Ultraviolet Background Radiation

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

5.5 Evidence For Scattering From Dust?

Fix, Craven & Frank (25) used the imaging experiment aboard Dynamics Explorer I to obtain the ultraviolet background data that appear in Figure 16. In this experiment the use of an imager renders stars less of a concern; also use of a very high orbit means that airglow and time-dependent dark current should not be factors. The residual they report at 1500 Å, shown in Figure 16, consists of a component that is independent of Galactic latitude, plus a component that shows a clear dependence on Galactic latitude. A natural interpretation is an extragalactic component plus a component originating in the light of Galactic-plane OB stars scattering from dust. The correlation with neutral hydrogen column density that they find gives the strength of the extragalactic component as 530 ± 80 units, in reasonable agreement with other determinations (longward of 1216 Å) reported above.

Figure 16. Fix, Craven & Frank (25) obtained these data showing the diffuse ultraviolet background as a function of galactic latitude. The distribution is very smooth, and at moderate latitudes exceeds the Apollo-Soyuz upper limits of Figure 15. This figure is from Fix et al (25), with permission.

Consider first these observations at face value and ask, what can we conclude concerning the optical properties of interstellar dust? The argument has already been rehearsed above: the lack of detectable signal from Voyager at b = 40° means that a drastic change must take place in the scattering properties of the interstellar grains between 1500 and 1100 Å, in the sense that either the grains have a much lower albedo at 1100 Å, or are much more strongly forward scattering at 1100 Å, or some combination of the above.

Can the observations be criticized? Note that only photometry is involved, not spectroscopy, so the signal has no internal character that can be examined in hope of gaining an understanding of the signal's origin. Next, the authors determine their dark current through studying the count rate with different filters in place, and say that the count rate due to cosmic rays should be independent of the filter. But this is not necessarily so, because part of the dark count could well originate in ultraviolet fluorescence in the material of the filter itself, and it is known that different filters may have very different rates of fluorescence. If this is a factor, however, it will only affect the ``extragalactic'' component, not the ``dust scattering'' component of the signal. With those exceptions, there is nothing else internally to criticize in the observations. In particular, the observations lack the ``patchy'' appearance expected if undetected-star contamination is a problem.

Externally two problems exists with the data. The first is the disagreement with the Apollo 17 upper limit, at moderate latitudes, of 400 units at the same wavelength. Fix et al find 800 units. In view of the stellar correction problem on Apollo 17, however, some do not find this argument compelling. Less easy to dismiss is the clear disagreement with the upper limits of 300 units (91) from Apollo-Soyuz, that appear in Figure 15. As argued before, the problem is to prove that the higher background is not spurious; it has been argued above that the Berkeley upper limits are reliable. Also, the data shown in Figure 16 represent the average, with ± 1 sigma error bars, of four different cuts through the galactic plane, so ``looking in different directions'' is unlikely to be the answer as to why Figure 16 differs from Figure 15, but this problem cannot be studied in detail because the galactic longitudes of the Berkeley observations are not published.