This research work presents a relative flux calibration method for LAMOST, which may be applied to connect a blue spectrum with a red spectrum together to build the whole spectrum across the total wavelength range (3700 ~ 9000Å). In each spectrograph, researchers estimate the effective temperatures of selected stars using a grid of spectral line indices in the blue spectral range and a comparison to stellar atmosphere models. For each spectrograph, stars of A and F types are selected as standard stars, and the theoretical spectra are used to calibrate both the blue (3700 ~ 5900 Å) and red spectrograph arms (5700 ~ 9000 Å). Then the spectral response function for these standard stars can be used to correct the raw spectra provided by the other fibers of the spectrograph, after a fiber efficiency function has been derived from twilight flat-field exposures. A key problem in this method is the fitting of the stellar continuum, some give also a detailed description of this step. The method is tested by comparing a small sample of LAMOST spectra calibrated in this way to stars also observed by the Sloan Digital Sky Survey. 

From the comparison with SDSS calibrated spectra, the accuracy of our relative calibration is good enough, and the effective temperature estimation is also reliable enough to pick out standard stars. The more spectra in each sub-field regarded as standard stars, the better the accuracy of the flux calibration will be theoretically. Researchers need to do more experiments to study the statistical results of the numbers of standard stars and flux accuracy. The paper titled “Relative flux calibration for GSJT (LAMOST)” has been published in Research in Astronomy and Astrophysics.

Top row: left panel: LAMOST spectrum of a star after relative calibration, right panel: SDSS spectrum of the same target. The temperature of this star is 5313K from SDSS SSPP. Bottom row: An 8230K star. left: LAMOST, right: SDSS