First in any IDL session using these routines one should execute procedure sumer_init to set up directory references for calibration files.
A set of study-specific procedures has to exist in the working directory:
sumer_item.pro
sumer_id.pro
sumer_px.pro
sumer_cont.pro
A typical reduction session consists of:
1. ftp FITS files. The FITS files that correspond to the
observations should be copied over with ftp in binary mode. It is
recommended to place the files in a directory tree separate from
the working directories. The FITS files should then be linked over to a local
working directory. That way one avoids having many copies of the same
FITS files.
(in Oslo in the achernar/u4/sumer tree
(e.g. /mn/achernar/u4/sumer/fits_oct96).
All belonging to the cds group may create directories in
/mn/achernar/u4/sumer. Make sure to set group to cds (e.g. chgrp cds
fits_nov96)).
2. make an index of FITS files. To find out which FITS files that belong to a given observing programe one may use the procedure sumer_fits_list. First produce a list of file names in a file called index.txt (can be done with the UNIX command ls *.FITS >! index.txt). The output is written to file fits.list. This should be done in the fits-file directory tree.
3. link FITS files. The FITS files that belong to a specific observing sequence are linked to a local working directory. Further reference will be made using date and item number (see below) to distinguish between different data sets.
4. make index of files. Make a list of file names again with the name
index.txt. A procedure
sumer_item.pro has to be written. It is used by the other procedures to
identify the observations. The procedure
sumer_index is then used to make an index of the FITS files with information
on the data structure. The output is written to file
index.tab. If the
same sequence was run several times it may be necessary to manually edit this
file. In the example files above the observing program consisted of
0. raster of region
1. full detector exposure
2. time sequence with fixed slit
3. full detector exposure
4. raster of region
Only items 0, 1 and 2 are recognized and 3 (same as 1) and 4 (same as 0) has to be edited. Note from the example index.tab file that the FITS files may have very different organization for the same kind of data. This is a challenge for the next procedure in the series.
5. concatenate data of the same item. This is done by the procedure sumer_cat. The calling is typically sumer_cat,'960913',0 with date and item_nr as arguments.
6. flat-fielding. This is done by first picking/making a flat-field with the help of the procedure sumer_mkflat. Options 'nearest' and '/shift_obs' are recommended and option '/ff_t_warp' is not. sumer_mkflat has to be run for each date and item_nr separately. When the flat-field has been constructed it is used to flat-field the data with the help of procedure sumer_ff.
7. geometric correction. sumer_geo.
8. wavelength calibration. sumer_idwave. sumer_wavecal. sumer_wave.
9. absolute calibration. sumer_abscal.
10. extraction of total intensities and velocities.
sumer_veli.
Credits
Most of the reduction steps are quite standard except for the flat-fielding
using option '/shift_obs' and the wavelength calibration. Papers based on
option '/shift_obs' should reference
Carlsson, M., Judge, P.G., Wilhelm, K.: 1997,
Astrophys. J. Letters 486, L63-L67
where the procedure is described.
Papers using the wavelength calibration
should reference
Hansteen, V.H., Betta, R., Carlsson, M.: 2000,
Astron. Astrophys. 360, 742-760
Last modified: Sun Apr 19 16:24:14 MET