This Instrument Science Report discusses a phenomenon of GHRS performance that arose as a result of observations as part of Cycle 5 program 6075. The PI for that program is R. Gilliland, an STScI astronomer and former Instrument Scientist for the GHRS. Program 6075 was highly unusual in that it obtained a series of short-exposure spectra of a star over many orbits in ACCUM mode and during the occultation portion of each orbit a similar series of wavelength calibration exposures were obtained, also with ACCUM. Ordinarily the occultation portion of the orbit is not used because of the scheduling difficulties imposed, but the science goals of this program required this to be done. Also, more typically a PI will obtain a single (or a few) long exposures as an ACCUM.
The observations obtained are detailed in Table 1. As a first look at the data, Gilliland plotted the total counts in each spectrum as a function of time, to see if stellar variability was present. What was revealed instead was a regular decline of the observed counts of roughly 10% over the course of each orbit (Figure 1). A plot of counts in the wavecals showed a decline of about the same percentage (Figure 1). Similar declines have been seen previously in stellar spectra obtained with the GHRS Small Science Aperture (SSA) and have been attributed to motions of the star in the aperture that reduce the throughput. In this case that explanation could not account for declines in both the star and wavecal simultaneously. Dr. Gilliland brought this to our attention, and here we provide additional information and a likely explanation.