ISSA Explanatory Supplement
D. Cautionary Notes
- Absolute Radiometry
- Point Source Photometry
- Photometric Errors in Low Latitude Images, ISSA Reject Set
- Confirmation of Sources
- Solar System Debris
- Residual Photon-Induced Responsivity Effects
- Calibration Change Due to Improvements in the Accuracy of Detector Solid Angles
- Ascending vs. Descending Scans
§III.C.2 and Appendix G) and at 60 and 100 µm (§ II.B.4 and III.A.2) by imperfect knowledge of the detector offsets. All bands are affected to some extent by both of these error sources.
An uncertainty of as much as 30% at 60 µm and 60% at 100 µm exists in the frequency response correction, which affects the relative brightness measurements at spatial scales above a few degrees. Read § II.B before using ISSA images for quantitative measurements.
There is an ongoing effort with COBE scientists to understand the calibration differences between the IRAS and DIRBE data for measurements of sky brightness on large spatial scales (§ IV.D.3). IPAC newsletters will contain any updated results of this work. Unless the IRAS-DIRBE comparison indicates an uncalibrated nonlinearity in the IRAS data, the accuracy of the IRAS point source calibration is not and will not be impacted by the results of this comparison.
ISSA is designed to study extended structures in the survey data and has not been optimized for accuracy for sources smaller than 5'. Point sources are not optimally analyzed with this product. The IRAS Point Source Catalog Version 2, and the IRAS Faint Source Survey (Moshir et al. 1992) should be consulted for survey information on point sources. See § IV.C for discussion of expected accuracy of point sources measured with ISSA. The IRAS Point Source Catalog and the IRAS Faint Source Survey are available for download.
The images within 20° of the ecliptic plane are of reduced quality compared to the rest of the ISSA due to contamination by zodiacal emission residuals and the zodiacal dust bands. The residual background errors at 12 and 25 µm in the reject region can be up to ten times worse than in the nonreject region. If special care is taken when estimating the background (§ IV.F) at 12 and 25 µm, the reject images remain scientifically useful. At 60 and 100 µm, the residual background errors are smaller than at 12 and 25 µm, making the reject images especially useful at the longer wavelengths.
Due to the increased sensitivity of the ISSA images, more nonconfirming objects or anomalies are visible than in the SkyFlux images. Nonconfirming objects are those objects that appear in only one HCON intensity image. These objects may be orbiting satellites, asteroids or space debris. If not removed, they can appear in the co-added image. No confirmation co-adder was implemented in producing the co-added ISSA image (§ III.C.4). Therefore, for each ISSA field, all individual HCON intensity images were examined visually to identify objects appearing in only one HCON image. When found, the contaminated portion of the offending scan was identified and removed. Less than 1% of the entire survey database was removed by this visual inspection process (§III.D.3). Even though great care was taken to remove anomalies, some remain in the co-added image. Individual HCON images should be examined to verify the reality of unusual features in the co-added image.
Emission from some solar system material remains in the data and can cause confusion. This includes some asteroids, zodiacal dust bands, comet tails, comet trails and planets. Some asteroids remain in the co-added images since only known asteroids as of the 1986 publication of the IRAS Asteroid and Comet Survey (Matson 1986) were removed.
The zodiacal dust bands are seen in different aspects in different HCONS and will appear as nonconfirming extended emission bands parallel to the ecliptic plane at ecliptic latitudes less than 15°. The images affected by the dust bands are in the ISSA Reject Set.
Both comet tails and trails are seen in the data. Dust associated with comet tails, caused by either charged ionized particles blown by solar wind or neutral particles blown by radiation pressure, appear when the comet is closest to the sun. The tail of comet IRAS-Araki-Alcock is visible in some images near the north ecliptic pole (fields 416 and 418). Comet trails, composed of larger debris insensitive to radiation pressure, spread along the orbit of the comet and accumulate over a long period of time. In the ISSA images comet trails appear as streaks crossing the image nearly perpendicular to the scan direction. A list of ISSA fields affected by known comet trails is found in § III.C.4. Planets are also visible in ISSA fields that cover the lower latitude sky, § III.C.4.
Artifacts resembling tails appear around point sources in the ISSA images. The tails are due to a photon-induced responsivity enhancement, or hysteresis effect, that is a function of source strength and background. Sources brighter than 15 Jy at 12 µm and 20 Jy at 25 µm are expected to have tails. Some sources at less than these thresholds may have tails. Point source tails were not removed from the data. More than one tail may radiate from a single point source in the co-added images due to point sources being scanned in several directions.
At 60 and 100 µm, hysteresis effects remain around bright areas (within ~ 6°) such as the Galactic plane. The effect of the photon-induced brightness around sources will change depending on the scan direction. See href="../../exp.sup/ch4/A.html#8">Main Supplement § IV.A.8 for explanation of photon-induced responsivity enhancement.
Improved solid angle estimates were derived for each detector based on two-dimensional response functions (see Explanatory Supplement to the IRAS Faint Source Survey Version 2 1992, § II.D.2). The improved solid angles differ from those used in making the original image products, e.g., SkyFlux and IRAS Zodiacal History File (ZOHF) Version 2.0. Compared to Main Supplement Table IV.A.1 the average of new effective solid angles for full size detectors increased by 13%, 8% and 6% at 12, 25 and 100 µm, respectively, and decreased by 3% at 60 µm. A change in the solid angles has an inverse effect on calculated intensity values. Therefore, the values in the ZOHF Version 3.1 will be fainter at 12, 25 and 100 µm compared to ZOHF Version 2.0 and slightly brighter at 60 µm.
Several users of the ZOHF Version 2.0 and 3.0 (§ I.F and Appendix H) have found that the descending scans (scans which progress with decreasing ecliptic latitude) are systematically brighter at the ecliptic plane than the ascending scans (scans which progress with increasing ecliptic latitude). In the IRAS orbit, descending scans always look behind the Earth in its orbit while ascending scans always look ahead. The magnitude of the effect is about 2% at 12 and 60 µm, 1.5% at 25 µm, and 4% at 100 µm as seen at the north ecliptic pole between the ascending and descending scans. Analysis as described in Appendix H shows that a large part of the ascending-descending asymmetry can be attributed to uncorrected calibration drifts. However, the possibility that some small part of the asymmetry is a real feature of the sky cannot be ruled out (Dermott et al. 1994).