Section 9. Space Telescope Imaging Spectrograph

Datasets and associations produced by the Space Telescope Imaging Spectrograph (STIS) are calibrated by the various programs that comprise the calstis task. The calibration process has been defined in a series of STIS Instrument Science Reports (ISRs): ISR 95-006 (Baum and Hodge), ISR 95-007 (Hodge and Baum), ISR 96-013 (Baum, Zarate, and Hodge), ISR 96-017 (Hodge, Baum, and Zarate), ISR 96-018 (Baum, Hsu, Hodge, and Ferguson), ISR 96-019 (Hulbert, Hodge, and Baum), and ISR 96-036 (Hulbert, Hodge, and Busko). The reference files required for the STIS calibration pipeline are described in this section. Section 9.1 defines the general file formats and keyword conventions, Section 9.2 describes the calibration reference images, while Section 9.3 describes the calibration reference tables. Section 9.4 documents the current STIS generic image headers for the CCD and MAMA detectors.

9.1 File Formats and Conventions

The STIS calibration reference files (which encompass both reference images and reference tables) are all written as FITS files in either IMAGE or BINTABLE extensions, as appropriate. The FITS primary header unit contains all keywords not specific to individual extensions; these keywords apply to all extensions in the file. Calibration reference files that are images are referred to as calibration reference images and are described in Section 9.2. Reference files that are tables are referred to as calibration reference tables and are described in Section 9.3.

9.1.1 File Naming Convention

The calibration reference file names are constructed from the standard CDBS unique name, an underscore, a three-letter suffix, and the extension .fits. Each reference file name has an identifying suffix (as defined in Sections 9.2 and 9.3) in parentheses after the descriptive file name.

9.1.2 Keyword Conventions

The headers of the reference images follow the generic STIS file format as described in Sections 9.2 and 9.4 and as defined in ICD-19; the headers for the reference tables are given in generic form in Section 9.3. To support the operation of calstis, some additional keywords may be needed: these keywords are defined in Section 9.2. Also, history header records documenting the source(s) of the calibration data at a level that would make it possible to recreate the reference file are required.

The science and reference images have various keywords that describe the detector and world coordinate systems. Two types deserve some explanation. The LTVi and LTMi_j keywords give a linear mapping from a reference pixel coordinate system to the current pixel coordinate system of an image. The reference coordinate system is detector pixels (illuminated portion only) for the CCD, and it is low- resolution detector pixels for the MAMA. In IRAF notation, the first pixel is number one, and the pixel number is an integer at the center of the pixel (e.g., the first pixel runs from 0.5 to 1.5). Generally, the matrix terms LTMi_j specify a rotation and rescaling of the image pixels. For the case of STIS images, there is no rotation, so the cross terms are always zero; and there is no image inversion, so the diagonal terms are always positive. For this case the elements LTMi_i are the reference pixel size in units of the current pixel size. LTVi is the location, in the current coordinate system, of pixel zero of the reference coordinate system. Note that “pixel zero” does not mean the first pixel using zero indexing; it means the location one pixel to the left of (or below) the first pixel in the image.

The situation is straightforward for MAMA data or for unbinned CCD data, and the values of LTVi and LTMi_i are as tabulated below. For STIS data, LTM1_2 and LTM2_1 are zero (no rotation or skew). Reference images can have the following values of LTVi and LTMi_i:

Detector	Bin	Axis Lenghts	LTV1	LTV2	LTM1_1	LTM2_2
MAMA	High	2048 x 2048	-0.5	-0.5
MAMA	Low	1024 x 1024
MAMA	2 x 2	512 x 512	0.25	0.25	0.5	0.5
MAMA	4 x 4	256 x 256	0.375	0.375	0.25	0.25
MAMA	8 x 8	128 x 128	0.4375	0.4375	0.125	0.125
CCD	None	1024 x 1024

The complication for binned CCD observations is that the physical overscan region is 19 pixels, so when binning there is a pixel near each end of the line that consists of part overscan and part illuminated. Such pixels are removed with the overscan when the bias level is subtracted, so the first pixel in the output image does not begin at the beginning of the illuminated portion. This fractional pixel results in some peculiar values of LTVi, which can depend on which amplifier was used for readout. After stripping off the serial and parallel overscans (which will always be the case for reference images), the LTV/LTM keyword values are as tabulated below:

Detector	Bin	Axis Lenghts	LTV1	LTV2	LTM1_1	LTM2_2
LTM2_2	Any	2 x 2	255 x 256	-0.375	0.4375	0.125
CCD	Any	4 x 4	255 x 256	-0.1875	0.4375
CCD	A or C	4 x 4	127 x 128	0.0625		0.5
CCD	B or D	8 x 8	127 x 128		0.5	0.25
CCD	A or C	8 x 8		0.25	0.25	0.25
CCD	B or D	1024 x 1024	-0.25	0.375	0.25	0.125
CCD	None	511 x 512	0.125	0.375	0.125	0.125

9.1.3 Selection Criteria

Calibration files (whether images or tables) are selected based on the value of particular header keywords. For reference tables, calstis will often retrieve information from particular rows within the table, which are selected based on the value found in one or more scalar (or character) columns in that row. The selection criteria for the file and (if relevant) table row are specified in the description of each file in Sections 9.2 and 9.3.

It is often convenient to use special wildcard values in file headers and in table scalar columns in order to minimize redundant or irrelevant information. The convention in character-type fields is to use “ANY” if they apply to any mode, and “N/A” if they are not relevant to that mode. For fields with integer-type fields, the values are -1 and -999, respectively. These conventions are generally noted in keyword comments of Sections 9.2 and 9.3.

Note: It is not permitted to mix the use of specific values and wildcard values in the same table column for otherwise identical selection criteria. For example, in the photometric conversion table (PHT) in Section 9.3.5, rows are selected on OPT_ELEM and (if OBSTYPE = SPECTROSCOPIC) on CENWAVE and SPORDER. If in some row a value of -1 appears in the table column CENWAVE, values different from that are not permitted in this column for any row that has the same value of OPT_ELEM and SPORDER.

9.1.4 Reference Image File Organization

All STIS observation datasets include FITS files that consist of one or more sets of three image extensions. (This format has been documented by Baum and Hodge in STIS ISR 95-006.) These triplets of image extensions are known as imsets. Most reference images will contain one such imset with a similar structure, the exception being the small scale distortion (SSD) image, described in Section 9.2.7. The names of the extensions are specified in the EXTNAME keyword and have the following possible values:

Science Image (SCI): The science image contains the data from one of the detectors (FUVMAMA, NUV-MAMA, or CCD). These may be represented as data numbers (DN) or as detected photons (which are related to DN through the ATODGAIN setting).

Error Array (ERR): The error array contains an estimate of the random statistical error at each pixel. It is expressed as the square-root of the expected variance, based upon the detected photons, the read noise, and a scale-noise that scales linearly with DN. Note that this image may be represented with a null array (i.e., NAXIS = 0 and no pixel array, following STScI conventions) if all the values are identically zero. In most cases the error array for the reference images will have been populated by appropriately combining the error arrays in the constituent calibration exposures. The reference file error array should be combined appropriately with the science error array (usually in quadrature) during the course of calibration.

Data Quality Flags Array (DQ): The data quality flags array provides independent flags for each pixel to indicate which (if any) of 15 possible pathologies may have affected the corresponding pixel in the SCI image. Each flag has a true (set) or false (unset) state and is encoded as a bit in a 16-bit (signed short integer) word. Users are advised that this word should not be interpreted simply as an integer. The meaning of each bit is specified in Section 9.1.5. This image may also be represented with a null array, as with the ERR image.

9.1.5 STIS Data Quality Flags

The data quality flags for STIS science images are described in detail by Baum and Hodge in STIS ISR 95-006 and are listed in the tabulation below. These flags apply to images obtained with all STIS detectors and observation modes alike. These flags will be set and examined during the course of calibration and may likewise be interpreted and used by downstream analysis applications. These flags also apply to the data quality array in extracted 1-D spectra but are not used in calibration reference tables since arrays stored in these tables are generally functional characterizations of some relationship. Note that calibration reference images will, by construction, rarely be affected by certain pathologies (e.g., Reed-Solomon error or pixels compromised by cosmic rays), but the meaning associated with each bit is the same for reference images as it is for science images. Thus, the reference file DQ array is usually ORed with the science array during the course of calibration. Each flag occupies one bit in a 16-bit signed integer word, where bit 0 is the least significant bit:

Bit	Value	Condition
0	1	Reed-Solomon decoding error
1	2	Data replaced by fill value
2	4	Bad detector pixel or beyond aperture
3	8	Data masked by occulting bar
4	16	Hot pixel
5	32	Pixel affected by large flat field blemish
6	64	<Reserved>
7	128	<Reserved>
8	256	Saturated pixel
9	512	Bad pixel in reference file
10	1024	Pixel affected by small flat field blemish
11	2048	Bad pixels in background region
12	4096	Pixel discarded from extraction region
13	8192	Pixel rejected during image combination
14	16384	Pixel not CTI corrected

9.2 Calibration Reference Images

Individual reference files will be uniquely identified by the FILETYPE keyword. Additional keywords required for selection of files are included in the text description and must be inserted into the generic headers.

9.2.1 Bias Image File (BIA): <unique name>_BIA.FITS

The bias image file consists of an image of the additive stationary pattern in the electronic zero point of the CCD; the units are data numbers (DN).

Format: Bias level images are zero exposure time, unbinned, full-frame images. Bias image files are rotated to the Amp A orientation to ensure consistent placement of the overscan regions.

Selection Criteria: Files are selected by CCDAMP, CCDGAIN, CCDOFFST, and BINAXIS1 and BINAXIS2.

Restrictions: This file is only used with DETECTOR = CCD observations.

