BayeSED-GALAXIES. II. Bayesian Full Spectrum Analysis of Galaxies and Application in the CSST Wide-field Slitless Spectroscopy Survey

Abstract The China Space Station Telescope (CSST) will simultaneously conduct wide-field multiband photometric imaging and spectroscopic slitless surveys, poised to significantly advance cosmology and galaxy evolution research. Meeting CSST’s cosmological goals requires precise redshifts ( σ NMAD ≲ 0.002–0.005) from its low-resolution ( R ∼ 200) and potentially blended slitless spectra. We present BayeSED3, extended for Bayesian full spectrum analysis, incorporating detailed nebular emission modeling (via CLOUDY) and a robust Bayesian treatment of the model scaling factor, enhancing reliability over optimization techniques for low signal-to-noise ratio (SNR) spectra. Validated on realistic mock data generated with the CSST Emulator for Slitless Spectroscopy (CESS; median SNR of 1.65, instrumental and self-blending effects included), our method achieves excellent redshift precision with three-band (GU+GV+GI) spectroscopy, exceeding requirements: σ NMAD = 0.0008 (∼80% success rate) for star-forming galaxies and σ NMAD = 0.0015 (∼50% success rate) for quiescent galaxies. Stellar mass ( M * ; σ NMAD ≈ 0.015 dex for star-forming galaxies, ≈0.016 dex for quiescent galaxies) and star formation rate ( σ NMAD ≈ 0.05 dex for star-forming galaxies, especially at SNR > 1) can also be reliably recovered. While spectral self-blending increases scatter ( σ NMAD ) by ≳30%, combining spectroscopy with CSST’s seven-band photometry crucially improves accuracy, especially for quiescent galaxies and in data-limited scenarios. Combining photometry with single-band spectroscopy leads to reasonable redshift estimation. While GU+photometry shows limited performance, GI+photometry yields success rates of >60% (for star-forming galaxies) and >40% (for quiescent galaxies) with σ NMAD ≲ 0.002, and GV+photometry achieves success rates of >35% (for star-forming galaxies) and ∼40% (for quiescent galaxies) at similar precision. Our comprehensive Bayesian framework thus provides a robust tool for precise galaxy characterization, enhancing scientific returns from CSST’s unique dataset despite inherent slitless spectroscopy challenges.