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

3. OBSERVATIONS

3.1. Measurement Challenges

The CIB has few defining observational characteristics on which to base a detection. The radiation is of extragalactic origin and is therefore expected to be isotropic on large scales. There is no distinctive spectral signature. The spectrum will depend in a complex way on the characteristics of the luminosity sources, on their cosmic history, and on the history of dust formation and the distribution of dust relative to the luminosity sources. Because discrete sources contribute at least part of the CIB, the background will have fluctuations superimposed on the isotropic signal. In Sections 3.1 to 3.5 we review direct measurements of the CIB based on searches for isotropic, extragalactic infrared radiation. Studies of the fluctuations are discussed in Section 3.6.

Direct measurement of the infrared background is both technically and astrophysically challenging. The technical challenge is to make absolute sky brightness measurements relative to a well-established zero-flux level. Emission from telescope and instrument components and the Earth's atmosphere must be eliminated. Scattered and diffracted light from the very bright local sources (Sun, Earth, and Moon) must also be strongly rejected. In practice, this requires that observations be conducted with carefully designed, cryogenically cooled instruments located above Earth's atmosphere. Confident measurement of the CIB requires sufficient observation time to identify and eliminate potential sources of systematic measurement errors.

The fundamental astrophysical challenge for direct CIB measurement is definitive discrimination of the CIB from the many bright celestial contributors to the sky brightness. These include discrete sources, such as stars and other compact sources within the Galaxy, and diffuse sources such as light scattered and emitted by interplanetary dust (IPD) and emitted by interstellar dust. At wavelengths longer than ~ 400 µm, the CMB becomes dominant and must be discriminated from the CIB. Figure 2 shows the measured spectral energy distribution of the sky in the direction of minimum H I column density in the Galaxy, the "Lockman Hole" at Galactic coordinates (l, b) ~ (150°, +53°) [geocentric ecliptic coordinates (, ) ~ (137°, +45°)] (Lockman et al. 1986). Figure 2 also shows the individual contributions from the foreground sources as determined by the COBE team (Section 3.4.2). Even at this high ecliptic and Galactic latitude, the largest contribution to the sky brightness from 1.25 to 140 µm comes from the IPD. Starlight is substantial from 1.25 to 3.5 µm, and interstellar dust emission is strong for wavelengths greater than 60 µm. There are two spectral windows most favorable for finding a faint extragalactic background: (a) the near-infrared window near 3.5 µm, which is the minimum between scattered and emitted light from the IPD, and (b) the submillimeter window between ~ 100 µm, the peak of the interstellar dust emission, and the CMB.

Figure 2. Foreground contributions to the DIRBE (Diffuse Infrared Background Experiment) data at 1.25-240 µm in the Lockman Hole area: observed sky brightness (open circles), interplanetary dust (triangles), bright Galactic sources (squares), faint Galactic sources (asterisks), and the interstellar medium (diamonds). (Solid circles) The residuals after removing all foregrounds from the observed brightness. [Adapted by permission from Hauser et al. (1998).]