INT-DAS-2
FITS Headers for
INT Observations
Guy Rixon, Mike Irwin, Peter Bunclark & James Lewis
Issue 3.3; 1997-09-12
Royal Greenwich Observatory,
Madingley Road,
Cambridge CB3 0HJ
1 Introduction
1.1
Purpose of this documentThis is the Interface-Control Document for the passing of FITS headers from the various INT observing systems to the associated data-export systems and thence to the observers. As such, the document defines the set of descriptors that a `consumer' of the data may expect to find in each FITS file. Once offically issued, this document is an authoritative specification. It is appropriate for authors of observing systems to claim conformance to a specific issue of this document.
1.2
Scope of the interfaceThe data-sets defined are those written to D-tapes and C-tapes and those presented to the observers as disk-FITS files. No mention is made of the way in which the data get into those files and any intermediate files are not covered by this specification. It may be assumed that FITS files are not accessible to the observers, or to software other than observing and logging systems until the files are complete.
1.3
Division of headers into packetsFor convenience of description, each FITS header may be built from separate packets, each packet containing one or more descriptors. Once assembled, the boundaries between packets must be invisible and it must be possible for external software to process the header without knowing about the packets. The packets should appear in the order given, and within each packet, the descriptors must appear in the order specified below.
The names of the packets form part of the interface between the DAsS and other sub-systems. Tasks that write packetsmust produce them with the names given below.
After the last descriptor of the last packet, shall be present a descriptor with the HISTORY keyword marking the end of the part of the header that was written during observation. Any subsequent descriptors are then clearly the result of data reduction.
These are the lists of packets that need to be collected:
WFC: observation wcs telescope autoguider wfc exposure ccd ccdproc
IDS: observation wcs telescope autoguider cagb ids exposure ccd ccdproc
2
Image-size packetThe descriptors listed in this section shall be present at the start of every FITS-file written at the INT. They indicate how to read the image from the file.
Keyword |
Constant value |
Variable type |
Notes |
SIMPLE |
T |
All files are simple FITS. |
|
BITPIX |
16 |
All files contain 16-bit images. |
|
NAXIS |
integer |
Number of axes in the image. |
|
NAXISn |
integer |
Number of pixels in nth axis. n descriptors as appropriate. |
|
BSCALE |
1.0 |
Image data have not been rescaled. |
|
BZERO |
32768.0 |
Image data have the offset required to fit 16-bit unsigned data into a 16-bit signed range. |
The use of these descriptors is strictly defined in the NASA FITS standard, NOST 100-1.0. The order of SIMPLE, BITPIX, NAXIS and NAXISn is part of the standard. The BLOCKED keyword is not used.
This packet has no formal name. The data are provided directly by the DAsS and the packet is not collected in thesame way as other packets.
3
Observation packetThis packet contains data that cannot be derived from any single sub-system of the observing system.
Keyword |
Constant value |
Variable type |
Notes |
RUN |
integer |
The run number. |
|
SYSVER |
character |
Version number of the observing system plus list of software patches. |
|
ORIGIN |
'ING La Palma' |
As used at the WHT. |
|
OBSERVAT |
'LAPALMA' |
Used by IRAF |
|
OBSERVER |
character |
Typically, this contains the observer's initials. |
|
PROPOSAL |
character |
This is the code for the PATT/CAT proposal. |
|
DATEARCH |
character |
Date of observation in format yyyy/mm/dd |
|
OBJECT |
character |
If the observer has entered a title for the observation, it appears here. Otherwise, the name of the current target should be used, or the words ARC, BIAS, FLAT, SKY etc. as appropriate. |
|
OBSTYPE |
character |
This should be one of TARGET, BIAS, FLAT, ARC, SKY, FLAT, DARK, MAP, FOCUS. |
|
IMAGETYPE |
character |
Same as OBSTYPE: needed by IRAF. |
All these descriptors can be built at the start of the observation.
ORIGIN is written to match WHT headers. OBSERVAT, with the precise value shown above, is used by some IRAF programs to select site-specific data.
