next previous contents



The concept of the IUE header label has been discussed in several sections of this document (see, for example, Sections 2.1 and 8.2.1). The label generated at the time of image acquisition is passed through to the processing system along with the image data and is appended in sequential fashion to record significant processing parameters.

The discussion of Section 8.2.1 focussed on the physical and logical record format of image labels on tape, without regard to their actual contents. Similarly, in other sections of this document, the kinds of data added to the processing history portion of the label were discussed generally, without regard to specific formats within the label records. In this section, the contents of image labels are presented and explained so that users may more usefully interpret the documentary information in the labels.

Throughout this section, the illustrative displays of labels and portions thereof are in the form in which they would appear on the labelprint listings described in Section 8.1.4. In particular, the character decoding is EBCDIC, so that binary-format label entries are not decoded properly. Since the binary portions of the label are, with the exception of the "camera snapshot" entries pertaining to camera temperature data, not used by the processing system, little documentation of the binary data is presented herein. Users are referred to Thomason and Cheng (1978) for the original format specifications for the science header.


The appearance of a typical raw-image label obtained from a labelprint listing of a GO tape is illustrated in Figure 9-1. The basic 100-logical-record format is readily apparent, as is the short image processing history portion appended as the file was written to tape. As will be apparent in Section 9.3 , this image processing history expands significantly for subsequently derived files (processed data). Table 9-1 contains an explanatory key to the most significant individual fields in the raw-image label.

Figure 9-1a: Labelprint Listing for Raw Image File (RI), Part 1.

Figure 9-1b: Labelprint Listing for Raw Image File (RI), Part 2.

file. Fields 1-4 collectively comprise the "size parameters" for the data file.
Table 9-1. Key to Figure 9-1
1Starting line (record no.) of data file(bytes 25 - 28)
2 Starting sample (byte no.) of data file (bytes 29 - 32)
3 Number of lines (records) in data file (bytes 33 - 36)
4 Number of samples (bytes per record) in data (bytes 37 - 40)
5 Camera scan step size (1-4) (byte 44)
6 EDS file no. (1 or 2) (byte 46)
7 Station flag (0 = HANDOVER, 1 = GSFC, 2 = VILSPA) (byte 49)
8 Camera no. (1 = LWP, 2 = LWR, 3 = SWP, 4 = SWR, 8 = FES1, 9 = FES2) (byte 50)
9 Dispersion flag (0 = high, 1 = low) (byte 51)
10 Image sequence no. (1-99999) (bytes 52 - 56)
11 Running number of label line (bytes 67 - 69)
12 Continuation character (C = more lines follow, L = this is last line of label) (byte 72)
13 SOC tape (raw image archive tape) no.
14 File no. of raw image on SOC tape
15 Total time camera was turned on (seconds), taken from FIN entry (see field no. 24). Sum of all exposures if more than one is taken. This is not a true exposure time for trailed spectra.
16 Guest Observer comments section, entered by telescope operator
17 Event "round-robin" section describing time- tagged sequence of procedures. Entries all begin with GMT time in hhmmss format. Oldest entries appear below the double blank lines. Note: SWLA = short wavelength large aperture, LWSA = long wavelength small aperture, etc.
18 Year of read
19 GMT day of read
20 Approximate time of read in hours, minutes, and seconds GMT
21 Exposure start tag. GMT time given is near start of exposure. Format is:
EXPOBC cam. no. Tmintsec gain lamps
where Tmintsec is the commanded duration of "camera-on" time (seconds are rounded). Actual duration may be modified by a subsequent MODTIME command.
22 Exposure end tag. GMT time given is usually near end of exposure but can be much later. Format is:
FIN cam. no. T t S sec voltage U uvc voltage,
where t is the total cumulative duration of "camera-on" time achieved in seconds (truncated) since last read of camera in question. Due to truncation, duration may not agree with that in EXPOBC tag.
23 Exposure start tag, in this case for the second aperture.
24 Exposure end tag, in this case following exposure in second aperture and showing truncated cumulative time for both apertures. Time in seconds here is passed to field no. 15.
25 Readprep tag. GMT time given is near start of image read process. Format is: READPREP cam. no. IMAGE image sequence no.
26 Program ID
27 Episode no. (1, 2, 3,. . . etc.)
28 Observer sign on name
29 Target list sequence no.
30 Catalog source (H, B, D,. . . etc.)
31 Object name
32 Object classification
33 Right ascension of object (hhmmsst where t is tenths of seconds of time)
34 Declination of object (± ddmmss of arc)
35 Spectral type
36 Luminosity class (1-9)
37 V magnitude or flux
38 Color excess E(B-V) or color B-V
39 Information from Preplanned Observation Tape (POT)
40 Binary section of label
41 Binary section of label
42 Orbital elements, periodically updated, for epoch specified by fields 43 and 44
43 Julian Date
44 Seconds since midnight of JD in field 43
45 a, semimajor axis (km)
46 e, eccentricity
47 i, inclination (deg)
48 OMEGA , longitude of ascending node (deg)
49 omega , argument of pericenter (deg)
50 M, mean anomaly (deg)
51 Spacecraft attitude commands sent to spacecraft (most recent set of four)
52 Day:hour:min:sec at which new attitude commanded
53 Right ascension commanded (hhmmsst where t is tenths of seconds of time)
54 Declination commanded (± ddmmss of arc)
55 Spacecraft roll angle (dddmmss of arc)
56 Binary section of label
57 Image processing history section of label
58 File type identifier and image processing scheme name (see Figure 9-2 and Table 9-2)
59 Identifier for image processing program name and time (GMT) of image processing scheme initiation

