There is one file in this section:
SWAA SWS Auto-Analysis Results
The AAR files contain primary headers described in Table 9.4.1. In this table example information is used to fill the header.
Most keywords are self explanatory, or have been described above. Those that are not include:
The AAR file contains records with the fields described in Table 9.12. In the FITS file they are defined in the binary header.
SWAALINE and SWAASCNT are both counts of the valid data present in an AAR.
SWAALINE is a count of the valid ranges in a dataset. For AOT 1, it is filled with the AOT band number (see table 3.2). For an AOT 2 observing X lines it will count from 1 to X, with 0 reserved for any datapoints not associated with a line. Note that the datapoints associated with SWAALINE set to, e.g. 1, may not correspond to the first entered line number in your AOT as the logic may re-arrange them to increase efficiency. For AOT 6 it counts the number of scans, for an example see table 9.13. Users should note that there is a difference in the behaviour of SWAALINE for AOT 2's and 6's - for AOT 2 there is one value per each up-down pair, whereas for an AOT 6 there is one value per individual up- or down-scan.
| Operation | line number for detector band | |||
| 1 | 2 | 3 | 4 | |
| REF | 1 | 2 | 3 | 4 |
| SCAN UP | 5 | 6 | 7 | 8 |
| REF | 9 | 10 | 11 | 12 |
| SCAN DOWN | 13 | 14 | 15 | 16 |
| REF | 17 | 18 | 19 | 20 |
| . | . | . | . | . |
SWAASCNT has nothing to do with the number of scans required for an AOT. It is a count of the number of lines/bands containing requested (i.e. what was requested in the AOT) data, starting with 1. Any data outside the requested ranges will have SWAASCNT set to 0. Note that even if SWAASCNT is greater than 0, the flag and status word should be checked to ensure the data is okay. For AOT 1 SWAASCNT is filled with 1.
The decision whether to increment SWAALINE or SWAASCNT depends on whether a scan starts and ends at the same grating position as the preceeding scan. Initially both are set to one. If a scan starts and ends at the same grating position as the preceeding scan the line counter is not incremented but the scan counter is. If the wavelength range is different the line counter is incremented and the scan counter is (re-)set to one.
A slight problem exists in the counting of SWAALINE in that the counter can count up too much, resulting in a higher count than the number of lines. This will be fixed in a later release of OLP.