This packet is named observation.
4
World-coordinate-system packetThis packet defines the registration of the image to celestial coordinates. The form of the packet should be reviewed as the WCS conventions for FITS evolve.
Keyword |
Constant value |
Variable type |
Notes |
CTYPE1 |
character |
Type world coordinates mapped to x axis. Usually 'RA', or 'DEC'. |
|
CRPIX1 |
real |
Reference pixel for world coordinates on x axis. |
|
CRVAL1 |
real |
World-coordinate value at reference pixel, x axis. |
|
CDELT1 |
real |
Interval in world coordinates between pixel centres, x axis. |
|
CROTA1 |
real |
Rotation of x axis from the world-coordinate type stated in CTYPE1. |
|
CTYPE2 |
character |
Type of world coordinates mapped to y axis. Usually 'RA', or 'DEC'. |
|
CRPIX2 |
real |
Reference pixel for world coordinates on y axis. |
|
CRVAL2 |
real |
World-coordinate value at reference pixel, y axis. |
|
CDELT1 |
real |
Interval in world coordinates between pixel centres, y axis. |
|
CROTA1 |
real |
Rotation of y axis from the world-coordinate type stated in CTYPE1. |
The axis-type keywords can be omitted for axes that are not mapped to position on the sky. Where world coordinates are RA or dec., CDELTn and CVALn are given in degrees. CROTAn is always in degrees.
All these descriptors can be built at the start of the observation.
This packet is named wcs.
5
Telescope packet
Keyword |
Constant value |
Variable type |
Notes |
TELESCOPE |
INT |
Name of telescope. |
|
LATITUDE |
'28:45:38.5' |
Latitude of telescope. |
|
LONGITUD |
'17:52:39.5' |
Longitude of telescope. |
|
SLATEL |
'LPO2.5' |
The name for the telescope as understood by the library subroutine sla_OBS(). |
|
CAT-NAME |
character |
The name of the target as stored in the target list. May be different from the name stored under OBJECT. |
|
CAT-RA |
character |
Taken from the target list. In the format hh:mm:ss.ss. |
|
CAT-DEC |
character |
Taken from the target list. In the format sdd:mm:ss.s. |
|
CAT-EQUI |
character |
Equinox of the target's coordinates, taken from the target list. In the format Jnnnn.n, Bnnnn.n or A, where nnnn.n is in years AD. |
|
CAT-EPOC |
real |
Epoch of the target's coordinates, taken from the target list. In years AD as a decimal fraction. |
|
PM-RA |
real |
Proper motion in R.A. In seconds of time per year. |
|
PM-DEC |
real |
Proper motion in dec. In seconds of arc per year. |
|
PARALLAX |
real |
Half-annual parallax in seconds of arc. |
|
RADVEL |
real |
Heliocentric radial velocity of the target in km?s-1. |
|
RATRACK |
real |
Differential-tracking rate in arcsec/second in r.a. |
|
DECTRACK |
real |
Differential-tracking rate in arcsec/second in dec. |
|
RA |
character |
R.A. that the telescope is tracking. In the format hh:mm:ss.ss. |
|
DEC |
character |
Dec. that the telescope is tracking. In the format sdd:mm:ss.s. |
|
EQUINOX |
character |
Equinox of the coordinates in RA, DEC. Ideally, these coordinates should be given to the same equinox as CAT-RA and CAT-DEC. Same format as CAT-EQUI. |
|
STSTART |
character |
Local sidereal time at the start of the observation. |
|
ST |
character |
Same as STSTART |
|
AZIMUTH |
real |
Mean azimuth of telescope during observation. In degrees. |
|
ZD |
real |
Mean zenith distance during observation. In degrees. |
|
AIRMASS |
real |
Mean airmass during observation. |
|
XAPOFF |
real |
Offset from nominal pointing in `x'. In seconds of arc. |
|
YAPOFF |
real |
Offset from nominal pointing in `y'. In seconds of arc. |
|
ROTTRACK |
T |
Always set to true, since an equatorial telescope does not need a tracking rotator. |
|
ROTSKYPA |
real |
Mean position of rotator during observation. In degrees. |
|
PARANGLE |
real |
Parallactic angle. In degrees. |
|
VIGNETTE |
boolean |
True if the mirror petals or the dome vignette the telescope's beam. |
|
DOMEAZ |
real |
Mean azimuth of dome during observation. In degrees. |
|
TELFOCUS |
real |
Position of telescope's focus drive. In metres. The sense of focus movement may vary between foci. |
|
TEMPTUBE |
real |
Temperature of telescope tube in degrees Celsius. |
The quantities defined as `average...during observation' are the mean of samples at regular intervals, starting with the clearing of the CCD for each exposure and ending at the start of CCD readout. It is assumed that the sampling rate is high enough to track non-linear variations in these quantities.
