SDSS-IV MaNGA: Modeling the spectral line-spread function to subpercent accuracy

The Sloan Digital Sky Survey IV Mapping Nearby Galaxies at APO (MaNGA) program has been operating from 2014 to 2020, and has now observed a sample of 9269 galaxies in the low redshift universe (z ∼ 0.05) with integral-field spectroscopy. With rest-optical (λλ0.36-1.0 μm) spectral resolution R ∼ 2000 the instrumental spectral line-spread function (LSF) typically has 1σ width of about 70 km s⁻¹, which poses a challenge for the study of the typically 20-30 km s⁻¹ velocity dispersion of the ionized gas in present-day disk galaxies. In this contribution, we present a major revision of the MaNGA data pipeline architecture, focusing particularly on a variety of factors impacting the effective LSF (e.g., under-sampling, spectral rectification, and data cube construction). Through comparison with external assessments of the MaNGA data provided by substantially higher-resolution R ∼ 10,000 instruments, we demonstrate that the revised MPL-10 pipeline measures the instrumental LSF sufficiently accurately (≤0.6% systematic, 2% random around the wavelength of Hα) that it enables reliable measurements of astrophysical velocity dispersions σ_Hα ∼ 20 km s⁻¹ for spaxels with emission lines detected at signal-to-noise ratio > 50. Velocity dispersions derived from [O II], Hβ, [O III], [N II], and [S II] are consistent with those derived from Hα to within about 2% at σ_Hα > 30 km s⁻¹. Although the impact of these changes to the estimated LSF will be minimal at velocity dispersions greater than about 100 km s⁻¹, scientific results from previous data releases that are based on dispersions far below the instrumental resolution should be reevaluated.