Annu. Rev. Astron. Astrophys. 2001. 39: 249-307
Copyright © 2001 by . All rights reserved

Next Contents Previous


In this review we have shown that knowledge of the CIB, its sources, and its implications has advanced dramatically in the past few years. We now have claimed detections of the background at near-infrared and far-infrared/submillimeter wavelengths based on data from the COBE and IRTS missions. Useful upper limits at other wavelengths come from these direct measurements and indirectly from observations of TeV gamma-rays, whereas lower limits from ground-based galaxy counts and counts from ISO and SCUBA surveys further limit the observational uncertainty in the CIB (Section 3.10). Although observational knowledge of the CIB still has much room for improvement, the detections and limits presented in this review already have many important implications.

  1. The total intensity of the EBL from UV to millimeter wavelengths lies in the range 45-170 nW m-2 sr-1. The energy in the 0.16 to 3.5 µm range is still quite uncertain, with present limits of 19-100 nW m-2 sr-1. The energy in the 3.5 to 140 µm range, 11-58 nW m-2 sr-1, remains similarly uncertain because of the dominance of the interplanetary dust emission. The most certain background determination is in the 140 to 1000 µm range, which contains 15 ± 2 nW m-2 sr-1. The nominal value for the total CIB intensity (1-1000 µm) is 76 nW m-2 sr-1. The nominal value for the total extragalactic background (0.16-1000 µm) is 100 nW m-2 sr-1.

  2. Using nominal EBL values, the energy in the background appears about equally divided between direct starlight (52%) and starlight that has been absorbed and reemitted by dust into the 3.5 to 1000 µm wavelength regime (48%). However, the uncertainties in the present measurements allow the fraction from dust emission to range from 20%-80%.
  3. The dominant energy source for the EBL from 0.16 to 1000 µm is apparently thermonuclear fusion reactions that convert hydrogen into heavier elements, whereas AGN contribute about 10%-20% of the total background. If all of the nucleosynthetic energy release occurred at a redshift z = 1, then a fraction between 0.02 and 0.06 of the total baryonic matter was converted into elements heavier than hydrogen.

  4. The nominal fraction of the EBL intensity consisting of thermal emission from dust (48%) is higher than the ~ 30% found in the local universe, which suggests that dust-enshrouded objects made a larger contribution to the luminosity density in the past than at the present epoch.

  5. SCUBA observations at 850 µm identified such a population of luminous, dusty objects at high redshift. The light from these objects is comparable to the CIB intensity at 850 µm, indicating that most or all of the background at this wavelength has been resolved.

  6. ISO source surveys from 7 to 170 µm reveal steeply rising counts of dusty sources at these wavelengths. The ISO surveys near 170 µm have resolved ~ 10% of the measured CIB.

  7. Because the EBL is an integral quantity, the measured EBL spectrum does not imply a unique star formation history. Definitive determination of the CSFR history will require direct observations of the sources that make up the EBL. However, the integrated EBL implies that the average star formation rate over cosmic history is about an order of magnitude larger than the current rate.

  8. If the correlation between radio and far-infrared emission found in nearby galaxies applies to the star-forming galaxies that dominate the CIB, then these galaxies contribute about half of the cosmic radio background.

  9. The CIB provides a significant source of opacity for ultrahigh-energy gamma-rays due to electron-positron pair production. Blazars within distances of ~ 100 h-1 Mpc have therefore been used to probe the CIB, constraining the CIB intensity in the mid-infrared (~ 5-60 µm) spectral range.

  10. We have discussed four distinct approaches that have been used to model the evolution of cosmic sources with varying degrees of physical realism and astrophysical detail. Although most of these models were not primarily constructed to predict the CIB, they have been tuned to be generally consistent with the CIB intensity and spectral shape. This tuning has provided insight into the history of star formation, metal production, and dust formation and distribution.