The definitions of CAT-EQUI and EQUINOX conflicts, by the presence of the prefix-letter for the calender system, with NASA's standard for the use of FITS. It may be better to state the calender system under a separate keyword.
This packet is named telescope.
6
Autoguider packet
Keyword |
Constant value |
Variable type |
Notes |
AUTACQT |
real |
Integration time (in seconds) for acquisition. |
|
AUTGUIT |
real |
Integration time (in seconds) for guiding. |
|
AUTWINDO |
integer |
Number of autoguider window. |
|
AUTMODE |
character |
Name of autoguider mode. |
|
AUTXSEE |
real |
Mean FWHM (arcsec) of guide star, x-axis. |
|
AUTYSEE |
real |
Mean FWHM (arcsec) of guide star, y-axis. |
|
AUTXSRMS |
real |
RMS on FWHM (arcsec) of guide star, x-axis. |
|
AUTYSRMS |
real |
RMS on FWHM (arcsec) of guide star, y-axis. |
|
AUTMAGNI |
real |
Mean magnitude of guide star. |
|
AUTTRANS |
real |
RMS variation of magnitude of guide star. |
|
AUTMAGOF |
real |
Magnitude zero-point. |
|
AUTSTATN |
integer |
Number of readings used for statistical calculations. |
|
AUTOMODE |
character |
Mode of autoguiding. |
|
AUTOX |
real |
Guiding position on autoguider, x-axis. |
|
AUTOY |
real |
Guiding position on autoguider, y-axis. |
The mean and RMS value above are calculated from the readings made between the start and end of each exposure,
Some descriptors in this packet cannot be collected until the end of the exposure.
This packet is named autoguider.
7
Exposure packetThis packet records the timing of the exposure. Typically, these data come from the CCD controller, but some observations may have other sources.
Keyword |
Constant value |
Variable type |
Notes |
DATE-OBS |
character |
Date of observation in the FITS-approved format. |
|
UTSTART |
character |
UTC of start of observation. In the format hh:mm:ss.s. |
|
ELAPSED |
real |
Length of exposure, including pauses, from end of clearing to start of readout. In seconds. |
|
DARKTIME |
real |
Same as ELAPSED. |
|
EXPOSED |
real |
Length of exposure, excluding pauses, from the first opening of the shutter to its final closing. In seconds. |
|
EXPTIME |
real |
Same as EXPOSED. |
This packet cannot be completed until the end of the observation.
DATE-OBS is the UT date and changes at 0000hrs UTC. The FITS standard specifies two exact formats for this datum: observation dates before 1st January 1998 shall be in the format dd/mm/yy; later dates shall be in the format ccyy-mm-dd. DATE-OBS and UTSTART must be derived from the same time-stamp. It is an error for UTSTART and DATE-OBS to refer to instants on opposite side of the date-change at midnight.
This packet is named exposure.
8
CCD packetThis packets describes the CCD and how it was read out. Details of how the pixel data were finally stored on disk are in the CCDPROC packet, below.