Note that lines 3-9 are entered by the telescope operator at the console and may occasionally contain errors. Lines 36-37, normally input from the POT, may be modified by the telescope operator and hence are also subject to errors. The automatic entries on the other lines are more accurate but can be affected, for instance, by ground computer problems (see also Section 9.5.1).

As is apparent from Figure 9-1, the binary-format portion of the raw-image label in logical records 38-82 and 86-100 is not usefully decoded when interpreted in EBCDIC characters. The camera snapshot entries in logical records 86-100 are scanned by the processing system as described in Turnrose, Harvel, and Mallama (1982) in order to extract information on the camera head- amplifier temperature (THDA) which is in turn used to correct geometric distortion and dispersion-constant files for thermal effects (Sections 4 and 6). Users are referred to Turnrose, Harvel, and Mallama (1982) for details.


As images proceed through the various reduction steps performed by IUESIPS, records are appended to the raw image label to document the parameters describing the processing. This is a cumulative process so that the label of the end product (extracted spectral data) contains all entries added at earlier stages. In the following sections, examples of only the final low and high dispersion merged extracted spectra (MELO, MEHI) labels are discussed.


A labelprint listing of an SWP merged low dispersion extracted spectral (MELO) file is reproduced in Figure 9-2. Note that, as pointed out in Section 8.1.4, logical records 6-100 are suppressed in such listings; none of the image- processing history records are suppressed, however. Table 9-2 contains an explanatory key to the fields marked in Figure 9-2.

Information identifying lines in the raw image affected by microphonic noise (see Section 3.1 ) is currently provided for LWR images only; an example is given in Section 9.3.2.

Figure 9-2: Labelprint Listing for Merged Low Dispersion Extracted Spectral File (MELO)

Table 9-2. Key to Figure 9-2
Field Contents
1 Number of records in data file
2 Number of bytes per record in data file
3 Image processing history section of label
4 Image processing scheme name, in format shown below.

		(1)   (2)   (3)   (4)   (5)   (6)
		 T     n     H     L     A     C
			     L     S     M     S
      (1)       T = new software (F formerly used for old software)
      (2)       n = camera number (1-4)
      (3)       H = high dispersion
		L = low dispersion
      (4)       L = large aperture point source
		S = small aperture point source
		B = both apertures point source
		E = large aperture extended source
		T = large aperture trailed source
		X = both apertures extended source
		Y = both apertures trailed source
      (5)       A = automatic registration of spectral order(s)
		M = manual registration of spectral order(s)
      (6)       C = current production processing calibration
		S = special calibration

