optical light from objects having continuous spectra with absorption lines (galaxies, stars, etc.) and a surface brightness less than the natural night sky depend on the degree of light pollution. Furthermore, as it is difficult to extrapolate astronomical instrumentation over 20 years, we will consider the environmental impact of the SPS on data acquisition using existing techniques. Two types of detectors are routinely used for studies of faint objects— photographic emulsions and photon detectors. Photographic emulsions have a large capacity for information storage through the huge number of pixels available, low cost, and acceptable uniformity of response at the expense of linearity and efficiency (quantum efficiencies of at most a few percent). Plates also have the undesirable property that above a threshold, the signal- to-noise ratio for a given area essentially becomes independent of the amount of flux collected. In contrast, photon-counting detectors have higher quantum efficiencies which in principle, though not always in practice, are limited in accuracy only by photon statistics. Detectors of this type provide a best case for dealing with the problem of sky subtraction. Now consider an interesting object (star, galaxy, cluster, etc.) which is 2 seen against a natural sky background of surface brightness N photons/cm 2 s arcsec through some spectral bandpass defined by either a spectrometer or filter. We assume that we are in the optical spectral region and are not centered on a strong night sky emission line; so we will use V band brightness measurements as a representative case. The region of interest within a given object will have a surface brightness gN where often g«l. We will compare this circumstance with the same measurements taken against a sky with an increased brightness pN (p>l), in which case the object surface brightness is (g/p)pN. The critical question then is how do our observational capabilities depend on the magnitude of p? Since the maximum signal-to-noise ratio on a photographic plate over some fixed physical area is set by the emulsion and processing and not by the photon flux, then the detectable signal against sky is always the same fraction f of the sky brightness. Thus, under ideal conditions a given platetelescope combination might allow an intensity level of fN to be recorded,
RkJQdWJsaXNoZXIy MTU5NjU0Mg==