Keyword |
Constant value |
Variable type |
Notes |
BUNIT |
'ADU' |
Unit of the array of image-data. |
|
GAIN |
real |
Electrons per ADU. |
|
READNOISE |
real |
Readout noise per pixel. In electrons. |
|
CCDSPEED |
character |
Readout speed of CCD. |
|
CCDNCHIP |
integer |
Number of CCDs in this head. |
|
CCDCHIP |
integer |
Number of this CCD within its head. |
|
CCDTYPE |
character |
Name of the this CCD chip (technically, the name of the cryostat such as EEV7). |
|
CCDXSIZE |
integer |
Size in pixels of digitized CCD frame; x-axis. |
|
CCDYSIZE |
integer |
Size in pixels of digitized CCD frame; y-axis. |
|
CCDXIMSI |
integer |
Size in pixels of useful imaging-area; x-axis. |
|
CCDYIMSI |
integer |
Size in pixels of useful imaging-area; y-axis. |
|
CCDXIMST |
integer |
Start pixel of useful imaging-area; x-axis. |
|
CCDYIMST |
integer |
Start pixel of useful imaging-area; y-axis. |
|
CCDWMODE |
boolean |
True if windowing is in force. |
|
CCDWXO1 |
integer |
Offset of window 1, x-axis. |
|
CCDWYO1 |
integer |
Offset of window 1, y-axis. |
|
CCDWXS1 |
integer |
Size in pixels of window 1, x-axis. |
|
CCDWYS1 |
integer |
Size in pixels of window 1, y-axis. |
|
CCDWXO2 |
integer |
Offset of window 2, x-axis. |
|
CCDWYO2 |
integer |
Offset of window 2, y-axis. |
|
CCDWXS2 |
integer |
Size in pixels of window 2, x-axis. |
|
CCDWYS2 |
integer |
Size in pixels of window 2, y-axis. |
|
CCDWXO3 |
integer |
Offset of window 3, x-axis. |
|
CCDWYO3 |
integer |
Offset of window 3, y-axis. |
|
CCDWXS3 |
integer |
Size in pixels of window 3, x-axis. |
|
CCDWYS3 |
integer |
Size in pixels of window 3, y-axis. |
|
CCDWXO4 |
integer |
Offset of window 4, x-axis. |
|
CCDWYO4 |
integer |
Offset of window 4, y-axis. |
|
CCDWXS4 |
integer |
Size in pixels of window 4, x-axis. |
|
CCDWYS4 |
integer |
Size in pixels of window 4, y-axis. |
|
CCDXPIXE |
real |
Size in microns of pixels, x-axis, before binning. |
|
CCDYPIXE |
real |
Size in microns of pixels, y-axis, before binning. |
|
CCDXBIN |
integer |
Binning factor, x-axis. |
|
CCDYBIN |
integer |
Binning factor, y-axis. |
|
CCDSTEMP |
integer |
Required temperature of CCD, in kelvin. |
|
CCDATEMP |
integer |
Actual temperature of CCD in kelvin. |
|
PREFLASH |
real |
Length of preflash in seconds. |
When considering position of windows and imageing-areas on the CCD, the following point should be noted.
z For any CCD, there exists a notional full-frame which is the set of pixels read out when there are no readout windows and the binning factors are both set to one. This full frame typically has more pixels than the physical CCD because it includes bias regions produced as overscan in the preamplifier. Binning and windowing create an actual CCD frame which can be no larger than the full frame, but may, because of overscan, be larger than the physical CCD.
z The descriptors CCDWxxn, which define the readout windows used, refer to regions of the full frame. If the actual frame is both windows and binned, the region read out for each window will be smaller than that described by CCDWxSn. The data-acquisition system arranges the windows to form an actual frame that is the smallest possible rectangle and hence CCDWxxn do not indicate where the windows are in the actual image.
z The descriptors CCDSEC, TRIMSEC, BIASSEC and WINSECn (ccdproc packet, below) refer to regions of the actual frame. They all have the format '[x1:x2,y1:y2]', where x1, x2, y1, y2 are the outermost rows and columns of the region.
