To be more precise, it would be better to use the relativistic convention where possible when converting velocities to frequencies. In particular for when the wide-bandwidth SKARAB spectra are used. This is because the further we get from the central frequency, the more velocity-channels will differ (assuming constant frequency-spaced channels). If the v_obs is calculated from astropy, then one of the few instances where this conversion is done is
line 454 in discosscan.py for the spectral resolution.
obs.head.spe.vres = - (self.freq_resolution / self.central_frequency) * CLIGHT
Formula for the relativistic vel resolution can be found on https://www.gb.nrao.edu/~fghigo/gbtdoc/doppler.html
Alternatively, this can be implemented allowing the user to pick the velocity convention.
To be more precise, it would be better to use the relativistic convention where possible when converting velocities to frequencies. In particular for when the wide-bandwidth SKARAB spectra are used. This is because the further we get from the central frequency, the more velocity-channels will differ (assuming constant frequency-spaced channels). If the v_obs is calculated from astropy, then one of the few instances where this conversion is done is
line 454 in discosscan.py for the spectral resolution.
obs.head.spe.vres = - (self.freq_resolution / self.central_frequency) * CLIGHT
Formula for the relativistic vel resolution can be found on https://www.gb.nrao.edu/~fghigo/gbtdoc/doppler.html
Alternatively, this can be implemented allowing the user to pick the velocity convention.