Required Additional Keywords:

FILETYPE= CCD BIAS IMAGE

9.2.2 Dark Image File (DRK): <unique name>_DRK.FITS

The dark current image file consists of an image of the dark signal (i.e., the signal detected in the absence of photons from the sky).

Format: Each dark image is an unbinned, full-frame image. Units are electrons/s for the CCD and DN/s for the MAMAs.

Selection Criteria: Files are selected by DETECTOR and, in the case of CCD observations, by

CCDAMP and CCDGAIN.

Restrictions: None.

Required Additional Keywords:

FILETYPE= DARK IMAGE

Required Additional Keywords for MAMA Dark Images:

CCDAMP = N/A           CCD amplifier read out (A,B,C,D; N/A)
CCDGAIN = -999         commanded gain of CCD (-999 for N/A)

9.2.3 Pixel-to-Pixel Flat Image File (PFL): < unique name>_PFL.FITS

Description: The pixel-to-pixel flat field image file consists of the relative variations in pixel- topixel sensitivity of the detector. These images contain the wavelength-dependent, high spatial frequency information about the uniformity of the detector response. This image is divided into the science images during the course of calibration.

Format: Each P-flat image is an unbinned, full-frame image.

Selection Criteria: Files are selected on DETECTOR, OPT_ELEM, OBSTYPE, and on CENWAVE if OBSTYPE = SPECTROSCOPIC or APERTURE if OBSTYPE = IMAGING.

Restrictions: None.

Required Additional Keywords:

FILETYPE= PIXEL-TO-PIXEL FLATFIELD IMAGE

9.2.4 Delta Flat Image File (DFL): <unique name>_DFL.FITS

Description: The delta flat field image file consists of an image containing changes to the small scale flat field contained in the PFLTFILE.

Format: Each delta flat is an unbinned, full-frame image. Selection Criteria: Files are selected on DETECTOR.

Restrictions: None.

Required Additional Keywords:

FILETYPE= DELTA FLATFIELD IMAGE

9.2.5 Low-Order Flat Image File (LFL): < unique name>_LFL.FITS

Description: The low-order flat field image file consists of a map of the large scale variations in the sensitivity across the detector. These images contain the wavelength-dependent, low spatial frequency information about the uniformity of the detector.

Format: Each L-flat is a binned, full-frame image.

Selection Criteria: Files are selected on DETECTOR, OPT_ELEM, OBSTYPE, and CENWAVE if OBSTYPE = SPECTROSCOPIC or APERTURE if OBSTYPE = IMAGING.

Restrictions: None.

Required Additional Keywords:

FILETYPE= LOW-ORDER FLATFIELD IMAGE

9.2.6 Shutter Shading Correction Image File (SSC): <unique name>_SSC.FITS

Description: The shutter shading correction file consists of an image with exposure time correction (in seconds) used to correct for shading caused by the CCD shutter during very short exposures. Each pixel in the image gives the additional time that a given pixel was exposed above the nominal exposure.

Format: Each image is an unbinned, full-frame image. Selection Criteria: Files are selected on DETECTOR.

Restrictions: This file is only used with CCD observations.

Required Additional Keywords:

FILETYPE= SHUTTER SHADING CORRECTION IMAGE

9.2.7 Small Scale Distortion Correction Image File (SSD):<unique name>_SSD.FITS

Description: The small scale distortion correction file contains correction deltas to be applied to the detector (A1, A2) positions to correct for small scale detector distortions that are not adequately modeled by the mapping from the detector coordinates system to calibrated wavelength and spatial position.

Format: Each file consists of a pair of full-format images, each stored as an image extension. The first extension contains the deltas along axis 1 for each pixel in the detector and has EXTNAME = ‘AXIS1’; the second extension contains the deltas along axis 2 and has EXTNAME = ‘AXIS2’. Image values specify the displacement in pixels from the rectified image to the distorted image. That is, the coordinates in the rectified image are obtained by subtracting the SSD images (per axis) from the raw science image coordinates.

Selection Criteria: Files are selected on DETECTOR.

Restrictions: None.

Required Additional Keywords:

FILETYPE= SMALL SCALE DISTORTION CORRECTION IMAGE

9.3 Calibration Reference Tables

Information in STIS reference tables is often selected both on keyword values in the header and on values in one or more columns within table rows; the selection criteria will state explicitly the source of the information. In order to support this file/row selection mechanism, the keywords APERTURE (which is a string containing the imaging aperture name) and CENWAVE (which is a 16-bit integer specifying the central wavelength of the spectroscopic setup) will be present in the primary headers of all reference tables. The convention for interpreting these keywords is as follows:

Keyword	Type	OBSTYPE	Wildcard	NULL
CENWAVE	S	SPECTROSCOPIC	-1	-999
APERTURE	CH*8	IMAGING	‘ANY’	‘N/A’

In the above tabulation, “Wildcard” gives the keyword value when it is meant to match all possible values, while “NULL” gives the keyword value when that keyword is not used. The following keyword list describes the generic headers for the calibration reference tables. Other keywords that may appear in the extension header for particular reference tables are described in the subsections that follow.

SIMPLE  =                 T / file conforms to FITS standard?
BITPIX  =                16 / bits per data value
NAXIS   =                 0 / number of data axes
EXTEND  =                 T / FITS extensions present?
FILENAME=                   / disk filename
TELESCOP= HST               / telescope used to obtain data
INSTRUME= STIS              / instrument used to obtain data
DATE    =                   / date this file was written (dd/mm/yy)

                            / DATA DESCRIPTOR KEYWORDS
FILETYPE=                   / type of reference file

                            / CALIBRATION STATUS
USEAFTER=                   / use this file for obs taken on/ after this date
DESCRIP =                   / description of reference data
APERTURE=                   / imaging aperture for which this file is useful
CENWAVE =                   / central wavelength (-1 for any; -999 for N/A)
END

XTENSION= BINTABLE          / binary table extension
BITPIX  =                 8 / bits per data value
NAXIS   =                 2 / number of data axes
NAXIS1  =                   / bytes per row
NAXIS2  =                   / number of rows
PCOUNT  =                 0 / size of special data area
GCOUNT  =                 1 / one data group (required keyword)
TFIELDS =                   / number of table columns
EXTNAME =                   / name of this FITS extension
EXTVER  =                 1 / first extension of this type
INHERIT =                 T / inherit keywords from PHU?
END

9.3.1 Bad Pixel Table (BPX): <unique name>_BPX.FITS

Description: The bad pixel table contains the location of known problematic pixels and the codes indicating the nature of the problem. The pixel locations refer to the full-frame, unbinned image. The data quality codes are 16-bit signed integers with each bit representing a separate problem. The encoding of the data quality flags is given in Section 9.1.4.

Format: Each row in the table lists the detector coordinates of the bad pixel and the data quality code. If a range of pixels along a row or column can be flagged with the same code, they are entered into the bad pixel table with a starting coordinate pair, a length in pixels, and a detector axis to indicate the direction. Pixels are one-indexed, and, in the case of the CCD, the first pixel means the first pixel in the illuminated portion of the CCD.

Selection Criteria: Tables are selected on DETECTOR.

Restrictions: None.

Keywords: The FILETYPE keyword must have the value BAD PIXEL TABLE and the EXTNAME keyword must have the value BPX. The following primary header keyword is also required:

DETECTOR= detector in use: NUV-MAMA, FUV-MAMA, or CCD

Table 9-1 describes the column definitions.

Table 9-1. Bad Pixel Table (BPX), <unique name>_BPX.FITS

Column Name	Data Type	Units	Description
OPT-ELEM	CH*8	—	Grating, prism, or mirror name
PIX1	S	pixel	Starting pixel in the axis 1 direction
PIX2	S	pixel	Starting pixel in the axis 2 direction
LENGTH	S	pixel	Number of pixels to be assigned value
AXIS	S	—	Detector axis: 1 = axis 1, 2 = axis 2
VALUE	S	—	Value to be ORed with data quality array

9.3.2 CCD Parameters Table (CCD): <unique name>_CCD.FITS

Description: The CCD parameters table contains various CCD operating parameters and is indexed on the amplifier used for readout, as well as the commanded gain and bias level.

Format: Each row of the table contains the CCD amplifier, commanded gain, number of unbinned detector pixels in a bin, commanded offset, actual gain, average bias level, read noise, and saturation threshold.

Selection Criteria: The table selection criteria do not depend upon keyword values; rows within the table are selected on CCDAMP, CCDGAIN, CCDOFFST, BINAXIS1, and BINAXIS2.

Restrictions: This table is only used for DETECTOR = CCD observations.

Keywords: The FILETYPE keyword must have the value CCD PARAMETERS TABLE, and the EXTNAME keyword must have the value CCD. The following additional primary header keyword is also required:

DETECTOR= CCD

Table 9-2 describes the column definitions.

Table 9-2. CCD Parameters Table (CCD), <unique name> _CCD.FITS

Column Name	Data Type	Units	Description
CCDAMP	CH*3	—	CD amplifier readout (A, B, C, D, ANY)
CCDGAIN	S	—	Commanded gain of CCD
CCDOFFST	S	—	Commanded bias offset index
BINAXIS1	S	—	Number of unbinned detector pixels (axis 1)
BINAXIS2	S	—	Number of unbinned detector pixels (axis 2)
ATODGAIN	R	electrons/DN	Actual gain of CCD
CCDBIAS	R	—	Actual average bias level
READNSE	R	electrons	CCD read noise
SATURATE	R	DN	CCD saturation threshold
PEDIGREE	CH*67	—	Pedigree of calibration data
DESCRIP	CH*67	—	Description of calibration data

9.3.3 MAMA Linearity Table (LIN): <unique name>_LIN.FITS Description:

Description: The MAMA linearity table contains the global and local count rate limits for the MAMA detectors and the value for the time constant, t, in the nonlinearity correction:

observed\_counts = true\_counts * exp(-t * true\_counts)

The count rate limits are used to assign data quality values to individual pixels or to the entire exposure. The time constant is used to correct for the nonlinearity in counting.