z CCDSEC and TRIMSEC describe the region containing all the illuminated pixels. Pixels outside this region are blank or bias level. When multiple windows are used, the `illuminated' region may contain unilluminated pixels in the corners.
z BIASSEC describes the region from which the bias level should be measured.
z WINSECn describe the region containing readout window n.
z In descriptors CCDWxxx, the window start-pixels are expressed as offsets from the edge of the actual frame. A window with CCDWXOx set to zero includes the leftmost column of the frame. A window with CCDWYOx set to zero includes the bottom row of the frame. However, CCDSEC, TRIMSEC, BIASSEC and WINSECn count the first pixel in the frame as (1,1).
This packet can be completed at the start of the observation.
This packet is named ccd.
9
CCDPROC packetThis packet defines how the CCD data are arranged in the FITS file. The name of hte packet reflects the influence of IRAF in defining the descriptors.
Keyword |
Constant value |
Variable type |
Notes |
TRIMSEC |
character |
Region of actual CCD frame containing the image; in `section' format (see below). |
|
CCDSEC |
character |
Same data as TRIMSEC. |
|
BIASSEC |
character |
Region of actual CCD-frame containing the bias level; in `section' format (see below). |
|
WINSECn |
character |
Region of actual CCD-frame containing data from readout window n, n = 1..4. In `section' format (see below). |
|
READAXIS |
character |
The direction of readout w.r.t the axes of the image: `COLUMN' or `LINE'. |
When considering position of windows and imageing-areas on the CCD, the following point should be noted.
z For any CCD, there exists a notional full-frame which is the set of pixels read out when there are no readout windows and the binning factors are both set to one. This full frame typically has more pixels than the physical CCD because it includes bias regions produced as overscan in the preamplifier. Binning and windowing create an actual CCD frame which can be no larger than the full frame, but may, because of overscan, be larger than the physical CCD.
z The descriptors CCDWxxn (ccd packet, above) which define the readout windows used, refer to regions of the full frame. If the actual frame is both windows and binned, the region read out for each window will be smaller than that described by CCDWxSn. The data-acquisition system arranges the windows to form an actual frame that is the smallest possible rectangle and hence CCDWxxn do not indicate where the windows are in the actual image.
z The descriptors CCDSEC, TRIMSEC, BIASSEC and WINSECn refer to regions of the actual frame. They all have the format '[x1:x2,y1:y2]', where x1, x2, y1, y2 are the outermost rows and columns of the region.
z CCDSEC and TRIMSEC describe the region containing all the illuminated pixels. Pixels outside this region are blank or bias level. When multiple windows are used, the `illuminated' region may contain unilluminated pixels in the corners.
z BIASSEC describes the region from which the bias level should be measured.
z WINSECn describe the region containing readout window n.
z In descriptors CCDWxxx, the window start-pixels are expressed as offsets from the edge of the actual frame. A window with CCDWXOx set to zero includes the leftmost column of the frame. A window with CCDWYOx set to zero includes the bottom row of the frame. However, CCDSEC, TRIMSEC, BIASSEC and WINSECn count the first pixel in the frame as (1,1).
This packet can be completed at the start of the observation.
This packet is named ccdproc.
10
Wide-field-camera packet
Keyword |
Constant value |
Variable type |
Notes |
INSTRUME |
WFC |
Name of instrument. |
|
WFCFNAME |
character |
Type of filter e.g. Bkp |
|
WFCFPOS |
integer |
Filter position (1..6). |
|
WFCFID |
character |
Identifier for individual filter. |
|
SECPPIX |
0.37 |
Image scale in arcseconds per pixel. |
These descriptors can be built at the start of the exposure.
The identifier for the filter is assumed to be some kind of serial number or code that uniquely identifies a particular item. Thus, there may be two filters available for which WFCNAME is Bkp but the have different values of WFCID. It is hoped that the unique identifiers can be extened to cover all filters used at ING.
This packet is named wfc.
