1 and dimensions ～10¹⁴ cm would constitute such sources. If the AGN are powered by ultra-massive black holes, then these numerical conditions are satisfied, and at high energies the flux J _v >J _γ . Berezinsky and Ginzburg have pointed out that the photon intensity around spinars is not sufficient to cause gamma-ray degradation. These authors have demonstrated that the measurement of neutrino flux, combined with the measurement (or upper limit) of gamma-ray flux would show whether the active galactic nuclei are powered by massive black holes or spinars. We estimate that gamma rays would be degraded at spinars, too, at energies >1 GeV.     

Abell 1367 and 1060 observed with EXOSAT

Summary Two clusters of galaxies have been observed with EXOSAT to map the distribution of cool gas. For A1367 we have detected the two brightest sources found by EINSTEIN. We confirm the identification of a X-ray source with a blue object near NGC3842. By comparing the EXOSAT and the EINSTEIN count rates we can state, either that the source is variable, or that the HI column density is low and the spectrum is either a power-law with a spectral index > 1.5 or thermal with a temperature below 6×10⁶ K. For A1060 we confirm the classification from optical data of NGC3311 as the dominant galaxy but we find no evidence for a central source as inferred from the EINSTEIN data. We derive an accretion rate of 10M_o/yr.     

The Energy Spectra and Anisotropies of Cosmic Rays

The existing paradigm of the origin of Galactic cosmic rays places strong supernovae shocks as the acceleration site for this material. However, although the EGRET gamma-ray telescope has reported evidence for GeV gamma rays from some supernovae, it is still unclear if the signal is produced by locally intense cosmic rays. Although non-thermal X-ray emissions have been detected from supernova remnants and interpreted as synchrotron emission from locally intense electrons at energies up to 100 TeV, the inferred source energy spectral slopes seem much steeper than the electron source spectrum observed through direct measurements. It remains the case that simple energetics provide the most convincing argument that supernovae power the bulk of cosmic rays. Two characteristics which can be used to investigate this issue at high energy are the source energy spectra and the source composition derived from direct measurements.     

Model-independent requirements for the source of positrons in the galactic centre

Model-independent requirements for the positron source in the galactic centre are formulated. From the known physical processes of positron production the most probable seems to be the e ⁺ – e ^– pair production as a result of photon-photon collisions. When certain conditions are satisfied, the efficiency of positron creation due to this mechanism can reach values 10%, which is comparable with the observed ratio of the annihilation line photon luminosity to the continuum one at E > 511 keV. Such a situation can be realized: (i) in a thermal pair-dominated mildly relativistic plasma, and (ii) on the development of a nonthermal electromagnetic cascade, initiated by relativistic particles in the field of ambient X-rays. Future gamma-ray observations at ultrahigh energies can be crucial to the choice of the model.