Published in the June, 1999 of Pub. ASP (vol. 111, 722)
Myron A. Smith 1
1STScI/CSC, Science Programs, Space Telescope Science Institute, 3700 San Martin Dr. Baltimore, MD 21218; Email: email@example.com
The extraction of net spectral fluxes from echellogram images for the creation of the final IUE archive was accomplished by the processing system ``NEWSIPS" computing a background surface from unilluminated pixels over a wide area of the image for any given echellogram order. Although his technique facilitates the computation of net fluxes by automated computer-processing techniques, it does not always lead to optimal background solutions, especially when interorder flux between closely spaced orders at short-wavelengths is comparable to the net flux to be determined. In this study we investigate the reason for a reported spurious time-dependence in NEWSIPS-computed background fluxes of high dispersion Short Wavelength Prime (SWP) camera images in the region of Lyman obtained for the the calibration target Scorpii (B0.2 V). We show that the temporal change of the zero-flux (``null") surface of the SWP camera changes markedly with position across the camera field. As this IUE camera aged, these changes caused a relaxation in a steep gradient in the region where short-wavelength fluxes were imaged. The NEWSIPS background-determining algorithm could not follow this gradient accurately for pre-1982 high-dispersion images, a fact which produced an undercorrection for the background near the Lyman line for these images. Ironically, the changed null surface actually permits better background flux solutions in the short-wavelength region of the camera for late-epoch images. As a consequence of time-dependent errors in both the background-determination and in the Intensity Transfer Function for the SWP camera, both gross and background fluxes of spectra may exhibit spurious changes in time. We quantify these changes for the continuum net fluxes of high-dispersion SWP images of Sco over the lifetime of the IUE mission.
stars: individual Sco - stars: individual: Cas - techniques: data analysis, techniques: image processing