Note: not all combinations are possible.
5 Data pertinent to ITF application (see Section 5)
6 Effective ITF exposure times in units of 0.01 seconds
7 ITF MULT values
8 ITF FACTOR value (rounded)
9 Identifier for image processing program name and time
10 Descriptor identifying aperture from which spectrum was extracted
11 Identifier for image processing program name and time
12 GMT "midpoint" of observation (see Section 6.4.1)
13 Target coordinates extracted from line 37 of raw-image label (fields 33 and 34 in Figure 9-1)
14 Height (length) of gross extraction slit in units of SQRT(2) pixels (see Section 7)
15 Height of background extraction slit in units of show SQRT(2) pixels (see Section 7)
16 Distance from dispersion line to center of background slit in units of SQRT(2) pixels (see Section 7)
17 Value of "OMEGA" angle in low dispersion extraction (angle of the line of constant wavelength relative to dispersion direction) (see Section 7)
18 Identification of mean reseau data set (see text)
19 THDA temperature used for reseau correction
20 Space for message if no reseau temperature correction was applied (see Figure 9-3).
21 Identification of mean dispersion constant set (see text)
22 THDA temperature used for dispersion-constant correction
23 Values of time/temperature computed shifts applied to zero point of dispersion relations
24 Values of residual shifts applied (either manually or automatically) to zero point of dispersion relations to register spectrum
25 Array of line-direction dispersion-constant values (after all shifts) used to extract spectrum
26 Array of sample-direction dispersion-constant values (after all shifts) used to extract spectrum
27 Identifier for image processing program name and time
28 File-type identifier message
29 Identifier for image-processing program name and time

The information pertaining to the reseau displacements and thermal corrections thereof is documented in two separate lines of the label (fields 18-20). In this example, the label data state the following: the mean reseau displacements used were calculated from images acquired between GMT day 085 of 1978 and day 334 of 1979 (note that the entries in the label are not in decimal-year notation), 18 flat-field images from this time period were used to calculate the mean, and the average 1 sigma scatter values among these images in the sample and line directions are 0.134 and 0.138 pixels, respectively.

The camera head amplifier temperature (THDA) at the time of image read obtained from the camera snapshot data (see Turnrose, Harvel and Mallama, 1982) is documented in field 19. This temperature is used as described in Section 4 to calculate thermally-corrected displacements, as appropriate. Should a specific THDA value ever be manually entered into the processing scheme to override camera snapshot data, the words "MANUAL OVERRIDE" would be added to this label record. Furthermore, when mean reseau displacements rather than temperature-corrected displacements are used, the message "MEAN RESEAU USED" would appear in field 20 (see Figure 9-3 for such a message). Such a situation will currently always occur for LWR or LWP data and will also occur for SWP data for which a valid THDA value is not available from either the camera snapshots or manual input.

The information pertaining to the dispersion relations and thermal/temporal corrections is documented in a number of lines of the label (fields 21-26). In this example, the label data indicate that the mean dispersion relations used were calculated from images acquired between GMT day 001 of 1980 and day 222 of 1982 (again this is not a decimal-year notation); that 44 wavelength calibration images from this time period were used to calculate the mean; and that the average 1s residual scatter values after correction for time and temperature effects in the sample and line directions are 0.287 and 0.252 pixels, respectively.

The camera head amplifier temperature (THDA) at the time of the end of the exposure, as obtained from the camera snapshot data, is documented in field 22 and used as described in Section 6 to calculate the thermal shift of the dispersion relation. If the THDA at the end of exposure is not available from the camera snapshots, the THDA at the time of image read is used and the words "AT TIME OF READ" would be added to the label record. In the event that a specific THDA value is manually entered into the processing scheme to override camera snapshot data, the words "MANUAL OVERRIDE" would be added to this label record. If the THDA is not available from the header label or manual input but the time of read is determined (either specified manually or extracted from the label) then a correction for time only is applied and the message "MEAN THDA" will appear in field 22 signifying a default to the mean THDA value. If neither the THDA nor the time of read is available from either the header label or manual input, the message "MEAN DC USED" is added to indicate that the processing defaulted to mean dispersion constants. Note that even in such an instance, the sigma values given in the label are still the average residuals described above.

The information labeled "THERMAL SHIFTS" in field 23 documents the net time/temperature shifts applied to the mean dispersion constants (see Section 6). The shifts labeled "REGISTRATION SHIFTS" in field 24 represent the residual shifts needed after the "THERMAL SHIFTS" listed above are applied to register the extraction slit with the order (see Section 6). For registration shifts determined manually (i.e., by operator intervention) the word "MANUAL" appears after the shift values; for registration shifts determined automatically by the software system, the word "AUTO" appears instead, as in this example.

