A star known as 10 Lacertae, or 10 Lac is a particularly good star to use for wavelength calibration. It is bright (magnitude 4.8) and hot (spectral type O9V) which means it has strong hydrogen and helium emission lines. The specINTI Toolbox2, in section 5. Considerations Regarding the Wavelength calibration of Spectra p8 shows steps for using absorption lines of H, He and O2 to obtain dispersion coefficients for ones spectrograph. The annotated profile on p15 gives the published values for these lines.
I started by taking spectra of 10 Lac (11 30s exposures), which gave me this raw profile:
As you can see, there are a few absorption lines visible on the left, in the blue region of the spectrum, but nothing obvious in the middle or right, in the red region. If we ignore the absorption lines (the sharp dips) for a moment, we can see an overall continuum that has a maximum on the left at about 200 pixels and the drops rapidly and levels out toward the right. What we need to do is to divide out this continuum curve, so that the dips are more prominent. We may even see some absorption lines that aren’t visible at all in the initial raw profile.
We can use the configuration file CONF_REMOVE_CONTINUUM on p14 of Toolbox2. When we run specINTI, we get a much more useful profile, which clearly shows absorption lines:
the, using the annotated plot on p15 of the Toolbox2. we can identify and label absorption lines on our own profile of 10 Lac, as shown below:
We open our rectified spectrum in specINTI and click successively on the locations of the 13 absorption lines. That gives us 13 position values corresponding to 13 wavelength values:
#--------------------------------------------------------------------------------
# Wavelengths of measured points
# 13-lines from 2023-12-20 10 Lac Rectified (p2.fits)
#--------------------------------------------------------------------------------
fit_wavelength: [3835.39, 3889.05, 3970.08, 4026.19, 4101.75, 4340.48, 4685.68, 4861.34, 4921.93, 5875.61, 6562.81, 6869.1, 7065.19]
#--------------------------------------------------------------------------------
# Measured points' coordinates
#--------------------------------------------------------------------------------
# Corresponding 13 absorption lines in 10 Lac
# by reading off x-values in Profile display
fit_posx: [47.211, 83.091, 137.856, 177.513, 232.278, 400.349, 638.292, 759.152, 800.697, 1445.858, 1891.105, 2096.972, 2221.450]Using these lines in the CONF_COMPUTE_POLY configuration file and pressing Run, we get the following result in the console:
-------------------------------------
Polynomial fit
calib_coef: [-1.120594104431658e-09, 3.631287942199457e-05, 1.4071364519892544, 3772.5743186764466]
-------------------------------------
O-C: [-3.697 -0.695 2.836 2.693 0.384 -1.188 0.439 0.098 -0.04 -4.009
6.907 -3.545 -0.181]
Root Mean Square Error = 2.8768 A
End.Return to Workflow