|Annu. Rev. Astron. Astrophys. 1991. 29:
Copyright © 1991 by . All rights reserved
5.3 Example of a Conflict
The data that have been discussed so far accord well with a picture in which (1) a uniform extragalactic background of ~ 400 units is present from near 1216 Å to at least 3200 Å; (2) only an upper limit of 100 units exists below 1216 Å; and (3) no strong evidence exists for any starlight scattered from dust, except at the very lowest Galactic (or Gould) latitudes, and in a limited longitude range.
In this section we present an example of conflicting data. We focus first on a case in which several observers look at the same location, with similar field of view, and at the same wavelength.
Weller (114) published an important measurement of the diffuse background made from a spacecraft whose altitude (130,000 km) was so high that airglow and time-variable dark count could not be deleterious factors, and important because the instrument field of view was large (8° FWHM), which made the stellar correction very insensitive to precise knowledge of any faint stars that might inadvertently be in the field of view. Weller compared his results (Figure 14) with Apollo 17 results, with results from the Apollo-Soyuz mission (91) (see next section) and also with another Johns Hopkins Aries rocket result (1). As is apparent in Figure 14, the results from Weller, Aries, and Apollo 17 agree on a low background, of order 300 units; while in contrast Apollo-Soyuz, in repeated measurements, finds backgrounds of 900 to 1500 units.
Figure 14. Four observations of diffuse ultraviolet background radiation near the north galactic pole are compared by Weller (114). The data of Weller (114, oval at 180 units), Henry et al (37, quadrilateral at 300 units), and Anderson et al (1, filled rectangles at 285 units) agree reasonably well, while Apollo-Soyuz observations (filled circles give much higher values. This figure is from Weller (114), with permission.
In earlier discussion I argued that a large false background is not unlikely in any experiment while the reverse (missing a large signal that is actually there) is unlikely. It is the Apollo-Soyuz result that gives the high count rate. If the ``low-background'' observers subtracted out their signal, erroneously attributing it to stars (as occurred to some extent in Apollo 17), then the argument does not hold. But in the case at hand, the total stellar corrections are much less than the claimed Apollo-Soyuz backgrounds, and it would seem that no reasonable case can be made that the Apollo-Soyuz results are correct.