Format: Each row contains the detector name, the global and local count rate limits, the value for the time constant, and the value for the radius in pixels of the region around a flagged pixel that should also be flagged. The reference pedigree and description are also provided per row.

Selection Criteria: The table selection criteria do not depend upon keyword values; rows within the table are selected on DETECTOR.

Restrictions: This table is only used with DETECTOR = FUV-MAMA or NUV-MAMA observations.

Keywords: The FILETYPE keyword must have the value MAMA LINEARITY TABLE, and the EXTNAME keyword must have the value LIN.

Table 9-3 describes the column definitions.

Table 9-3. MAMA Linearity Table (LIN), <unique name> _LIN.FITS

Column Name	Data Type	Units	Description
DETECTOR	CH*10	—	Detector in use: FUV-MAMA or NUV-MAMA
GLOBAL_LIMIT	D	count/s	Count rate resulting in 10% global nonlinearity
LOCAL_LIMIT	D	count/s	Count rate resulting in 10% local nonlinearity
TAU	D	s	Time constant in global nonlinearity expression
EXPAND	R	pixel	Radius for expanding the region flagged as nonlinear
PEDIGREE	CH*67	—	Pedigree of calibration data
DESCRIP	CH*67	—	Description of calibration data

9.3.4 Analog-to-Digital Correction Table (A2D): < unique name>_A2D.FITS

Description: The A-to-D correction table contains corrected data number (DN) values that will replace the input DN values that were assigned by the analog-to-digital conversion during readout. The correction will compensate for any bias in the assignment of certain DN numbers.

Format: Each row consists of the CCD amplifier in use, the commanded gain, a reference keyword, the value of the reference keyword, the number of elements in the array of DN corrections, and the array of DN corrections. The reference pedigree and description are also provided per row.

Selection Criteria: The table selection criteria do not depend upon keyword values; rows within the table are selected on CCDAMP and CCDGAIN.

Restrictions: This table is only used with DETECTOR = CCD observations.

Keywords: The FILETYPE keyword must have the value A-TO-D CORRECTION TABLE, and the EXTNAME keyword must have the value A2D.

Table 9-4 describes the column definitions.

Table 9-4. Analog-to-Digital Correction Table (A2D), <unique name> _A2D.FITS

Column Name	Data Type	Units	Description
CCDAMP	CH*67	—	CCD amplifier readout (A, B, C, D, ANY)
CCDGAIN	I	electron/DN	Commanded gain of CCD
REF_KEY	CH*12	—	Header keyword to use as reference
VALUE	R	—	Value of reference keyword
NELEM	S	—	Number of elements in A-to-D correction array
ATOD	R[65536]	—	Array of A-to-D corrections
PEDIGREE	CH*67	—	Pedigree of calibration data

9.3.5 Photometric Conversion Table (PHT): <unique name>_PHT.FITS

Description: The photometric conversion table contains the throughput of the instrument configuration as a function of wavelength assuming an infinite aperture centered on the detector. The table contains the photometric correction for a point source. Effects such as vignetting and echelle blaze function are folded into this table.

Format: Each row contains the optical element, the central wavelength and spectral order (if a spectroscopic mode observation), the number of array elements, and arrays for the transmissions and corresponding wavelengths. The reference pedigree and description are also provided per row.

Selection Criteria: Tables are selected on DETECTOR and OBSTYPE; rows within the table are selected by OPT_ELEM, and also by CENWAVE and SPORDER when OBSTYPE = SPECTROSCOPIC.

Restrictions: None.

Keywords: The FILETYPE keyword must have the value PHOTOMETRIC CONVERSION TABLE and the EXTNAME keyword must have the value PHT. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA or CCD
OBSTYPE = type of observation: IMAGING or SPECTROSCOPIC

Table 9-5 describes the column definitions. The following columns are relevant for echelle mode observations and are only included in the FUV-MAMA and NUV-MAMA spectroscopic tables: REFORD, REFWAV, REFY, REFMJD, BSHIFT_VS_X, BSHIFT_VS_Y, BSHIFT_VS_T, and BSHIFT_OFFSET.

Table 9-5. Photometric Conversion Table (PHT), <unique name> _PHT.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	C*12	—	Optical element in use
CENWAVE	S	Å	Central wavelength
SPORDER	S	—	Spectral order
NELEM	S	—	Number of data points in throughput array
WAVELENGTH	D[500]	Å	Wavelength array
THROUGHPUT	R[500]	QE	Total system throughput @wavelength
ERROR	R[500]	QE	Error associated with throughput
REFORD	S	—	Reference spectral order
REFWAV	D	—	Wavelength value at pixel 512 in reford
REFY	R	—	Extraction location of pixel 512 in reford
REFMJD	D	—	MJD of the sensitivity curves
BSHIFT_VS_X	D	—	Spatial x component of echelle blaze shift
BSHIFT_VS_Y	D	—	Spatial y component of echelle blaze shift
BSHIFT_VS_T	D	—	Temporal component of echelle blaze shift
BSHIFT_OFFSET	D	—	Temporal component of echelle blaze shift
PEDIGREE	CH*67	—	Pedigree of calibration data
DESCRIP	CH*67	—	Description of calibration data

9.3.6 Aperture Throughput Table (APT): <unique name>_APT.FITS

Description: The aperture throughput table consists of wavelength-dependent transmissions for each aperture with respect to a reference aperture. This table is used in conjunction with PHOTTAB to convert observed counts to absolute flux.

Format: Each row consists of an aperture name, a wavelength and a transmission array, and the array size. The reference pedigree and description are also provided per row.

Selection Criteria: The table selection criteria do not depend upon keyword values; rows within the table are selected on APERTURE.

Restrictions: None.

Keywords: The FILETYPE keyword must have the value APERTURE THROUGHPUT TABLE, and the EXTNAME keyword must have the value APT.

Table 9-6 describes the column definitions.

Table 9-6. Aperture Throughput Table (APT), <unique name> _APT.FITS

Column Name	Data Type	Units	Description
APERTURE	CH*16	—	Aperture name
NELEM	S	—	Number of data points in throughput array
WAVELENGTH	D[65535]	Å	Reference wavelength array
THROUGHPUT	R[65535]	transmission	Total system throughput at each wavelength
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.7 Template Cal Lamp Spectra Table (LMP): < unique name>_LMP.FITS

Description: The template calibration lamp spectra table contains spectra of the calibration lamps obtained at each of the operating currents. These spectra are used to create templates, which in turn are used to determine the wavelength corrections from which automatically generated or user-requested WAVECALs are to be applied to the default dispersion solutions.

Format: Each row of the table consists of the lamp current setting, the number of elements in the arrays, an array of spectrum flux values, and an array of corresponding wavelengths. The wavelength sampling below 3150 Angstrom must accommodate the highest resolution echelle mode (0.01 Angstrom), while the sampling above 3150 can be coarser (0.1 Angstrom or greater).

Selection Criteria: The reference file itself has no selection criteria; however, as noted below it is only used for OBSTYPE = SPECTROSCOPIC observations. Rows within the table are selected by SCLAMP. The reference pedigree and description are also provided per row.

Restrictions: This table is only used for OBSTYPE = SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword must have the value TEMPLATE CAL LAMP SPECTRA TABLE, and the EXTNAME keyword must have the value LMP. The following additional primary header keyword is required for this table:

OBSTYPE = SPECTROSCOPIC

Table 9-7 describes the column definitions.

Table 9-7. Template Cal Lamp Spectra Table (LMP), <unique name> _LMP.FITS

Column Name	Data Type	Units	Description
SCLAMP	CH*12	—	Name of cal lamp that is on
LAMPSET	CH*6	mA	Spectral cal lamp current value
NELEM	I	—	Number of data points in spectrum
WAVELENGTH	D[300000]	Å	Reference wavelength
FLUX	R[300000]	counts	Lamp flux
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.8 Aperture Description Table (APD): <unique name>_APD.FITS

Description: The aperture description table contains the geometric description of the apertures and their offsets (in arcseconds) from a reference aperture.

Format: Each row of the table contains the aperture name, size (length and width) of the aperture, offsets from the center of the aperture to the center of a reference aperture (which is named in the header), offsets from the aperture center to the center of the occulting bars, and the widths of the occulting bars. The reference pedigree and description are also provided per row.

Selection Criteria: The table selection criteria do not depend upon keyword values; rows within the table are selected on APERTURE.

Restrictions: None.

Keywords: The FILETYPE keyword must have the value APERTURE DESCRIPTION TABLE, and the EXTNAME keyword must have the value APD.

Table 9-8 describes the column definitions.

Table 9-8. Aperture Description Table (APD), <unique name> _APD.FITS

Column Name	Data Type	Units	Description
APERTURE	CH*16	—	Aperture name
WIDTH1	R	arcsec	Width along axis 1
WIDTH2	R	arcsec	Width along axis 2
ANGLE	R	degrees	Orientation of long axis of aperture
OFFSET1	R	arcsec	Offset from nominal position in axis 1
OFFSET2	R	arcsec	Offset from nominal position in axis 2
NBARS	S	—	Number of occulting bars
BAR1LOCN	R	arcsec	Location of bar 1 from aperture center along axis 2
BAR2LOCN	R	arcsec	Location of bar 2 from aperture center along axis 2
BAR3LOCN	R	arcsec	Location of bar 3 from aperture center along axis 2
BAR1WIDTH	R	arcsec	Width of occulting bar 1
BAR2WIDTH	R	arcsec	Width of occulting bar 2
BAR3WIDTH	R	arcsec	Width of occulting bar 3
PEDIGREE	CH*67	—	Pedigree
DESCRIP	CH*67	—	Description

9.3.9 Image Distortion Correction Table (IDC): <unique name>_IDC.FITS

Description: The image distortion correction table consists of a set of world coordinate information and fits to a distortion function that are used to construct the rectified, linearized 2-D image.