11
Meteorological packetAt present, meteorological data are not available. Were they available, they would be compiled into this packet:
Keyword |
Constant value |
Variable type |
Notes |
TEMPOUT |
real |
Temperature outside the dome, in degrees C. |
|
TEMPDOME |
real |
Temperature inside the dome in degrees C. |
|
TEMPMIRR |
real |
Temperature of the primary mirror in degrees C. |
|
TEMPTUBE |
real |
Temperature of the telescope tube in degrees C. |
|
DIMM |
real |
Seeing measured on the Differential Image Motion Monitor; in seconds of arc. |
|
WSPEED |
real |
Wind speed in km?h-1 |
|
WDIRECT |
real |
Wind direction in degrees east of north. |
|
HUMIDITY |
real |
Relative humidity as a percentage. |
|
PRESSURE |
real |
Atmospheric pressure in mbar. |
|
DUSTDENS |
real |
Atmospheric extinction in magnitudes. |
|
SKYBRGHT |
real |
Sky brightness in magnitudes per square arcsecond. |
|
METMOLD |
real |
Time in seconds since measurements were obtained from the met. system. |
|
METTOLD |
real |
Time in seconds since measurements were obtained from the TCS |
|
METDOLD |
character |
Time in seconds since measurements were obtained from the DIMM. |
|
METCOLD |
character |
Time in seconds since measurements were obtained from the meridian telescope. |
Meteorological data comes from the central meteorological system, from the TCS (certain temperature readings), from the DIMM tower and from the Carlesberg meridan-telescope. Some of these links may be difficult to maintain during observing, and it is expected that data may be updated manually, probably a few times per night, in files that are always accessible to the observing system. The descriptors METxOLD indicate how long ago these files were last updated.
This packet is named meteorology.
12
IDS packetThis packet covers both IDS and FOS-1. All the descriptors are collected for each instrument even though a few are inapplicable to each.
Keyword |
Constant value |
Data type |
Meaning |
INSTRUME |
IDS |
Name of instrument. |
|
SLITWID |
real |
Slit width in metres. |
|
PSLITWID |
real |
Slit-width projected on detector in metres. |
|
PAOFFSET |
0.0 |
Slit PA — telescope's sky PA in degrees. |
|
SLTSKYPA |
real |
Position angle of slit on sky in degrees. |
|
DEKKERID |
character |
Dekker-slide name. |
|
DEKPOS |
char |
Name of Dekker-slide position. |
|
BSCFILT |
character |
Name of colour filter in the beam below the slit. |
|
BSNDFILT |
character |
Name of ND filter in the beam below the slit. |
|
COLLNAME |
character |
Name of collimator. |
|
COLLFOC |
int |
Collimator focus in encoder units. |
|
HARTMANR |
character |
State of right Hartman-shutter: `OPEN' or `SHUT'. |
|
HARTMANL |
character |
State of left Hartman-shutter: `OPEN' or `SHUT'. |
|
CROSDISP |
character |
State of cross disperser: `IN' or `OUT'. |
|
CLINESMM |
int |
Ruling density of cross disperser in lines per milimetre. |
|
CDBLAZE |
real |
Blaze of cross disperser in Ângstroms. |
|
CAMERA |
character |
Name of spectrograph camera: `235' or `500'. |
|
GRATNAME |
character |
Name of grating. |
|
GLINESMM |
int |
Ruling density of grating in lines per millimetre. |
|
GRATBLAZE |
real |
Blaze of grating nm. |
|
GRATANGL |
real |
Angle of grating in degrees. |
|
GRATSHUT |
character |
State of grating shutter: `OPEN' or `SHUT'. |
|
CENWAVE |
real |
Central wavelength of spectral image in nm. |
|
DISPERSI |
real |
Spectra dispersion in nm of wavelength per mm of length projected on the detector. |
|
FOSMIKER |
real |
Reading of FOS micrometer `R'. |
|
FOSMIKEB |
real |
Reading of FOS micrometer `B'. |
|
FOSMIKES |
real |
Reading of FOS micrometer `S'. |
This packet is named ids.
