Wavelength calibration

After identifying the arc lines you may now start the actual wavelength calibration with e.g.

Midas ...> SET/MOS wlcmet=IC

Midas ...> CALIBRATE/MOS

The first character of gives the way the first lines are identified (Identify, Linear, Fors), the second character gives the mode of fitting (Constant or Variable linelist, Fit all). For a detailed description see subsection L.3.3. This program fits a dispersion relation at first to the row with the identified lines, then to the slitlet, and afterwards to all slitlets (row by row, separately for each slitlet). The fitted line positions are compared to those listed in .tbl (hear). The type of polynomials (Legendre or Chebyshev) is read from (CHEBYSHEV), the fit order from (1), and the tolerance for automatic line identification from ((2)). can be given in pixel (> 0) or wavelength units (< 0). In case you know the central wavelength and the mean dispersion of the grism you used and you determined the offsets of the slitlets with OFFSET/MOS you have to correct the central wavelength for the offset between the center of the CCD and the reference slitlet used by OFFSET/MOS.

Example:

Your reference slitlet has an offset of -100 relative to the center of the CCD in x-direction and you have a mean dispersion of 2Å/pixel and a central wavelength of 5500Å. This central wavelength will always lie at the x-position of the respective slitlet, which is in this case -100pixels (i.e. -200Å) from the center of the CCD. This means that you have a wavelength of 5700Å at the center of the CCD within the reference slitlet. This wavelength should be used as wcenter, since the program assumes that :xoffset = 0 means that the slitlet is at the center of the CCD in x-direction.

Then you can use the following command

Midas ...> SET/MOS wcenter=central_wavelength

Midas ...> SET/MOS avdisp=mean_linear_dispersion

Midas ...> SET/MOS wlcmet=LC

Midas ...> CALIBRATE/MOS

which will read the two parameters from and . If you have FORS data use

Midas ...> SET/MOS wlcmet=RT

Midas ...> CALIBRATE/MOS

This will read the grism number from the header of and then take the parameters stored in the respective keyword. The plot option (N) decides whether or not you get a plot of the residuals and the display option (0) influences the amount of intermediate results displayed on screen. CALIBRATE/MOS stores the results in .tbl (coerbr) in the columns

:SLIT	number of slitlet
:ROW	row for which dispersion relation was fitted (pixel coordinates)
:Y	row for which dispersion relation was fitted (world coordinates)
:RMS	r.m.s. error of fit
:COEF_i	fit coefficients

and adds to .tbl (linpos) the columns

:WAVE	wavelength identification for :X
:WAVEC	fitted wavelength for :X
:RESIDUAL	:WAVE-:WAVEC
:REJECT	mark for rejected lines

Slitlets where no dispersion relation could be fitted are marked with '-1' in column :SLIT. With

Midas ...> RESPLOT/MOS ? slit

you can also get a plot of the residuals for a certain slit (slit = number) that are stored in .tbl (linpos) (or in the first parameter given on the command line).