Format: Each row of the table contains the direction of the mapping (forward or inverse), filter name, and the coefficients to the fits to the distortion functions.

Selection Criteria: Tables are selected on DETECTOR; rows within the table are selected on DIRECTION (“INVERSE” is selected) and FILTER.

Restrictions: This table is only used for OBSTYPE = IMAGING observations.

Keywords: The FILETYPE keyword must have the value IMAGE DISTORTION CORRECTION TABLE, and the EXTNAME keyword must have the value IDC. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA or NUV-MAMA
OBSTYPE = IMAGING

Table 9-9 describes the column definitions. Some columns are included that are not actually used; these were included for consistency with the general IDC table format.

Table 9-9. Image Distortion Correction Table (IDC), <unique name> _IDC.FITS

Column Name	Data Type	Units	Description
DIRECTION	CH*8	—	Mapping direction (FORWARD or INVERSE)
DETCHIP	CH*8	—	(not used)
OPT_ELEM	CH*8	—	Mirror name (not used)
FILTER	CH*8	—	Filter name
WAVELENGTH	D	angstrom	Nominal wavelength (not used)
XSIZE	I	pixel	Size of uncorrected image in X
YSIZE	I	pixel	Size of uncorrected image in Y
XREF	D	pixel	Reference position in X
YREF	D	pixel	Reference position in Y
XCOM	D	pixel	Reference position in X
YCOM	D	pixel	Reference position in Y
V2REF	D	arcsec	V2 corresponding to (XREF,YREF)
V3REF	D	arcsec	V3 corresponding to (XREF,YREF)
THETA	D	degrees	Angle wrt V2/V3 axes (not used)
SCALE	D	arcsec/pixel	Pixel scale for corrected image
CXSIZE	I	pixel	Size of corrected image in X
CYSIZE	I	pixel	Size of corrected image in Y
CXREF	D	pixel	Output pixel corresponding to (XREF,YREF)
CYREF	D	pixel	Output pixel corresponding to (XREF,YREF)
CX00	D	—	Coefficient in X
CX10	D	—	Coefficient in X
CX11	D	—	Coefficient in X
CX20	D	—	Coefficient in X
CX21	D	—	Coefficient in X
CX22	D	—	Coefficient in X
CX30	D	—	Coefficient in X
CX31	D	—	Coefficient in X
CX32	D	—	Coefficient in X
CX33	D	—	Coefficient in X
CX40	D	—	Coefficient in X
CX41	D	—	Coefficient in X
CX42	D	—	Coefficient in X
CX43	D	—	Coefficient in X
CX44	D	—	Coefficient in X
CY00	D	—	Coefficient in Y
CY10	D	—	Coefficient in Y
CY11	D	—	Coefficient in Y
CY20	D	—	Coefficient in Y
CY21	D	—	Coefficient in Y
CY22	D	—	Coefficient in Y
CY30	D	—	Coefficient in Y
CY31	D	—	Coefficient in Y
CY32	D	—	Coefficient in Y
CY33	D	—	Coefficient in Y
CY40	D	—	Coefficient in Y
CY41	D	—	Coefficient in Y
CY42	D	—	Coefficient in Y
CY43	D	—	Coefficient in Y
CY44	D	—	Coefficient in Y

9.3.10 2-D Spectrum Distortion Correction Table (SDC): <unique name>_SDC.FITS

Description: The 2-D spectrum distortion correction table consists of a set of world coordinate information used to construct rectified, linearized 2-D spectra. One corrected image is produced per spectral order.

Format: Each row of the table contains the optical element, central wavelength, spectral order, the nominal axis 2 position of the center of the spectrum, and the WCS information that defines the output rectified spectra. The reference pedigree and description are also provided per row.

Selection Criteria: Tables are selected on DETECTOR; rows within the table are selected on OPT_ELEM, SPORDER, and CENWAVE.

Restrictions: This table is only used for OBSTYPE = SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword must have the value 2-D SPECTRUM DISTORTION CORRECTION TABLE, and the EXTNAME keyword must have the value SDC. The following additional primary header keyword is required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA, or CCD

Table 9-10 describes the column definitions.

Table 9-10. 2-D Spectrum Distortion Correction Table (SDC), <unique name>_SDC.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Spectroscopic element in the grating wheel
CENWAVE	S	Å	Central wavelength
SPORDER	S	—	Spectral order
A2CENTER	R	pixel	Nominal axis 2 coordinate for center of spectrum
NPIX1	S	pixel	Number of axis 1 pixels in rectified image
NPIX2	S	pixel	Number of axis 2 pixels in rectified image
CRPIX1	D	pixel	Axis 1 coordinate at reference pixel in rectified image
CRPIX2	D	pixel	Axis 2 coordinate at reference pixel in rectified image
CRVAL1	D	Å	Axis 1 value at reference pixel in rectified image
CRVAL2	D	arcsec	Axis 2 value at reference pixel in rectified image
CDELT1	D	Å/pixel	Axis 1 pixel spacing in rectified image
CDELT2	D	arcsec/pixel	Axis 2 pixel spacing in rectified image
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.11 Incidence Angle Correction Table (IAC): <unique name>_IAC.FITS

Description: The incidence angle correction table contains coefficients to fits of the change in two dispersion coefficients (the zero-point and the first-order terms) as a function of angular offset from the reference position. These corrections are applied to the default dispersion coefficients.

Format: Each row consists of the optical element employed, the observation type, the central wavelength, spectral order, the number of coefficients, and the two coefficients used to derive the corrections to each of the zero-point and first-order terms in the dispersion relation. The reference pedigree and description are also provided per row.

Selection Criteria: Tables are selected on DETECTOR; rows within the table are selected on OPT_ELEM, SPORDER, and CENWAVE.

Restrictions: This file is only used for OBSTYPE = SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword must have the value INCIDENCE ANGLE CORRECTION TABLE, and the EXTNAME keyword must have the value IAC. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA, or CCD
OBSTYPE = SPECTROSCOPIC

Table 9-11 describes the column definitions.

Table 9-11. Incidence Angle Correction Table (IAC), <unique name> _IAC.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Spectroscopic element in the grating wheel
CENWAVE	S	Å	Central wavelength
SPORDER	S	—	Spectral order
NCOEFF1	S	—	Number of coefficients in IAC solution for first term
COEFF1	D[8]	—	Incidence angle correction coefficients for first term
NCOEFF2	S	—	Number of coefficients in IAC solution for second term
COEFF2	D[8]	—	Incidence angle correction coefficients for second term
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.12 Dispersion Coefficients Table (DSP): < unique name>_DSP.FITS

Description: The dispersion coefficients table consists of the coefficients to the nominal dispersion solution to be applied to extracted 1-D spectra.

Format: Each row consists of grating, central wavelength, spectral order, the reference aperture name, and the coefficients to one of the dispersion functions (gratings or prism). The reference pedigree and description are also provided per row.

Selection Criteria: Tables are selected on DETECTOR; rows within the table are selected by OPT_ELEM, CENWAVE, SPORDER, and A2CENTER.

Restrictions: This table is only used with OBSTYPE = SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword must have the value DISPERSION COEFFICIENTS TABLE, and the EXTNAME keyword must have the value DSP. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA or CCD
OBSTYPE = SPECTROSCOPIC

Table 9-12 describes the column definitions.

Table 9-12. Dispersion Coefficients Table, <unique name> _DSP.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Spectroscopic element in the grating wheel
CENWAVE	S	Å	Central wavelength
SPORDER	S	—	Spectral order number
REF_APER	CH*12	—	Name of reference aperture
A2CENTER	R	pixel	Nominal axis 2 coordinate for center of spectrum
NCOEFF	S	—	Number of coefficients in dispersion solution
COEFF	D[10]	—	Dispersion solution coefficients
MREF	I	—	Reference spectral order number
YREF	D	—	Reference Y location
A4CORR	D	—	Correction to coefficient 4
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.13 1-D Spectrum Trace Table (1DT): < unique name>_1DT.FITS

Description: The 1-D spectrum trace table consists of displacements of spectra along axis 2 for determining the location of a spectrum prior to extracting the 1-D spectrum.

Format: Each row of this table consists of an array of the y-displacements of the center of the spectrum from the nominal center, tabulated by position along the slit (axis 2). The reference pedigree and description are also provided per row.

Selection Criteria: Tables are selected on DETECTOR; rows within the table are selected by OPT_ELEM, CENWAVE, and SPORDER.

Restrictions: This table is only used with OBSTYPE = SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword must have the value 1-D SPECTRUM TRACE TABLE, and the EXTNAME keyword must have the value 1DT. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA, or CCD
OBSTYPE = SPECTROSCOPIC

Table 9-13 describes the column definitions.

Table 9-13. Spectrum Trace Table (1DT), <unique name>_1DT.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Spectroscopic element in the grating wheel
CENWAVE	S	Å	Central wavelength
SPORDER	S	—	Spectral order number
NELEM	S	—	Number of data points in spectrum
A2DISPL	R[1024]	pixel	Displacement along axis 2
A1CENTER	R	pixel	Nominal axis 1 coordinate of center of spectrum
A2CENTER	R	pixel	Nominal axis 2 coordinate of center of spectrum
SNR_THRESH	R	—	S/N ratio threshold for point-source test
MJD	D	day	MJD at zero point for trace rotation
DEGPERYR	D	degree/year	Rate of trace rotation
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.14 1-D Extraction Parameter Table (1DX): <unique name>_1DX.FITS

Description: The 1-D extraction parameter table describes the extraction apertures (or “boxes”)

and methods used in the extraction of 1-D spectra.

