Year: 2026
Type: article
Abstract: We present SM-Net , a machine learning model that learns a continuous spectral manifold from multiple high-resolution stellar libraries. SM-Net generates stellar spectra directly from fundamental stellar parameters such as effective temperature ( upper T Subscript normal e normal f normal f T e f f $T_{\mathrm{eff}}$ ), surface gravity ( log g identical to log Subscript 10 Baseline left bracket g divided by left parenthesis normal c normal m normal s Superscript negative 2 Baseline right parenthesis right bracket log g ≡ log 10 [ g / ( c m s − 2 ) ] ${\log}\, g \equiv {\log}_{10}[g/(\mathrm{cm}\,\mathrm{s}{^{-2}})]$ ), and metallicity ( log upper Z identical to log Subscript 10 Baseline left parenthesis upper Z divided by upper Z Subscript circled dot Baseline right parenthesis log Z ≡ log 10 ( Z / Z ⊙ ) $\log Z \equiv \log_{10}\kern-1.5pt(Z/Z_{\odot})$ ), trained on a combined grid derived from PHOENIX–Husser , C3K–Conroy , OB–PoWR , and TMAP–Werner libraries. By combining their respective parameter spaces to achieve broader coverage, we construct an extended composite dataset that spans a significantly larger, more continuous region of stellar parameter space than any single library. The unified grid spans upper T Subscript normal e normal f normal f Baseline equals 2 000 T e f f = 2 000 $T_{\mathrm{eff}} = 2{\,}000$ – 190 000
Institution: The University of Western Australia
Language: English
Subfield: Astronomy and Astrophysics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,550
OpenAlex ID: W7140417088
FWCI:
Cites:
Cited by:
Related to: