|
|
IRS: Data Products |
|
There are three basic IRS spectroscopic data products. The IRS Raw Data consists of one or more Data Collection Events (DCE's) that have been converted to FITS format. Each IRS DCE consists of a number of reads (samples) of the array at fixed time intervals without a reset between the samples. All IRS spectra will either contain four, eight or 16 reads of the array between resets, and thus the raw FITS data will be a multi-plane data cube with each data plane representing a read of the array. Any ramp duration (exposure time) longer than 16 seconds still results in a sixteen plane raw FITS data product.
Basic Calibrated Data (BCD)The IRS Basic Calibrated Data (BCD) is a FITS file made from each DCE, which has all the "Spitzer-specific" effects removed. For the IRS spectra, the basic calibrated data are either (1) a calibrated, 2-D echellogram for the high-resolution data, or (2) a calibrated, long-slit 2-D spectra for the low-resolution data. The BCD FITS files are accompanied by additional FITS files used in the processing (e.g. the wavelength map, flat field, efficiency correction and traceable error images). Thus, the observer can easily see the entire process used to take the raw data and create the BCD. A single DCE results in a single BCD. An IRS AOR can be made up of many DCE's.
In the Figure, BCD (1) and BCD (2) represent planar images that have been corrected and uncorrected for peak-up stray light or echelle order optical cross talk. Shadowed steps are meant to indicate processes that are performed on the RAW (128 x 128 x N) data cubes, while the other steps are performed on the 128 x 128 planar images. Calibration inputs to the science pipeline are indicated on the left. Post-BCDThe IRS post-BCD product is intended to convey the richness of the target spectrum and allow for a "first look" scientific analysis. For both the high and low resolution IRS data, the post-BCD product will be: (1) a 2-D, spatially-rectified spectrum obtained by coadding all observations obtained at the same position for a given slit, with a linear wavelength scale along the columns and the spatial dimension along the rows, (2) an extraction of all 1-D spectra derived from (1), and (3) a final co-addition of all 1-D spectra for the entire AOR for each slit (regardless of position). The observer is also supplied with a co-added 2-D product. Flux is assigned during the extraction (assembly) process, and is directly related to assumptions about the target (e.g. point source, or extended). The extracted 1-D spectra will be in flux density vs. wavelength units. The 2-D, post-BCD data will contain a header keyword that allows conversion between DN per pixel and flux density per arcsec, per micron. The observer can create the final spectrum for an entire AOR by co-adding the individual post-BCDs. A subset of modules employed by the post-BCD pipeline will be made available to users to be incorporated into their own analysis environments (e.g. IDL or IRAF running on Solaris OS).
Peakup dataThe above discussion applies to IRS spectroscopic data only. Although no pipeline processing is applied, the IRS peak-up data is transferred to the ground and distributed to the observer. The IRS peak-up process uses the Double Correlated Sampling technique, and all data processing is handled on board. Each IRS peak-up produces six, unprocessed DCS (difference) images, as well as two dark-subtracted, cosmic-ray cleaned, flat-fielded and background subtracted "average" frames. These average frames are the ones that are used by the algorithm to find the peak-up target. Therefore the observer selecting an IRS peak-up will receive, in addition to their spectral data, a set of eight peak-up images.
Other Data ProductsAlong with the Raw, BCD, and post-BCD data products, the IRS data in the Spitzer archive will include intermediate data products (e. g. the flat fields, variance images and mask files) used in the calibration and science threads. These will allow observers to immediately see which pixels are processed in the pipeline, and where junctures suitable for re-processing are located. On-the-fly calibration will be performed when needed, i.e., when the calibration in the archive is out-of-date.
Go back to IRS Data Products page
|
help@spitzer.caltech.edu http://ssc.spitzer.caltech.edu/irs/products/overview.html