Format: Each row of the table describes the size and orientation of the target extraction aperture and the sizes, locations, and orientation of background extraction regions. The orientations of the spectrum and the background boxes are expressed as coefficients to a low-order polynomial that specifies each orientation as a function of position along the order. The orientation angles are with respect to axis 2, in units of degrees. Also included are the order of the polynomial to be used in fitting the background for each wavelength bin, the algorithm to be used in extracting the order, and the maximum search extent for locating the orders (relative to the expected location). Finally, the reference pedigree and description are also included per row.

Selection Criteria: Tables are selected on DETECTOR; rows within the table are selected on APERTURE, OPT_ELEM, CENWAVE, and SPORDER.

Restrictions: This table is only used with OBSTYPE = SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword must have the value 1-D EXTRACTION PARAMETER TABLE, and the EXTNAME keyword must have the value 1DX. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA, or CCD
OBSTYPE = SPECTROSCOPIC

Table 9-14 describes the column definitions.

Table 9-14. 1-D Extraction Table (1DX), <unique name> _1DX.FITS

Column Name	Data Type	Units	Description
APERTURE	CH*16	—	Aperture name
OPT_ELEM	CH*8	—	Spectroscopic element in the grating wheel
CENWAVE	S	Å	Central wavelength
SPORDER	S	—	Spectral order
EXTRSIZE	R	pixel	Height of spectrum extraction box
NCOEFFSL	S	—	Number of coefficients in solution to slit tilt
SLTCOEFF	D[8]	—	Spectrum extraction box tilt with respect to axis 2
BK1SIZE	R	pixel	Height of background extraction box 1
BK2SIZE	R	pixel	Height of background extraction box 2
BK1OFFST	R	pixel	Offset of background extraction box 1 from spectrum
BK2OFFST	R	pixel	Offset of background extraction box 2 from spectrum
NCOEFFBK	S	—	Number of coefficients in solution to background tilt
BKTCOEFF	D[8]	—	Background extraction box tilt with respect to axis 2
BACKORD	S	—	Order of polynomial fit to background
XTRACALG	CH*12	—	Extraction algorithm to use
MAXSRCH	S	pixel	Maximum search extent for order finding
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.15 MAMA Offset Correction Table (MOC): <unique name>_MOC.FITS

Description: The MAMA offset correction table contains coefficients to fits of the change in the dispersion coefficients as a function of commanded offset from the reference position. These corrections are applied to the default dispersion coefficients.

Format: Each row consists of the optical element employed, the central wavelength, spectral order, and the number of coefficients plus two arrays of coefficients that are used to derive the corrections to each of the terms in the dispersion relation. The reference pedigree and description are also provided per row.

Selection Criteria: Tables are selected on DETECTOR; rows within the table are selected on OPT_ELEM, SPORDER, and CENWAVE.

Restrictions: This file is no longer used. It was originally used for MAMA OBSTYPE = SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword must have the value MAMA OFFSET CORRECTION TABLE, and the EXTNAME keyword must have the value MOC. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA, or CCD
OBSTYPE = SPECTROSCOPIC

Table 9-15 describes the column definitions.

Table 9-15. MAMA Offset Correction Table (MOC), <unique name> _MOC.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Spectroscopic element in the grating wheel
CENWAVE	S	Å	Central wavelength
SPORDER	S	—	Spectral order
NCOEFF	S	—	No. coefficients in MOC solution: both terms
COEFF1	D[10]	—	Change to disp coeff due to axis 1 offset
COEFF2	D[10]	—	Change to disp coeff due to axis 2 offset
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.16 Cosmic Ray Rejection Parameters Table (CRR): <unique name>_CRR.FITS

Description: The cosmic ray rejection table contains the parameters used in the cosmic ray rejection algorithm.

Format: Each row of the table contains the value for the number of CRSPLIT images, the mean exposure time of the splits, the method used for the initial guess at the CR-rejected solution, the method for subtracting the sky, the statistical rejection thresholds, the rejection propagation radius and relative threshold, the data quality bit mask that designates bad input pixels, and a switch to control whether the data quality values are set for CR-rejected pixels in the input data.

Selection Criteria: The table selection criteria do not depend upon keyword values; rows within the table are selected on CRSPLIT and MEANEXP.

Restrictions: This table is only used with DETECTOR = CCD observations.

Keywords: The FILETYPE keyword must have the value COSMIC RAY REJECTION TABLE, and the EXTNAME keyword must have the value CRR. The following additional primary header keyword is required for this table:

DETECTOR= CCD

Table 9-16 describes the column definitions.

Table 9-16. Cosmic Ray Rejection Parameters Table (CRR),<uniquename>_CRR.FITS

Column Name	Data Type	Units	Description
CRSPLIT	S	—	Number of images to combine
MEANEXP	R	sec	Mean exposure time
SCALENSE	CH*8	—	Multiplicative scale factor applied to noise
INITGUES	CH*8	—	Initial guess method (minimum or median)
SKYSUB	CH*4	—	Sky subtraction method (“mode” or “none”)
CRSIGMAS	CH*20	—	Statistical rejection thresholds
CRRADIUS	R	pixel	Rejection propagation radius
CRTHRESH	R	—	Rejection propagation threshold
BADINPDQ	S	—	Data quality flag bits to reject
CRMASK	B	—	Flag CR-rejected pixels in input files?

9.3.17 Photometric Correction Table (PCT): < unique name>_PCT.FITS

Description: The purpose of the photometric correction table (PCT) can best be understood in the context of the photometric conversion table (PHT), described in Section 9.3.5. The photometric conversion table (PHT) contains the throughput of the instrument configuration as a function of wavelength assuming a fixed height extraction aperture centered on the spectrum of the point source. This aperture height is currently (as of 23 Oct 97) 11 pixels for the MAMAs and 7 pixels for the CCD. To calibrate diffuse source spectra properly to absolute surface brightness and to allow for optional extraction apertures for point sources, a correction factor must be applied to the standard photometric point source calibration. The calibrated point source fluxes calculated must be divided by the correction factors in the PCT table for either of these two cases.

By definition, the correction factors for the default extraction apertures are unity. The entries in the table that are to be considered as “infinite” height extraction apertures are those apertures with an extraction height (EXTRHEIGHT) equal to MAXHGHT.

THROUGHPUT is tabulated as a function of WAVELENGTH: the particular entries in the table are derived from the values of the cubic spline nodes for a fit to the throughput as a function of wavelength. ERROR is an empirical estimation of the uncertainty associated with the throughput values at each of the nodes.

Format: Each row of the FITS binary table contains the central wavelength, aperture, extraction aperture height, number of array elements, and arrays for the corrections, corresponding wavelengths, and estimated errors for the corrections.

Selection Criteria: Tables are selected on OBSTYPE, DETECTOR, and OPT_ELEM. (Note: Selecting by OPT_ELEM allows for bundling corrections in a compact manner.) Rows within the table are selected by CENWAVE, APERTURE, and EXTRHEIGHT.

Restrictions: This table is only used with OBSTYPE = SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword must have the value PHOTOMETRIC CORRECTION TABLE, and the EXTNAME keyword must have the value PCT. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA, or CCD
OBSTYPE = SPECTROSCOPIC
OPT\_ELEM= optical element in use

Table 9-17 describes the column definitions.

Table 9-17. Photometric Correction Table (PCT), <unique name> _PCT.FITS

Column Name	Data Type	Units	Description
CENWAVE	S	Å	Central wavelength
APERTURE	CH*16	—	Aperture name
EXTRHEIGHT	S	pixel	Height of spectrum extraction box
NELEM	S	—	Number of data points in throughput array
WAVELENGTH	R[500]	Å	Reference wavelength
THROUGHPUT	R[500]	QE	Relative system throughput at each wavelength
ERROR	R[500]	QE	Error associated with throughput
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description

9.3.18 Time Dependent Sensitivity Table (TDS): < unique name>_TDS.FITS

Description: The purpose of the time-dependent sensitivity table (TDS) is to correct the flux calibration for either spectroscopic or imaging data for variations with time and optionally also for temperature.

The sensitivity is determined as a function of time and wavelength, and it is relative to a specified time (that is, it is 1 at the specified time). The sensitivity should be interpolated as a function of wavelength. For each wavelength in the WAVELENGTH array, the sensitivity is intended to be regarded as linear within segments between the times given in the TIME array. The sensitivity is given relative to 1 at the first value in the TIME array and is continuous in time, with the linear segments defined by the values in the SLOPE array. Thus, to determine the sensitivity at a given time (exposure start time), one must start at the first value in the TIME array and concatenate each linear segment until the one is reached that includes the exposure start time.

The correction for temperature (at the same wavelengths as for the time-dependent sensitivity) is included by multiplying by the following factor:

1. + TEMPSENS * (temperature - REFTEMP)

where temperature is the detector temperature in degrees Celsius.

Format: Each row of the FITS binary table contains the grating name, an array of wavelengths, an array of times (MJD), and corresponding 2D array of slope. There is also information regarding a temperature-dependent correction, using columns for the reference temperature and the change in sensitivity per degree of temperature.

Selection Criteria: Tables are selected on DETECTOR. Rows within the table are selected by OPT_ELEM.

Restrictions: None.

Keywords: The FILETYPE keyword must have the value TIME DEPENDENT SENSITIVITY TABLE, and the EXTNAME keyword must have the value TDS. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA, or CCD
FILETYPE= TIME DEPENDENT SENSITIVITY TABLE
OBSTYPE = type of observation: IMAGING or SPECTROSCOPIC

Table 9-18 describes the column definitions.

Table 9-18. Time Dependent Sensitivity Table (TDS), <unique name> _TDS.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*12	—	Grating or mirror name
NWL	S	—	Number of values in WAVELENGTH array
NT	S	—	Number of values in TIME array
WAVELENGTH	D[60]	angstrom	Array of wavelengths
TIME	D[12]	MJD	Array of times (Modified Julian Date)
SLOPE	D[60,12]	% / year	Slope of sensitivity vs time
REFTEMP	D	Celsius	Reference temperature
TEMPSENS	D	/degree	Slope of sensitivity vs temperature
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description

9.3.19 Wavecal Parameters Table (WCP): <unique name>_WCP.FITS

Description: The wavecal parameters table (WCP) contains parameters that control the wavecal processing.

When finding the shift in either the wavelength (dispersion) or spatial direction, it may be desired to reduce the image size by excluding data from the borders. One reason for trimming is that the distortion may not be as well determined near the image borders. Four parameters are included in this table for specifying regions to trim.

WL_TRIM1 and WL_TRIM2 are used when finding the shift in the wavelength direction. WL_TRIM1 pixels are excluded from each end of each image line, and WL_TRIM2 lines are excluded from the bottom and top of the image. Similarly, SP_TRIM1 and SP_TRIM2 are used when finding the shift in the spatial direction. The final digit (1 or 2) in these parameter names indicates the image axis number. None of these parameters can be negative.

This table is to be used by calstis4, which is run on the 2-D rectified output of calstis7, not on raw data. This is significant for two reasons. One is that calstis7 adds a border of about 88 pixels to each edge of the image, so a trim size less than this has no real effect. The second is that when calstis7 processes echelle data, the individual orders are written to separate images, and the height (NAXIS2) of each image will be much smaller than the detector height. (This is the reason for selecting rows depending on OPT_ELEM.) It should be possible to use a fixed trim size for all first order data, and another fixed trim size for all echelle data, for a given detector. This is because wavecals are taken in full format, not subarray, so the image size is predictable. WL_RANGE is the number of pixels over which a cross correlation will be performed between the observed spectrum and the template spectrum. The value must be an odd integer, and the minimum value is 3. WL_RANGE is the full range, not the maximum lag. For example, if WL_RANGE is 5, the lag (offset) for cross correlation runs from -2 to +2.

SP_RANGE is the search range (pixels) used when looking for the ends of a slit or the edges of occulting bars. For a short echelle slit, SP_RANGE is the range over which a cross correlation will be done between the slit illumination pattern and a model of the slit. The value must be an odd integer, and the minimum value is 3. SP_RANGE is the full range, not the amplitude. For example, if SP_RANGE is 5, the search range for an edge could be from 371 through 375, and the lag for cross correlation for finding a short slit would be from -2 to +2.

For the CCD detector, calstis4 has an option to reject cosmic rays. The wavecal parameters table includes four parameters for controlling this option. See ISR STIS 98-12 for a description of the algorithm. When computing the mean and standard deviation along a column, MAD_REJ is a rejection criterion for outliers, and MIN_MAD is the minimum value for the median of the absolute values of deviations (MAD). Both MAD_REJ and MIN_MAD must be greater than zero. MIN_MAD is needed to prevent the MAD from being zero, which would result in rejection of all data, regardless of the value of MAD_REJ. NSIG_CR is the number of standard deviations beyond which a data value will be flagged as a cosmic ray. In an obscured region (behind an occulting bar or beyond the end of the slit), the data value must also be more than NSIG_ILL times the standard deviation from the mean in the illuminated region. This is needed because at this point we don’t accurately know where the obscured regions are (that’s why calstis4 is run, after all), and we don’t want to mistake illuminated pixels for cosmic rays. Both NSIG_CR and NSIG_ILL must be greater than zero.

Format: Each row contains the detector name and grating name, and parameters for processing a wavelength calibration exposure. The reference pedigree and description are also provided per row.

Selection Criteria: The table selection criteria do not depend upon keyword values ; however, as noted below, they are only used for OBSTYPE=SPECTROSCOPIC observations. Rows within the table are selected on DETECTOR and OPT_ELEM.

Restrictions: This table is only used for OBSTYPE=SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword must have the value WAVECAL PARAMETERS TABLE, and the EXTNAME keyword must have the value WCP.

Table 9-19 describes the column definitions.

Table 9-19. Wavecal Parameters Table (WCP), <unique name> _WCP.FITS

Column Name	Data Type	Units	Description
DETECTOR	CH*10	—	Detector in use
OPT_ELEM	CH*10	—	Grating in use
WL_TRIM1	I	pixel	Pixels to trim at left end of first axis
WL_TRIM2	I	pixel	Pixels to trim at right end of first axis
SP_TRIM1	I	pixel	Pixels to trim at left end of second axis
SP_TRIM2	I	pixel	Pixels to trim at right end of second axis
WL_RANGE	I	pixel	Range in wavelength for cross correlation
SP_RANGE	I	pixel	Search range in spatial direction
NSIG_CR	D	—	Nsigma level for CR rejection
NSIG_ILL	D	—	Secondary nsigma rejection level
MAD_REJ	D	—	Number of MAD rejection level
MIN_MAD	D	—	Minimum MAD
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.20 Cross-Disperser Scattering Table (CDS) <unique name>_CDS.FITS

Description: Contains area-normalized scattering profiles for STIS cross-dispersers, as a function of offset measured in low-resolution MAMA pixels. Profiles must have an odd number of pixels with the midpoint of the profile corresponding to an offset of zero pixels. Tables were initially constructed from the STIS IDT reference files named xdisp_*.fits.

Format: Each row corresponds to a particular cross-disperser. Each row contains the optical mode (OPT_ELEM) which is used during row selection, the number of data points (NELEM) in the scattering profile, the scattering profile itself (CDSCAT), PEDIGREE and DESCRIP. The FUV-MAMA and NUV-MAMA scattering functions are segregated into two separate files.

Selection Criteria: Tables are selected based on DETECTOR. Rows within the table are selected based on OPT_ELEM.

Restrictions: This file is only used for OBSTYPE=SPECTROSCOPIC with either DETECTOR=FUV-MAMA or DETECTOR=NUV-MAMA.

Keywords:

FILETYPE= CROSS-DISPERSER SCATTERING TABLE
OBSTYPE = SPECTROSCOPIC
DETECTOR= detector: FUV-MAMA or NUV-MAMA

Table 9-20 describes the column definitions:

Table 9-20. Cross-Disperser Scattering Table (CDS), <unique name> _CDS.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Cross dispersion grating in use
NELEM	I	—	Number of elements in CDSCAT
CDSCAT	R[201]	—	Cross disperser scattering
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.21 Echelle Scattering Table (ECH) <unique name>_ECH.FITS

Description: Contains area-normalized scattering profiles for STIS echelle gratings, as a function of offset measured in low-resolution MAMA pixels. Profiles must have an odd number of pixels with the midpoint of the profile corresponding to an offset of zero pixels.

Tables were initially constructed from the STIS IDT reference files named ech_xdisp_*.fits.

Format: Each row corresponds to a particular echelle grating (OPT_ELEM) and echelle order (SPORDER). Each row contains the columns used during selection (OPT_ELEM and SPORDER), the number of data points (NELEM) in the scattering profile, the scattering profile itself (ECHSCAT), PEDIGREE and DESCRIP. The FUV-MAMA and NUV-MAMA scattering functions are segregated into two separate files.

Selection Criteria: Tables are selected based on DETECTOR. Rows within the table are selected based on OPT_ELEM and SPORDER.

Restrictions: This file is only used for OBSTYPE=SPECTROSCOPIC with either DETECTOR=FUV-MAMA or detector=NUV-MAMA.

Keywords:

FILETYPE= ECHELLE SCATTERING TABLE
OBSTYPE = SPECTROSCOPIC
DETECTOR= detector: FUV-MAMA or NUV-MAMA

Table 9-21 describes the column definitions:

Table 9-21. Echelle Scattering Table (ECH), <unique name> _ECH.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Grating in use
SPORDER	I	—	Spectral order number
NELEM	I	—	Number of elements in ECHSCAT
ECHSCAT	R[1225]	—	Echelle scattering array
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.22 Echelle Cross-Dispersion Scattering Table (EXS) <unique name>_EXS.FITS

Description: Contains area-normalized profiles describing smearing of echelle scattered light in the cross-dispersion direction. This is distinct from scattering by the cross-dispersing grating itself, which precedes the echelle grating in the optical path and is described in the CDS reference file. The physical origin of this component of scattering is not understood. Nonetheless, the scattering profile is given as a function of offset measured in low-resolution MAMA pixels. Profiles must have an odd number of pixels with the midpoint of the profile corresponding to an offset of zero pixels. Tables were initially constructed from the STIS IDT reference files named ech_xdisp_*.fits.

Format: Each row corresponds to a particular cross-disperser and hence a particular echelle grating. Each row contains the optical mode (OPT_ELEM) which is used during row selection, the number of data points (NELEM) in the scattering profile, the scattering profile itself (EXSCAT), PEDIGREE and DESCRIP. The FUV-MAMA and NUV-MAMA scattering functions are segregated into two separate files.

Selection Criteria: Tables are selected based on DETECTOR. Rows within the table are selected based on OPT_ELEM.

Restrictions: This file is only used for OBSTYPE=SPECTROSCOPIC with either DETECTOR=FUV-MAMA or DETECTOR=NUV-MAMA.

Keywords:

FILETYPE= ECHELLE CROSS-DISPERSION SCATTERING TABLE
OBSTYPE = SPECTROSCOPIC
DETECTOR= detector: FUV-MAMA or NUV-MAMA

Table 9-22 describes the column definitions:

Table 9-22. Echelle Cross-Dispersion Scattering Table (EXS), <unique name> _EXS.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Grating in use
NELEM	I	—	Number of elements in EXSCAT
EXSCAT	R[21]	—	Echelle cross dispersion scattering
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.23 Detector Halo Table (HAL) <unique name>_HAL.FITS

Description: Contains volume-normalized detector halo images at one or more reference wavelengths for each echelle grating. Images have a maximum size of 1024 by 1024 MAMA low-resolution pixels. Tables were initially constructed from the STIS IDT reference files named halo*.fits.

