There are three sources of background counts for data taken with the Copernicus spectrometer - particles (cosmic rays), stray light, and scattered light. Each of these presents specific reduction issues which are briefly discussed below, users of the Copernicus archive are urged to consult the references mentioned below and other papers from the literature before using the data.
The particle background was routinely calculated for observations taken after the first eight months of operation as a function of spacecraft latitude and longitude at the time of observation. To allow for the correction to the extrapolated predictions of particle background flux, dark counts were measured about every four orbits. For the U1 and U2 tubes the counts due to particles was generally insignificant except when the spacecraft was in the South Atlantic Anomaly where observations were normally not taken (see also Rogerson et al. 1973 and the Description of the Princeton Experiment in Section 3 of the Guest Investigator's Guide). For many stars, the data from the Copernicus team at Princeton includes the background contributions due to charged particles for each scan. For data that does not include this information in the file a look-up table is consulted which gives the background count rate as a function of spacecraft position and time of observation. The particle background was removed during creation of the stacked scans.
Stray light reached the phototubes through out-gassing holes in the spectrograph. For observations taken using the U1 tube the stray light from the out-gassing holes was often blocked by positioning carriage-2 just longward of the U1 exit slit to cover the holes; however, not all data was taken with the out-gassing holes blocked. The Copernicus archive lists data as being either blocked or unblocked to denote whether carriage-2 was known to be in the correct position to block the vent holes. Some scans are marked as unblocked that actually were blocked, however the position of the U2/V2 tubes is not known so the scans are flagged as unblocked. For example, the following scans of the SIII and SiII 1190Å lines have the same background level (as judged by the level of the continuum and the depth of the core of the strong SiII line), but only the first set of scans have corresponding U2 scans to give the position of carriage-2:
Stray light was generally the dominant background source, so it is important not to mix blocked and unblocked data when stacking. An empirical correction for stray light for the U1 tube in any given (unblocked) scan can be derived in a manner similar to that used for scattered light below. The correction of U2 data for stray and scattered light is described by Bohlin (1975).
In addition to the particle and stray light backgrounds described above, there is a residual background of about 10 percent or more of the continuum for U1 observations using the U1 tube, this background is generally attributed to scattered light from the grating. In most cases the scattered light correction must be derived from residual counts in the bottom of saturated stellar or interstellar lines. The observing scheme usually used included the observation of strong interstellar lines (including the Lyman lines of HI, CII 1036 and 1334Å, CIII 977Å, NII 1084Å, OI 1302Å, and SiII 1260Å), the background was then empirically determined by assuming that some or all of these features were saturated, and thus should have no residual counts in the line cores (see Rogerson et al. 1973b; Morton 1975; Bohlin et al. 1983). In determining the background it is important to not mix blocked and unblocked scans as the background level varies widely between the two (although as mentioned above some scans were blocked but the necessary carriage-2 position information was not recorded, such "unblocked" can be combined with the blocked scans after one has determined the scans were blocked by the continuum and background levels).
One additional consideration in determining the scattered and stray light backgrounds is the fact that the Copernicus spectrometer varied in sensitivity both as a function of wavelength and time. For this reason, one cannot use the ratio of the background to the local continuum level determined for one scan (or set of scans) to calculate the background for scans of the same wavelength range taken at a much different time. Instead, one must determine the background for each set of scans using only scans taken at approximately the same time.
material compiled by jtl