IRAS Explanatory Supplement
VII. Analysis of Processing
J. Extended Source Products
J.1 Zodiacal Emission Effects
Since the detector signals were DC coupled, the extended source data products include emission from sources on all angularscales. In particular, emission from interplanetary dust, or zodiacal emission was a prominent large scale signal component in all survey bands. The contribution of the zodiacal emission to the observed intensity toward any direction on the celestial sphere depended upon the integrated emission from the interplanetary dust along the line of sight at the time of observation, which in turn depended upon the Earth's orbital position within the dust cloud (time of year) and the spatial structure of the cloud. Hence, repeated measurements in a given celestial direction at substantially different times of year gave different results. for example, for elongations near 90°, the 1° daily motion of the Sun produced about a 2% change in the sky brightness near the ecliptic plane in all survey bands, with the brightness decreasing with increasing elongation. Because the dust symmetry plane near 1 AU from the Sun was inclined with respect to the ecliptic plane, the brightness at high ecliptic latitudes varied sinusoidally annually with a peak-to-peak variation of about 20% in the bands.For the purposes of preparing the extended source data products, the hours-confirming scans (typically separated by about two hours) were treated as if they had been obtained simultaneously and the results were averaged together. The difference in zodiacal brightness between one orbit and the next was less than 0.1% of the ecliptic plane brightness; within the maximum 36-hour spacing of an hours-confirming coverage, the brightness difference was less than 3%. The data from the three hours-confirming surveys were reduced separately and presented as three distinct sets of images. The data from the first two surveys, which each covered about 95% of the sky, were interleaved during the first 7 months of the survey, withthe two weeks-confirming surveys typically separated by about 10 days at any given celestial position. The third survey, which covered 72% of the sky, was obtained during the last four months.
As an aid to modeling and extracting the zodiacal emissioncontribution in any of the extended source products, the time-ordered Zodiacal Observation History file (ZOHF) was created. This file contains, at 0.5 degree sample spacing, the time (UTC), celestial coordinates, Sun-referenced observing angles, and measured brightnessesin the survey bands for all observations included in the extended source image products.
J.2 Effective Resolution
As discussed above, the data for the sky plates and Galactic plane maps were smoothed in the time domain to a sample spacing corresponding to 2'. Additional smoothing was inherent in theprocess of projecting these data into image grids. It was estimated that the resulting effective resolution, or, ability to distinguish point sources in these images is about three pixels, or 6'.J.3 Tests of Extended Source Calibration Consistency
The primary test of the extended source calibration consistency was an examination of the measured brightness in the survey data to show that the absolute baseline was being properly controlledthrough the daily observations of the baseline photometric reference area (the TFPR) as described in Section VI.B.3.One check consisted of observing the time variation of the brightness of the TFPR in the Zodiacal Observation History file (ZOHF). The mean of the ZOHF data followed the model of the TFPR given in Table VI.B.1 to within 5% in all bands except for a short period of at me about 20 days before the end of the mission where deviations approaching 10% brighter than the model occurred. The scatter of the measurements was non-Gaussian, most of the points scattered toward higher brightness, with 50% of the points lying within 7%, 5%, 12% and 12% of the model in the 12, 25, 60 and 100 µm, respectively.
As a second check the sum of the north and south ecliptic pole brightness, as measured from selected single scans of the telescope, was compared with parameter B0 of the TFPR model in Table VI.B.1. The geometrical basis of the TFPR brightness modelpredicts that the sum of the north and south brightness will be constant and equal to twice B0. This was found to be true to within 5% in all bands.