Format: Each row corresponds to a particular combination of optical mode (OPT_ELEM) and reference wavelength (HALOWAVE). Each row contains columns used during row selection (OPT_ELEM and HALOWAVE), number of pixels along an edge of the square halo image (HALDIM), the halo image itself (HALO), PEDIGREE and DESCRIP. The FUV-MAMA and NUV-MAMA scattering functions are segregated into two separate files.

Selection Criteria: Tables are selected based on DETECTOR. Rows within the table are selected based on OPT_ELEM and HALOWAVE.

Restrictions: This file is only used for OBSTYPE=SPECTROSCOPIC with either DETECTOR=FUV-MAMA or DETECTOR=NUV-MAMA.

Keywords:

FILETYPE= DETECTOR HALO TABLE
OBSTYPE = SPECTROSCOPIC
DETECTOR= detector: FUV-MAMA or NUV-MAMA

Table 9-23 describes the column definitions:

Table 9-23. Detector Halo Table (HAL), <unique name> _HAL.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Grating in use
HALOWAVE	S	angstrom	Halo reference wavelength
HALDIM	S	—	Length of each axis for HALO
HALO	D[1048576]	—	Cross disperser scattering
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.24 Echelle Ripple Table (RIP) <unique name>_RIP.FITS

Description: Contains peak-normalized ripple (or blaze) functions for STIS echelle gratings, as a function of wavelength, for each allowed echelle tilt (CENWAVE) and order (SPORDER). Tables were initially constructed from the STIS IDT reference files named ripple_*.fits.

Format: Each row corresponds to a unique combination of echelle grating (OPT_ELEM), tilt (CENWAVE), and order (SPORDER). Each row contains columns used during row selection (OPT_ELEM, CENWAVE, SPORDER), the number of wavelength points at which each ripple function is specified (NELEM), the wavelengths (in Angstroms) for each ripple function (WAVELENGTH), the actual ripple functions (RIPPLE), PEDIGREE and DESCRIP. The FUV-MAMA and NUV-MAMA scattering functions are segregated into two separate files.

Selection Criteria: Tables are selected based on DETECTOR. Rows within the table are selected based on OPT_ELEM, CENWAVE, and SPORDER.

Restrictions: This file is only used for OBSTYPE=SPECTROSCOPIC with either DETECTOR=FUV-MAMA or DETECTOR=NUV-MAMA.

Keywords:

FILETYPE= ECHELLE RIPPLE TABLE
OBSTYPE = SPECTROSCOPIC
DETECTOR= detector: FUV-MAMA or NUV-MAMA

Table 9-24 describes the column definitions:

Table 9-24. Echelle Ripple Table (RIP), <unique name> _RIP.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Grating in use
CENWAVE	I	—	Central wavelength
SPORDER	I	—	Spectral order number
NELEM	I	—	Number of elements in ripple array
WAVELENGTH	D[500]	—	Wavelengths corresponding to ripple array
RIPPLE	D[500]	—	Echelle ripple
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.25 Scattering Reference Wavelengths Table (SRW) <unique name>_SRW.FITS

Description: Contains list of halo reference wavelengths (HRWLIST) and PSF reference wavelengths (PRWLIST) to use when correcting for the two-dimensional effects of scattered light. Reference wavelength lists are given as a function of optical element (OPT_ELEM) and grating tilt (CENWAVE). Tables were initially constructed from the STIS IDT reference files named scat_*.

Format: Each row corresponds to a unique combination of optical element (OPT_ELEM) and grating tilt (CENWAVE). Each row contains columns used during selection (OPT_ELEM and CENWAVE), the number of reference wavelengths for the given configuration (NHRW), the list of halo reference wavelengths (HRWLIST) in Angstroms, the list of PSF reference wavelengths (PRWLIST) in Angstroms, PEDIGREE and DESCRIP. The FUV-MAMA and NUV-MAMA lists of reference wavelengths are segregated into two separate files.

Selection Criteria: Tables are selected based on DETECTOR. Rows within the table are selected based on OPT_ELEM and CENWAVE.

Restrictions: This file is only used for OBSTYPE=SPECTROSCOPIC with either DETECTOR=FUV-MAMA or DETECTOR=NUV-MAMA.

Keywords:

FILETYPE= SCATTERING REFERENCE WAVELENGTHS TABLE
OBSTYPE = SPECTROSCOPIC
DETECTOR= detector: FUV-MAMA or NUV-MAMA

Table 9-25 describes the column definitions:

Table 9-25. Scattering Reference Wavelengths Table (SRW), <unique name> _SRW.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Echelle grating in use
CENWAVE	S	—	Central wavelength
NRW	S	—	Valid size of HALOWAVES and PSFWAVES
HALOWAVES	S[10]	—	Halo reference wavelengths
PSFWAVES	S[10]	—	PSF reference wavelengths
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.26 Telescope Point Spread Function Table (TEL) <unique name>_TEL.FITS

Description: Contains volume-normalized telescope PSF images at several reference wavelengths. Images have a maximum size of 196 by 196 pixels. THE PSF PIXELS NEED NOT HAVE THE SAME PLATE SCALE AS STIS MAMA PIXELS! The actual plate scale (in arcsec per PSF pixel) for the PSF image is given in the PSCALE column. Tables were initially constructed from the STIS IDT reference files named stis*.

Format: Each row corresponds to a particular reference wavelength (REFWAVE). Each row contains a reference wavelength (REFWAVE) used during row selection, plate scale (PSCALE) for the PSF image, number of pixels along an edge of the square telescope PSF image (PSFDIM), the telescope PSF image itself (TELEPSF), PEDIGREE and DESCRIP. The FUV-MAMA and NUV-MAMA scattering functions are segregated into two separate files.

Selection Criteria: Tables are selected based on DETECTOR. Rows within the table are selected based on REFWAVE.

Restrictions: This file is only used for OBSTYPE=SPECTROSCOPIC with either DETECTOR=FUV-MAMA or DETECTOR=NUV-MAMA.

Keywords:

FILETYPE= TELESCOPE PSF TABLE
OBSTYPE = SPECTROSCOPIC
DETECTOR= detector: FUV-MAMA or NUV-MAMA

Table 9-26 describes the column definitions:

Table 9-26. Telescope Point Spread Function Table (TEL), <unique name> _TEL.FITS

Column Name	Data Type	Units	Description
PSFWAVE	S	angstrom	Reference wavelength
PSCALE	R	arcsec	Plate scale for PSF pixels
PSFDIM	S	—	Dimension of both TELEPSF axes
TELEPSF	R[38416]	—	Telescope PSF
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.27 Dark Correction Table (TDC) <unique name>_TDC.FITS

Description: The dark subtraction for the NUV-MAMA requires a special, time- and temperature-dependent characterization of the global rate of dark counts. This correction is distinct from the application of the DRK image, which corrects for the spatial distribution of the dark counts. This table contains the parameterization of the function:

Dark\_Rate = SCALE * NORM * exp(-THERMCST / (T\_Dark + 273.155) )

where:

T\_Dark = MAX (OM2CAT, T\_MIN)

and OM2CAT is the MAMA detector temperature, which is obtained from the first extension of the RAW file. Different dark normalizations apply to different epochs, but they are scaled so as to have same mean counts as the applicable pipeline NUV dark (DRK) file. The normalization (NORM), T_MIN, and thermal constant are all time dependant; their values will be determined by linear interpolation of the two temporally nearest entries in the table. Values in the last row will be used for observing epochs later than the largest MJD in the table.

Format: Each row of the table contains: the value for the Modified Julian Day for the epoch of calibration (MJD = JD-2400000.5), the reference DRK image scale factor, the function normalization, the minimum temperature below which the OM2CAT value is not used, and the thermal constant in the exponential of the expression for global dark rate.

Selection Criteria: This table is selected on DETECTOR; rows within the table are selected on the MJDs that most closely bracket the epoch of the observation.

Restrictions: This table is only used for NUV-MAMA observations.

Keywords:

FILETYPE= DARK CORRECTION TABLE
DETECTOR= NUV-MAMA

Table 9-27 describes the column definitions:

Table 9-27. Dark Correction Table (TDC), <unique name> _TDC.FITS

Column Name	Data Type	Units	Description
MJD	D	days	Date to compare with EXPSTART
SCALE	D	—	Overall scale factor
NORM	D	—	Time-dependent normalization factor
T_MIN	D	Celsius	Cutoff temperature
THERMCST	D	Kelvin	Band gap energy
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.3.28 Grating-Aperture Correction Table (GAC): < unique name>_GAC.FITS

Description: The grating-aperture correction table contains a correction factor to the throughput as a function of wavelength for various gratings and apertures.

The STIS spectroscopic flux calibration was set up under the assumption that the grating and aperture throughputs could be determined independently and then simply multiplied together. This turns out not to be true, especially for G430L and G750L modes which contain an extra Lyot stop. This effectively gives grating-dependent aperture throughputs. The correction is applied during absolute flux correction by dividing by the values in the THROUGHPUT column.

Format: Each row consists of grating, central wavelength, aperture, and the wavelength and throughput arrays. There are also pedigree and descrip for each row.

Selection Criteria: Tables are selected on DETECTOR; rows within the table are selected by OPT_ELEM, CENWAVE, and APERTURE.

Restrictions: This table is only used with OBSTYPE = SPECTROSCOPIC observations.

Keywords: The FILETYPE keyword should have the value GRATING-APERTURE CORRECTION TABLE. The following additional primary header keywords are required for this table:

DETECTOR= detector in use: FUV-MAMA, NUV-MAMA or CCD
OBSTYPE = SPECTROSCOPIC

Table 9-28 describes the column definitions.