PAOFFSET is an ING standard. It defines how far the slit position-angle is offset from the telescope's sky PA.
13
Cassegrain-A&G-box packet
Keyword |
Data type |
Meaning |
AGTVPOSX |
real |
TV's x-position in metres. |
AGTVPOSY |
real |
TV's y-position in metres. |
AGTVFILT |
character |
Name of TV filter. |
AGTVSHUT |
character |
State of TV shutter: `OPEN' or `SHUT'. |
AGPOSX |
real |
x -position of autoguider probe in metres. |
AGPOSY |
real |
y- position of autoguider probe in meters. |
ASNDFILT |
character |
Name of ND filter in the beam above the slit. |
ASCFILT |
character |
Name of colour filter in the beam below the slit. |
FLIPMPOS |
character |
Position of `flip' mirror: `IN" or `OUT'. |
AGARCLMP |
character |
Name of arc lamp in use (`NONE' if all lamps are off). |
AGARCND |
character |
Name of ND filter in front of arc lamp. |
COMPMPOS |
character |
Position of comparison mirror: `IN' or `OUT'. |
This packet is named cagb.
Appendix A.
Document HistoryIssue 1.1 22/12/94 First (incomplete) issue.
Issue 1.2 09/01/95 Description of image-size packet added (this changed the section numbering). In the observation packet, details of CTYPEn, CRPIXn, CDELTn, CROTAn altered. In the telescope packet, details of RA, DEC, RADVEL, EQUINOX, ZDSTART, ZDEND, altered; EPOCH added. In the CCD packet, details of CCDWxOn, CCDxPIXE altered; WINSECn added; notes defining CCDSEC, TRIMSEC, etc. changed. In the meterological packet, details changed for DIMM, PRESSURE, DUSTDENS, SKYBRGHT; METxOLD added.
Issue 2.1 20/06/96 Major changes after review by P. Bunclark, J. Lewis and G. Rixon. Marker introduced for end of ING-supplied header. BZERO corrected to floating point. FITSVER changed to SYSVER. FSTATION, DISPAXIS deleted and DATEARCH introduced. WCS packet split off from observation packet. LATITUDE, LONGITUD and SLATEL added. Corrections to CAT-EQUI, CAT-EPOC and TELFOCUS. UTOBS, EPOCH, ZDSTART, ZDEND, AMSTART, AMEND, MNTPATA, MNTPAEND, PAOFFSET, DAAZSTART and DAZEND all deleted. ZD and DOMEAZ added. Definition of AIRMASS changed. CCD packet split into Exposure, CCD and CCDPROC packets. WFC packet completely re-written. TEMPTUBE moved to the telescope packet.PFC packet (for the old PFCU) deleted.
Issue 3.1 1996-10-02 Descriptions of the IDS and CAGB packets were added. There is no longer a separate packet for FOS-1. The definition of the interface is now complete.
Issue 3.2 1997-03-24 The formal names are defined for all packets. SECPPIX was added to the wfc packet. The descriptors XXXSEC were removed from the packet ccd (they are now in ccdproc).
Issue 3.3 1997-09-12 The specification was updated following review of an IDS header from system s4.1a.
S
TELESCOPE, INSTRUME and DETECTOR were moved from observation to telescope, ids/wfc and ccd packets respectively.S
DATE-OBS was moved from observation to exposure and its formatting rules were extended to the new FITS convention.S
RUN was moved to the top of observation to match established practice.S
PREFLASH was moved from exposure to ccd.S
READAXI was deleted from ccd.S
READAXI in ccdproc was renamed READAXIS.S
SLTSKYPA was deleted from ids and PAOFFSET was added.S
DEKPOS is now the name of the dekker position, not the encoder value (this recognizes established practice).S
CAMFOCUS is deleted from ids. The IDS has no focus drive other than the collimator drive.S
GRATBLAZE and CENWAVE are changed from Ângstroms to nm to conform with IAU and FITS standards.S
DISPERSI was added to ids.S
ASCOLFILT was corrected to ASCFILT in ids.