Stuart Bowyer

The ultraviolet band of the spectrum is generally defined as extending from the end of the soft sky x-ray band at 100 Å to the cutoff imposed by the transmission of the atmosphere at ~ 3000 Å. It is best divided into two separate bands: the extreme ultraviolet (EUV) from 100- ~ 1000 Å and the far ultraviolet (FUV) from 1000-3000Å. If one considers the cosmic ultraviolet background , each of these bands is sampling vastly different regions; the EUV background is expected to be produced by emission from the hot component of the local interstellar medium, whereas the FUV background could be produced by a variety of mechanisms in our galaxy and may have contributions from a variety of exotic and cosmological processes. Studies of the EUV background are in their infancy but show great promise. The FUV background has been the subject of a substantial amount of work from the beginning of space research. The subset of the FUV band from ~ 1300-2000 Å has been studied most extensively. Scattering from the very intense, diffuse hydrogen Lyman alpha line at 1216 Å, which would severely compromise any observational data, can be eliminated through the use of crystal filters that block all radiation below ~ 1300 Å. Longward of 2000 Å, zodiacal light contributes strongly to the background in a complex and as yet poorly determined manner, and it is extremely difficult to separate the stellar contribution from the truly diffuse flux. It is a statement of the difficulty of this work that despite considerable effort substantial progress in our understanding of this background has only recently been achieved. This effort has clearly been worthwhile; recent results provide information on such diverse topics as the character of the long-prophesied but never established galactic halo, star formation rates in the universe, and the character of dust in our galaxy.


From the beginnings of space research, attempts were made to measure the cosmic FUV background. This work was strongly motivated by the hope that in this waveband a true extragalactic flux could be detected and characterized. Theoretical speculation as to possible sources for this radiation was unconstrained by the available data and included such diverse processes as emission from a lukewarm intergalactic medium, emission from hot gas produced in a protogalaxy collapse phase in the early universe, the summed emission from a star formation burst phase in young galaxies, and photons from the electromagnetic decay of real or hypothetical exotic particles that were produced, or may have been produced, in the early universe.

It was the expectation that all of the problems that had bedeviled attempts to measure the cosmic extragalactic flux in the optical band would be overcome: The zodiacal light would be absent because the Sun's radiation falls rapidly in the ultraviolet, the measurements would be made above the atmospheric airglow, and the difficulty of separating the diffuse background radiation from that produced by stars would be reduced or even eliminated because the UV-producing early stars would be limited to the Galactic plane.

The first 20 years of measurements, carried out by a number of groups in at least five major countries, indicated that the flux was uniform across the celestial sphere and hence was cosmological in origin. Estimates of the intensity of this flux, however, varied by three orders of magnitude with no clustering around a mean. In retrospect, there are a number of reasons for these discrepant results; perhaps it is most useful to state that these results provide empirical evidence that measurements of this type are intrinsically difficult.

A turning point in the study of this background occurred in 1980 when data obtained with a FUV channel of a telescope flown as part of the Apollo-Soyuz mission were analyzed and published. These data exhibited a correlation between intensity of the background and galactic neutral hydrogen column as derived from 21-cm radio measurements. Although these initial results were criticized on a variety of grounds, they were quickly confirmed by independent rocket and satellite measurements. The intensities obtained in these experiments were consistent within a factor of 2, and they varied from roughly 200-1500 photons cm-2 s-1 sr-1 Å-1 depending upon the galactic hydrogen column. These results showed that the vast majority of the FUV flux was connected with processes in our Milky Way galaxy and was not, in fact, extragalactic in origin. Nonetheless, all of the data sets were consistent with a small part of this flux being isotropic. However, there was no way to determine whether this component was due to residual airglow processes at satellite altitudes, due to processes occurring within the galaxy, or was extragalactic in origin.

In the first half of the 1980s, a variety of processes was suggested as possible sources for the galactic component of this radiation, including starlight scattered by dust, molecular hydrogen emission, line emission from a hot (~ 105 K) interstellar gas, two-photon recombination radiation, and others. A variety of extragalactic processes was suggested as the source of the small isotropic component. Given the character of the existing data; there was no way to establish which, if any, of the processes mentioned previously were contributing to the FUV background. Two experiments carried out in the second half of the 1980s proved to be especially productive in establishing the sources of this radiation.

The first of these was a FUV imaging detector flown as a piggyback experiment at the focal plane of a 1-m EUV/FUV telescope developed as a collaborative effort between the United States and Germany. Data from this imager were subjected to a power spectrum analysis in a search for any component of the FUV background flux that was correlated with angular separation. This search was motivated by the fact that galaxies are known to cluster, and the scale of this clustering is well characterized by a two-point correlation function derived from optical data. A component of the FUV background was, in fact, found with the same angular correlation as that found in the optical, leading to the conclusion that this component is indeed extragalactic in origin and is the summed flux of galaxies at great distances. Because of the cosmological redshift, the FUV flux emitted by galaxies is shifted out of the observed FUV band at distances corresponding to about one-third of a Hubble time.

The detection of a cosmological component has an important astrophysical consequence. The intensity of this flux (about 50 photons cm-2 s-1 sr-1 Å-1) is consistent only with a relatively low and constant star formation rate in galaxies for the time scales indicated. It also limits so-called star burst galaxies to be less than 10% of the total population of galaxies in this period.

A second major experiment that provided a variety of new data relevant to this field was a FUV diffuse spectrometer which flew in 1986 on the last shuttle flight before the Challenger tragedy. This instrument was sensitive from about 1350-1850 Å with about 16 Å resolution. Astronomical data were obtained from eight different directions in the celestial sphere with a wide range of neutral hydrogen column densities.

