[A Method to Estimate Metal Abundance from Stellar Spectra Using Ca Line Index]

Area of Science:

• Astronomy and Astrophysics
• Computational Astrophysics

Context:

• Stellar metallicity is a key parameter for understanding stellar evolution and galactic chemical history.
• Accurate metallicity estimation is crucial for large spectroscopic surveys like SDSS/SEGUE.
• Existing methods may have limitations with low-resolution spectra.

Purpose:

• To develop and validate a novel method for estimating stellar metallicity ([Fe/H]) using a Backpropagation (BP) Artificial Neural Network (ANN).
• To utilize the Ca line index and effective temperature (Teff) as input features for the ANN model.
• To enable accurate metallicity determination from low-resolution stellar spectra.

Summary:

• A BP ANN model was trained using stellar spectra and parameters from the Sloan Digital Sky Survey (SDSS)/Spectroscopic Evolution of the Extremely Galactic (SEGUE) survey.
• The model takes effective temperature (Teff) and Ca line index as inputs to predict the metallicity ([Fe/H]).
• The trained network effectively predicts stellar metallicity from low-resolution spectra, demonstrating high accuracy.

Impact:

• Provides an accurate and effective tool for measuring stellar metallicity ([Fe/H]) from low-resolution spectra.
• Enhances the capabilities of astrophysical research by enabling more precise analysis of stellar populations.
• Contributes to a better understanding of stellar evolution and the chemical enrichment of galaxies.