Overview
FRESCO employs the Maximum Correlation Method (MCM) (H.H. Aumann, J.W. Fowler
and M. Melnyk,
1990,AJ, 99,1674) to construct coadded images. This is the same
algorithm used to make
enhanced-resolution images (HIRES) at IPAC.
FRESCO is a three-step process:
- A program called SNIPSCAN is used to retrieve the raw survey data. The
IRAS raw scan data are known as
CRDD (Calibrated, Reconstructed Detector Data) and the full set of survey CRDD
is stored in a FITS table
format merged with the pointing (boresight) information. The collection of the
raw data in FITS table format
is known as the Level 1 Archive.
- A program called LAUNDR cleans the data
up
- A program called YORIC applies the MCM
algorithm to the LAUNDR'd data.
In FRESCO processing, YORIC does not iterate, but stops after it has built the
``iteration 1" image. At this
point, the image is a simple gridding of the raw scan data into a coadded map,
without any resolution
enhancement.
As of March 31, 1993, FRESCO is AC calibrated. The IRAS detectors exhibited
dwell-time dependent
responsivity changes, especially at 12 µm . This is the IRAS data
feature known as the "AC/DC effect". Previous IRAS image products have been
calibrated on the DC scale,
which is correct for structure > 30'. The AC scale gives flux densities
commensurate with the Point Source
Catalog (PSC2) for point sources.
FRESCO Processing Defaults
The default configuration for FRESCO is:
- Either a 1 degree by 1 degree or a 2 degree by 2 degree field with 15"
pixels
- Baseline removal de-striping
- Data for all four wavelength bands (12, 25, 60 and 100 µm)
processed
- All available survey data coadded
- Only intensity maps produced.
Commonly Used Processing Options
Only commonly used FRESCO processing options are discussed below. Since
FRESCO is essentially the first
iteration of HIRES, the documentation for HIRES may also be of interest.
- Field and Pixel Size and Wavelength
- FRESCO fields can be up to 2 degrees on a side, and need not be square.
Pixels must be
square and are usually 15" or 30". Since the resolution of FRESCO images is
dominated by the size of the
IRAS detectors these pixel sizes are small compared to the achieved resolution.
Fields are normally processed for all four IRAS wave bands but a subset may
be specified.
- Coverage Options
- IRAS surveyed most of the sky with three separate hours-confirming
observations HCONs. HCONs 1 and 2 were concurrent, HCON 3 was
done several months later and
only covers about 2/3 of the sky; because there is generally a larger time
difference between HCON 1 or 2 and
HCON 3, there is a greater difference in the zodiacal background. Because the
IRAS data display flux-dependent non-linear
response variations of the
detectors this can lead to a calibration difference for HCON 3 which can be
significant.
FRESCO can process data from any combination of HCONs. Popular combinations
are 1 and 2 together, 3
alone, which can be useful in examining the effects of the background on
processing, or 1, 2 and 3 all
together.
Output Options
There are several different types of images available. The default
processing produces only intensity
maps, one per band. Coverage and photometric noise maps, described below, may
also be specially requested
for FRESCO.
- Intensity Map
- Intensity in the map is in units of MegaJanskys per steradian.
Intensity maps are named with the
convention "img_name_bx" where "name" is the object name supplied by the
requester, x is the band number.
(Bands 1, 2, 3, and 4 refer to the 12, 25, 60, and 100 µm wavebands,
respectively).
- Coverage Map
- The coverage map gives the sum of the weighted detector responses at
each image pixel over all
detector samples. The actual number of detectors at a given point is roughly
twice the value of the pixel in
the coverage map. Examination of the coverage map can indicate non-uniformities
in detector coverage. The
naming scheme for coverage maps is "cvg_name_bx".
- Photometric Noise Map
- The photometric noise map indicates the internal photometric error of
the detector samples resulting
from the averaging of those that overlap, and is essentially the standard
deviation of contributions to each
pixel. It does not include absolute errors such as calibration errors. These
maps can thus be used to show
the relative noise across an image, but not the absolute level of the
photometric noise. The naming
scheme for these maps is "phn_name_bx".
|