Several important results came from the study of the data obtained with this instrument. Emission lines from a collisionally excited hot (~ 105 K) gas in the interstellar medium were discovered. Combining these results with absorption data obtained with the International Ultraviolet Explorer proved the existence of the long-postulated, but unproven, hot galactic halo. The mass infall rate of ~ 10 Msmsun yr requires that this halo must be continually replenished, which provides direct experimental confirmation of the ``galactic fountain'' model of the galactic halo gas. Data obtained with this instrument on FUV radiation scattered from dust in directions with relatively high column densities of neutral hydrogen showed that this dust has a relatively low albedo (omega ltapprox 0.2) and that it scatters FUV radiation almost isotropically, indicating the dust itself is relatively small and roughly spherical. There are strong indications that at least traces of this dust are present everywhere, including regions in directions of very low hydrogen column density.

The cosmic FUV flux from 1400-1850 Å is shown in Fig. 1. The lower curve shows the spectrum in a direction of low neutral hydrogen column. A variety of processes contributes to this flux; the most intense spectral lines are emitted by the hot interstellar medium. The upper curve shows the spectrum in a direction of higher neutral hydrogen column. The majority of this radiation is starlight scattered by dust in the interstellar medium, but other processes also contribute to this flux.

Figure 1

Figure 1. The cosmic FU flux from 1400-1850 Å. The lower curve shows the spectrum in a direction of low neutral hydrogen column. The broad line at 1550 Å is the C IV from hot interstellar gas. The upper curve shows the spectrum in a direction of higher neutral hydrogen column. The majority of this radiation is starlight scattered by interstellar dust, but other processes also contribute to this flux.


Early attempts at measurements of the cosmic EUV background were made with broadband filters which included the resonantly scattered solar lines of He II at 304 Å and He I at 584 Å. These lines, whose study is important in its own right, dominated the data and made estimates of a truly cosmic flux impossible. The first useful estimates of the cosmic diffuse EUV background below 500 Å were obtained with an EUV telescope flown on the Apollo-Soyuz mission. These data were also obtained with a broadband filter, but they had an important advantage over previous results, in that a small field of view was employed and a very extensive data set was obtained. The geocoronal ionized helium responsible for scattering the He II 304 Å flux is confined to a few Earth radii by the Earth's magnetic field. In the antisolar direction, the Earth's shadow leaves the helium in darkness, greatly reducing the scattered intensity. By analyzing the extensive data obtained in the antisolar direction, significant upper limits to this flux were obtained.

Several broadband measurements of the EUV background have been made at wavelengths longer that 500 Å. In addition to measurements in the 500-800 Å band, several groups have obtained upper limits to the diffuse flux from 912-1080 Å with instruments with relatively narrow bandpasses.

Two relatively coarse spectrographic measurements of the cosmic EUV background have been carried out. High galactic latitude observations were obtained with 40 Å resolution during the interplanetary cruise phase of the Voyager 2 mission near the orbit of Saturn. The EUV part of these observations was dominated by the solar He I 584 Å flux resonantly scattered from neutral helium flowing through the solar system. The effects of this bright He I line, along with several other sources of background, were removed from the raw spectra by analysis, and upper limits to the EUV background were derived.

Spectroscopy of the diffuse EUV background below 500 Å is in its infancy. Grazing incidence optics must be used in order to obtain any reasonable throughput, and grazing incidence spectrometers from diffuse radiation are technically difficult. An instrument that was flown on a sounding rocket in 1986 obtained upper limits on the EUV flux from 90-700 Å with about 20 Å resolution. The existing data on the diffuse cosmic EUV background are shown in Fig. 2.

Figure 2

Figure 2. The cosmic EUV flux from 100-1000 Å. Most of the measurements have been made with broadband detectors and are shown as solid horizontal bars. Most are upper limits to the diffuse flux. The curved dotted line is an upper limit to the diffuse flux obtained with a low-resolution (~ 40 Å) spectrometer. The solid curved line was also obtained with a low-resolution (~ 20 Å) spectrometer; some lines or line complexes may have been detected in this measurement.


Investigations of the FUV background were initiated in the early phases of space astronomy. Substantial efforts were devoted to this work with disparate results. The discovery that most of this flux is produced within our galaxy opened a new chapter in the study of this radiation. Subsequent work in this field has established the existence and character of the hot galactic halo, has provided limits on the star formation rate in the universe for the last third of a Hubble time, and has defined the character of a component of interstellar dust. Future work in this field will explore questions such as: Why is this flux only roughly, and not exactly, correlated with neutral hydrogen column? Is this correlation the same in all directions of the sky? What is the true extragalactic background level? What is the correlation of the FUV background with the infrared background? and higher-order questions. Answers to these questions will certainly lead us to new insights about our galaxy and the universe. The exploration of the cosmic EUV background is in its infancy, and the existing data are only suggestive of the underlying processes believed to be in operation. Another order of magnitude in sensitivity and spectral resolution may be required for these studies to provide definitive results.

Additional Reading
  1. Davidsen, A., Bowyer, S., and Lampton, M. (1974). Cosmic far ultraviolet background, Nature 247 513.
  2. Paresce, F. and Jakobsen, P. (1980). The diffuse ultraviolet background, Nature 288 119.
  3. See also Diffuse Galactic Light; Interstellar Medium, Galactic Corona; Zodiacal Light and Gegenschein.