We can anticipate that study of the CIB and its implications will continue to be vigorous for some time to come. Although there are no new space instruments under development for the specific purpose of diffuse infrared background measurements, one can expect there to be further improvements in limits or detections of the extragalactic infrared background based on the sky brightness measurements already in hand and better determinations of the foregrounds. Completion of the 2MASS survey at J, H, and K bands will permit accurate removal of the stellar foreground over the substantial sky areas needed for convincing demonstration of isotropy. The Space Infrared Telescope Facilty (SIRTF) has the potential of providing direct measurements of foreground sources at 3.5 µm, although clearly over limited sky areas (Werner et al. 2001). More extensive maps of the distribution of H+ will be available as the WHAM survey progresses, facilitating CIB discrimination particularly in the far infrared.

The largest obstacle to direct EBL measurements over much of the infrared spectrum and at UV-optical wavelengths remains the bright foreground due to scattering and emission from the interplanetary dust. This major problem could be substantially eliminated by making measurements with instruments located several astronomical units from the Sun in the ecliptic plane, or in a location out of the ecliptic plane. We are not aware of any current plans to accomplish such measurements.

Indirect constraints on the CIB from measurements of TeV gamma-rays will continue to improve as more sources over a greater range of distances are observed. Next-generation TeV gamma-ray telescopes, such as the Very Energetic Radiation Imaging Telescope Array System (VERITAS), will be able to detect fainter sources at higher redshift, providing needed confirmation of the intergalactic absorption signature in the source spectrum (Catanese & Weekes 1999). Space missions such as the Gamma-ray Large Area Space Telescope (GLAST) will help to clarify the intrinsic spectrum of the TeV gamma-ray sources from 20-200 GeV.

There promises to be continued rapid advance in measurements of discrete extragalactic sources contributing to the background. Such measurements will provide both increasingly robust lower limits to the CIB and essential information on the distance and character of these sources. Analyses of ISO data in the thermal and far-infrared are still in progress. Source counts will be carried to much fainter levels by missions such as SIRTF (Werner et al. 2001), the Far Infrared Space Telescope (FIRST; Pilbratt 2001), and the Infrared Imaging Surveyor (IRIS; Shibai 2001). The Atacama Large Millimeter Array (ALMA) will provide the sensitivity and angular resolution at submillimeter and millimeter wavelengths needed to clarify the nature of the sources being revealed in SCUBA observations. Deep, wide-field near infrared imagery from the Next Generation Space Telescope (NGST; Stockman & Mather 2001), combined with spectroscopy from NGST and large ground-based telescopes, will dramatically advance our understanding of the nature and distribution of sources at high redshift and their contribution to the infrared background. Within the next decade or so it should become clear whether the CIB arises entirely from discrete sources.

As observational knowledge of the cosmic histories of star formation and metal formation, and of the nature of the systems (AGN, starburst galaxies) making substantial contributions to the CIB improves, uncertainties in the models of background light generation associated with these processes will be reduced. This offers the prospect of a consistent, comprehensive history of the growth of cosmic structure and accompanying energy releases.


During the preparation of this manuscript we have greatly benefited from helpful and enlightening discussions with Rebecca Bernstein, Andrew Blain, Doug Finkbeiner, Dale Fixsen, Alex Kashlinsky, Richard Mushotzky, Sten Odenwald, Ray Protheroe, and Ned Wright. We thank Felix Aharonian, Rick Arendt, Chuck Dermer, Mike Fall, Jim Felten, Michel Fioc, Alex Konopelko, and Bob Silverberg for their helpful comments on sections of the manuscript. We thank Prof. T. Matsumoto for permission to reproduce the IRTS background spectrum. We thank Andrew Blain, James Bullock, Daniela Calzetti, Julien Devriendt, Alberto Franceschini, Guilain Lagache, Matt Malkan, Jonathan Tan, and Cong Xu for communicating their model results in digital form. Finally, we thank the scientific editor, Allan Sandage, for his careful review and valuable comments. Preparation of this review was partially supported by NASA grant NAG5-3899 and NASA contract NAS 5-26555 to the Association of Universities for Research in Astronomy, Inc. (MGH), and by NASA's Astrophysical Theory Program (NRA 99-OSS-01) (ED).

Next Contents Previous