Annu. Rev. Astron. Astrophys. 1991. 29: 89-127 Copyright © 1991 by Annual Reviews. All rights reserved

5.6 More Evidence For Scattering From Dust?

Joubert et al (54) present data (Figure 17) that also seem to argue for an origin of the diffuse ultraviolet background in an extragalactic component (of 400 units at 2200 Å, and 700 units at 1690 Å, at N_H = 0), plus a component that results from dust scattering.

Figure 17. Joubert et al (54) obtained this dependence of cosmic ultraviolet background radiation on neutral hydrogen column density. Each point represents an average over a 30-square-degree region. The scatter is real and must be compared with the smoothness of the data in Figure 16. Apollo-Soyuz data of Figure 15 give many well-established upper limits of 300 to 400 units. Experiments that individually seem clearly to indicate the presence of dust scattering disagree with each other profoundly in detail. This figure is from Joubert et al (54), with permission.

If we accept the dust-scattering component as real because of its correlation with N_H in Figure 17, then we must repeat our remarks about the very different optical parameters that must obtain for the interstellar grains at 1100 Å. On the other hand, there is a profound difference between the data of Figure 17 and those of Figure 16. In Figure 17, where each point represents an average over a large (30° x 30°) region of sky, a large scatter appears; a scatter entirely absent in Figure 16. That the Joubert et al data scatter is real, has been reaffirmed by Lequeux (61) and is inexplicable in terms of a dust scattering origin. It could be due to a dust-scattering origin for the lower envelope in Figure 17 plus a highly patchy extragalactic background. (Even that much light originating in dust scattering would require a large change in grain optical properties with wavelength.)

The data of Figure 17 again disagree with the Apollo 17 upper limit of 400 units, and with the more reliable Apollo-Soyuz upper limits of 300 units, because the signal in Figure 17, though patchy, is never as low as 300 units, even at moderate Galactic latitudes.

The Joubert et al data (54) came from an experiment located at very high altitude, so time-variable dark current should not be a problem. However, pointing was not completely stable, and undetected faint stars would of course give a patchy appearance as appears in the data. With the Joubert et al field of view of 4 x 10^-4 sr, an undetected unreddened 9th magnitude A0 star would produce a false background of 300 units, while a 10th magnitude B5 star would give 870 units. The experimenters have made every effort to assure themselves that that is not what is happening.