Simulated quantitative stellar classification at different spectral resolutions

摘要

We have simulated a set of "observed" spectra by combining synthetic 20 A resolution spectra (Kurucz models) with random variables drawn from the standard Gaussian distribution. The simulated spectra have been classified using Cayrel's perturbation method for deriving the stellar parameters: effective temperature, gravity and metallicity. Then we have decreased spectral resolutions artificially step by step. The full range of resolutions (from 20 A to lower resolution spectrophotometry, narrow-band, intermediate-band, and finally to broad-band photometry) has been covered, and the classification accuracies were estimated at every step. Useful features in the run of classification accuracies with spectral resolution have been noted and discussed. Selected photometric systems have been analyzed with the use of the same perturbation method, also. The results can help in designing optimized observing strategies in future spectral and photometric classification projects.