The final dispersion constants (after all shifts have been applied) actually used in the spectral extraction step are recorded in the B (line direction) and A (sample direction) arrays in fields 25 -26.


A labelprint listing of an LWR merged high dispersion extracted spectral (MEHI) file is reproduced in Figure 9-3, and an explanatory key is provided in Table 9-3. The majority of the fields in the image-processing history are the same as described for the low-dispersion case in Section 9.3.1, and so are not described again here. Note that the label of course reflects processing by the high dispersion programs SPECHI, SORTHI, and POSTHI instead of the low dispersion programs SPECLO and POSTLO. Also, in this example, the presence of microphonics is recorded (fields 1 and 2).

Figure 9-3: Labelprint Listing for Merged High Dispersion Extracted Spectral File (MEHI)

Table 9-3. Key to Figure 9-3
Field Contents
1 Identifier for image processing program name and time
2 Microphonics presence indicator; listed are lines of raw image affected by detected microphonics.
3 Ripple correction K value, evaluated by the expression given, where M = echelle order number
4 Ripple correction alpha value
5 Satellite velocity components
6 Earth velocity components
7 Net radial velocity correction which was applied to extracted wavelengths (km s-1)

The fields which are unique to high dispersion processing are those relating to the echelle ripple correction and to the velocity corrections made to reduce the wavelengths to a heliocentric frame of reference. The ripple correction parameters, applied as discussed in Section 6 , are given in fields 3 and 4.

The velocity components of the IUE satellite and the earth in a right-handed equatorial coordinate system, calculated as described in Section 6 , are given in fields 5 - 6. The resulting net radial velocity correction, whose wavelength equivalent is added to the extracted wavelengths in order to obtain the heliocentric wavelengths, is given in field 7; see Section 6 for computational details.



The single-record label associated with the Tape Header file described in Section is illustrated in the labelprint listing of Figure 9-4. Note that all information in the label is in EBCDIC format, and no size-parameter field is present.


The label associated with a file of reseau positions is illustrated by the labelprint listing of Figure 9-5 . The identifying information in bytes 41-72 of line 1 of the label is the same as that in the label of the image from which the reseau positions were derived.

Figure 9-4: Labelprint Listing for Tape Header File

Figure 9-5: Labelprint Listing of Reseau-Position File


9.5.1 LABEL LINES 1-100 Errors

There are occasions when the information contained in lines 1-100 of the image header label is either in error, unreadable or missing altogether. Although unusual, these problems can affect the processing of spectral data by IUESIPS, particularly since the implementation of the new software which relies more heavily than the earlier software on the data stored in the image header label. The cases in which these errors are known to occur are described below. Modifications

The dependence of the new software on information extracted from the image header label requires that IUESIPS have the ability to correct known label errors. For this reason, software was written to allow the image processing specialist to correct label entries pertaining to the camera number, image sequence number, program ID, exposure time, right ascension, declination, and date of observation. The applications program which allows the modification of these label entries, LABFIX, was first used in December 1981. All other label entries, such as the binary label data and the event status portion of the label were left uncorrected, primarily because of the difficulty in correcting these entries and in knowing the correct data to enter.

It was subsequently decided that a better policy would be to leave the original header intact and deal with corrections in an alternative way. The project therefore agreed that the IUESIPS software should eventually be modified to search for an appendage to the header label which would contain any corrections to data stored in lines 2 - 100. Line 1 would still be allowed to be modified directly, because of the importance of entries in this line as identifiers and keys used by numerous software systems. Because such an alternative procedure would require changes to almost every major IUESIPS applications program, an interim policy was agreed to in which although the label entries mentioned above would continue to be corrected in situ, the original (uncorrected) entries would also be stored in an appendage to the label. The applications program LABCOR which creates such an appendage was first used in February 1983.


The documentation and correction of errors discovered in the image-processing history portion of the IUE image header label is described in detail in Turnrose and Harvel (1982) and Turnrose, Thompson and Gass (1984). A summary of all changes to the image processing history portion of the label is contained in Thompson (1984d).

next previous contents