Physical Parameters of Stars in NGC 6397 Using ANN-Based Interpolation and Full Spectrum Fitting

Abstract

Stellar spectral interpolation is critical technique employed by fitting software to derive the physical parameters of stars. This approach is necessary because on-the-go generation of synthetic stellar spectra is not possible due to the complex and high cost of computation. The goal of this study is to develop a spectral interpolator for a synthetic spectral library using artificial neural networks (ANNs). The study aims to test the accuracy of the trained interpolator through self-inversion and, subsequently, to utilize the interpolator to derive the physical parameters of stars in the globular cluster NGC 6397 using spectra obtained from the Multi Unit Spectroscopic Explorer (MUSE) on the Very Large Telescope (VLT). In this study, ANNs were trained to function as spectral interpolators. The ULySS full-spectrum fitting package, integrated with the trained interpolators, was then used to extract the physical parameters of 1587 spectra of 1063 stars in NGC 6397. The trained ANN interpolator achieved precise determination of stellar parameters with a mean difference of 31 K for $T_{\rm eff}$ and 0.01 dex for [Fe/H] compared to previous studies. This study demonstrates the efficacy of ANN-based spectral interpolation in stellar parameter determination, offering faster and more accurate analysis.