Table 9-28. Grating-Aperture Correction Table, <unique name> _GAC.FITS

Column Name	Data Type	Units	Description
OPT_ELEM	CH*8	—	Spectroscopic element in the grating wheel
CENWAVE	S	Å	Central wavelength
APERTURE	CH*16	—	Name of aperture
NELEM	S	—	Number of elements in arrays
WAVELENGTH	D[514]	Å	Array of wavelengths
THROUGHPUT	D[514]	—	Array of corrections to throughput
PEDIGREE	CH*67	—	Pedigree of reference data
DESCRIP	CH*67	—	Description of reference data

9.4 Generic STIS Reference Image Headers

Generic reference image headers are given in the following subsections: one for the CCD detectors and one for the MAMA detectors. These headers give the minimal information necessary to define the observing mode to which the files apply.

9.4.1 Generic STIS CCD Reference Image Header

SIMPLE  =                 T / file conforms to FITS convention?
BITPIX  =                16 / bits per data value
NAXIS   =                 0 / number of data axes
EXTEND  =                 T / FITS extensions present?
FILENAME=                   / disk filename
TELESCOP= HST               / telescope used to obtain data
INSTRUME= STIS              / instrument used to obtain data
DATE    =                   / date this file was written (dd/mm/yy)

                            / DATA DESCRIPTOR KEYWORDS
FILETYPE=
                            / SCIENCE INSTRUMENT CONFIGURATION
DETECTOR= CCD               / detector in use: CCD
CCDAMP  =                   / CCD amplifier read out (A,B,C,D)
CCDGAIN =                   / commanded gain of CCD
CCDOFFST=                   / commanded bias offset of CCD
APERTURE=                   / aperture name
OPT_ELEM=                   / optical element used for observation

                            / OBSERVATION DESCRIPTION
OBSTYPE =                   / type of observation: IMAGING or SPECTROSCOPIC
MIRROR  =                   / mirror element used for imaging obs {OPTIONAL}
FILTER  =                   / filter name {OPTIONAL}
CENWAVE =                   / central wavelength {OPTIONAL}

                            / READOUT DEFINITION PARAMETERS
BINAXIS1=                   / axis1 data bin size in unbinned detector pixels
BINAXIS2=                   / axis2 data bin size in unbinned detector pixels

                            / CALIBRATION STATUS
USEAFTER=                   / use this file for obs taken on/after this date
PEDIGREE=                   / pedigree of reference data
DESCRIP =                   / description of reference data
END

XTENSION= IMAGE             / type of extension
BITPIX  =               -32 /
NAXIS   =                 2 / number of data axes
NAXIS1  =                   / size of first image axis
NAXIS2  =                   / size of second image axis
PCOUNT  =                 0 / size of special data area

GCOUNT  =                 1 / one data group (required keyword)
EXTNAME = SCI               / science array
EXTVER  =                 1 /
INHERIT =                 T / Inherit keywords from PHU?

                            / WORLD COORDINATE SYSTEM PARAMETERS
CTYPE1  = PIXEL             / coordinate type for first axis
CTYPE2  = PIXEL             / coordinate type for second axis

SUBARRAY=                 F / is image a subarray (T) or full frame (F)
LTV1    =                   / offset in axis1 to subarray start, detector coords
LTV2    =                   / offset in axis2 to subarray start, detector coords
LTM1_1  =                   / reciprocal of sampling rate, axis 1
LTM2_2  =                   / reciprocal of sampling rate, axis 2
END

XTENSION= IMAGE             / type of extension
BITPIX  =               -32 /
NAXIS   =                 2 / number of data axes
NAXIS1  =                   / size of first image axis
NAXIS2  =                   / size of second image axis
PCOUNT  =                 0 / size of special data area
GCOUNT  =                 1 / one data group (required keyword)
EXTNAME = ERR               / uncertainty array
EXTVER  =                 1 /
INHERIT =                 T / inherit keywords from PHU?

                            / WORLD COORDINATE SYSTEM PARAMETERS
CTYPE1  = PIXEL             / coordinate type for first axis
CTYPE2  = PIXEL             / coordinate type for second axis

SUBARRAY=                 F / is image a subarray (T) or full frame (F)
LTV1    =                   / location of detector pixel zero, for axis1
LTV2    =                   / location of detector pixel zero, for axis2
LTM1_1  =                   / reciprocal of sampling rate, axis1
LTM2_2  =                   / reciprocal of sampling rate, axis2
END

XTENSION= IMAGE             / type of extension
BITPIX  =                16 / bits per data value
NAXIS   =                 2 / number of data axes
NAXIS1  =                   / size of first image axis
NAXIS2  =                   / size of second image axis
PCOUNT  =                 0 / size of special data area
GCOUNT  =                 1 / one data group (required keyword)
EXTNAME = DQ                / data quality flag array
EXTVER  =                 1 /
INHERIT =                 T / inherit keywords from PHU?

                            / WORLD COORDINATE SYSTEM PARAMETERS
CTYPE1  = PIXEL             / coordinate type for first axis
CTYPE2  = PIXEL             / coordinate type for second axis

SUBARRAY=                 F / is image a subarray (T) or full frame (F)
LTV1    =                   / location of detector pixel zero, for axis1
LTV2    =                   / location of detector pixel zero, for axis2
LTM1_1  =                   / reciprocal of sampling rate, axis 1
LTM2_2  =                   / reciprocal of sampling rate, axis 2
END

9.4.2 Generic STIS MAMA Reference Image Header

SIMPLE  =                 T / Fits standard
BITPIX  =                16 / Bits per pixel
NAXIS   =                 0 / Number of axes
EXTEND  =                 T / File may contain extensions
FILENAME= '               ' / Disk filename
TELESCOP= 'HST     '        /
INSTRUME= 'STIS    '        /
DATE    = '                '
                            / DATA DESCRIPTOR KEYWORDS
FILETYPE= '          '      /
                            / SCIENCE INSTRUMENT CONFIGURATION
DETECTOR= 'NUV-MAMA'        / detector in use: FUV-MAMA or NUV-MAMA
CCDAMP  =                   / CCD amplifier read out (A,B,C,D)
CCDGAIN =                   / commanded gain of CCD
APERTURE= 'ANY     '        / aperture name
OPT_ELEM= 'ANY     '        / optical element
                            / OBSERVATION DESCRIPTION
OBSTYPE = 'ANY     '        / type of observation: IMAGING or SPECTROSCOPIC

                            / READOUT DEFINITION PARAMETERS
BINAXIS1=                 1 / axis 1 data bin size in unbinned det. pixels
BINAXIS2=                 1 / axis 2 data bin size in unbinned det. pixels
                            / CALIBRATION STATUS
USEAFTER= '              '  / use after this date
PEDIGREE= '              '  /
END
XTENSION= 'IMAGE   '        / Image extension
BITPIX  =               -32 / Bits per pixel
NAXIS   =                 2 / Number of axes
NAXIS1  =              2048 / Axis length
NAXIS2  =              2048 / Axis length
PCOUNT  =                 0 / No 'random' parameters
GCOUNT  =                 1 / Only one group
EXTNAME = 'SCI     '        / Extension name
EXTVER  =                 1 / Extension version
DATE    = '                '
INHERIT =                 T / Inherit keyword from PHU?
                            / WCS PARAMETERS
CTYPE1  = 'PIXEL   '        / the coordinate type for first axis
CTYPE2  = 'PIXEL   '        / the coordinate type for the second axis

SUBARRAY=                 F / is image a subarray (T) or full frame (F)
LTV1    =         -0.500000 / offset in X to subarray start, detector coords
LTV2    =         -0.500000 / offset in Y to subarray start, detector coords
LTM1_1  =           2.00000 / reciprocal of sampling rate in
LTM2_2  =           2.00000 / reciprocal of sampling rate in
END
XTENSION= 'IMAGE   '        / Image extension
BITPIX  =               -32 / Bits per pixel
NAXIS   =                 2 / Number of axes
NAXIS1  =              2048 / Axis length
NAXIS2  =              2048 / Axis length
PCOUNT  =                 0 / No 'random' parameters
GCOUNT  =                 1 / Only one group
EXTNAME = 'ERR     '        / Extension name
EXTVER  =                 1 / Extension version
DATE    = '               ' /
INHERIT =                 T / Inherit keyword from PHU?
                            / WCS PARAMETERS
CTYPE1  = 'PIXEL   '        / the coordinate type for first axis
CTYPE2  = 'PIXEL   '        / the coordinate type for the second axis

SUBARRAY=                 F / is image a subarray (T) or full frame (F)
LTV1    =         -0.500000 / offset in X to subarray start, detector coords
LTV2    =         -0.500000 / offset in Y to subarray start, detector coords
LTM1_1  =           2.00000 / reciprocal of sampling rate in
LTM2_2  =           2.00000 / reciprocal of sampling rate in
END
XTENSION= 'IMAGE   '        / Image extension
BITPIX  =                16 / Bits per pixel
NAXIS   =                 2 / Number of axes
NAXIS1  =              2048 / Axis length
NAXIS2  =              2048 / Axis length
PCOUNT  =                 0 / No 'random' parameters
GCOUNT  =                 1 / Only one group
EXTNAME = 'DQ      '        / Extension name
EXTVER  =                 1 / Extension version
DATE    = '               ' /
INHERIT =                 T / Inherit keyword from PHU?
                            / WCS PARAMETERS
CTYPE1  = 'PIXEL   '        / the coordinate type for first axis
CTYPE2  = 'PIXEL   '        / the coordinate type for the second axis

SUBARRAY=                 F / is image a subarray (T) or full frame (F)
LTV1    =         -0.500000 / offset in X to subarray start, detector coords
LTV2    =         -0.500000 / offset in Y to subarray start, detector coords
LTM1_1  =           2.00000 / reciprocal of sampling rate in
LTM2_2  =           2.00000 / reciprocal of sampling rate in
END