Classifying Single Stars and Spectroscopic Binaries Using Optical Stellar Templates

TL;DR

This paper introduces PyHammer v2.0, an advanced stellar spectral classification tool that now includes new spectral types, an empirical spectral library, and the ability to classify double-lined spectroscopic binaries with high accuracy.

Contribution

PyHammer v2.0 extends spectral classification capabilities to include C stars, DA white dwarfs, and SB2 binaries, with a new empirical library and improved classification accuracy.

Findings

Over 95% classification success rate for SB2 spectra

Includes new spectral types and an empirical spectral library

Accurate classification across various spectral types and S/N ratios

Abstract

Stellar spectral classification is a fundamental tool of modern astronomy, providing insight into physical characteristics such as effective temperature, surface gravity, and metallicity. Accurate and fast spectral typing is an integral need for large all-sky spectroscopic surveys like the SDSS and LAMOST. Here, we present the next version of PyHammer, stellar spectral classification software that uses optical spectral templates and spectral line index measurements. PyHammer v2.0 extends the classification power to include carbon (C) stars, DA white dwarf (WD) stars, and also double-lined spectroscopic binaries (SB2). This release also includes a new empirical library of luminosity-normalized spectra that can be used to flux calibrate observed spectra, or to create synthetic SB2 spectra. We have generated physically reasonable SB2 combinations as templates, adding to PyHammer the ability to spectrally type SB2s. We test classification success rates on SB2 spectra, generated from the SDSS, across a wide range of spectral types and signal-to-noise ratios. Within the defined range of pairings described, more than $95\%$ of SB2s are correctly classified.