SDSS-V/MWM DR19: Milky Way Mapper Release

• SDSS-V/MWM DR19 is a spectroscopic and catalog release that integrates near-IR APOGEE and optical BOSS data to derive detailed stellar parameters, elemental abundances, and kinematics.
• The release employs advanced pipelines such as ASPCAP, The Cannon, and SLAM to achieve high precision in measurements like T_eff, log g, and radial velocities across diverse stellar populations.
• DR19 provides extensive value-added catalogs for open clusters, M dwarfs, halo stars, and white dwarfs, enabling comprehensive studies in Galactic archaeology and chemical evolution.

Searching arXiv for SDSS-V/MWM DR19 papers and related sources to ground the encyclopedia entry. arxiv_search query="SDSS-V MWM DR19 Data Release 19 Milky Way Mapper ASPCAP OCCAM M dwarf MINESweeper", max_results=10 SDSS-V/MWM Data Release 19 (DR19) is the Milky Way Mapper component of the nineteenth data release of the Sloan Digital Sky Survey in its fifth phase, and the first public release of SDSS-V data from all three “mappers” (Collaboration et al., 9 Jul 2025). For Milky Way Mapper, DR19 represents the culmination of the APOGEE-2 survey plus the first two years of new MWM observations analyzed through the ASPCAP pipeline, combining near-infrared APOGEE spectroscopy and optical BOSS spectroscopy with derived stellar parameters, detailed abundances, radial velocities, distances, and quality flags (Mészáros et al., 9 Jun 2025). In practice, DR19 is both a spectroscopic release and a catalog release: it includes raw and reduced spectra, pipeline outputs, and Milky Way–focused Value-Added Catalogs (VACs) that extend the core products to open clusters, M dwarfs, halo stars, white dwarfs, and Bayesian distance inference (Collaboration et al., 9 Jul 2025).

1. Release definition and survey scope

Within SDSS-V, Milky Way Mapper is one of three scientific “mappers,” alongside the Black Hole Mapper and the Local Volume Mapper (Collaboration et al., 9 Jul 2025). DR19 is therefore a release-level milestone for SDSS-V as a whole, but its MWM component has a distinct role: it assembles a large, uniform stellar spectroscopic dataset across multiple Galactic environments and target classes.

For MWM, DR19 provides ~1.2 million near-IR APOGEE spectra at R≈22 500 over 1.51–1.70 µm and ~800 000 optical BOSS spectra at R≈2000 over 3600–10 000 Å, corresponding to ~400 000 and ~475 000 stars, respectively (Collaboration et al., 9 Jul 2025). The ASPCAP paper describes the APOGEE-based parameter release as containing atmospheric parameters and abundances for 964,989 stars, including all APOGEE-2 targets together with new observations of 336,511 stars from the Apache Point Observatory observed until 4 July 2023 (Mészáros et al., 9 Jun 2025). A plausible implication is that DR19 should be understood as a layered release in which spectra, per-visit measurements, coadded products, and downstream inference products are not interchangeable counts.

The scientific scope is correspondingly broad. DR19 includes derived stellar parameters such as $T_{\rm eff}$ and $\log g$, metallicity and element abundances, radial velocities, distances, extinctions, and multiple catalog-level quality indicators (Collaboration et al., 9 Jul 2025). It also distributes specialized VACs for distinct stellar populations and analysis tasks, allowing DR19 to function as a common reference point for Galactic archaeology, stellar population studies, and chemically resolved mapping of the Milky Way.

2. Observing system, spectral domains, and sky footprint

MWM DR19 is built on two spectroscopic systems. The twin APOGEE-N/S near-infrared spectrographs deliver R≈22 500 over the H band, with λ_{\min}=1.51 µm and λ_{\max}=1.70 µm, while the BOSS-N/S optical spectrographs deliver R≈2000 over ≈3600 Å to ≈10 000 Å (Collaboration et al., 9 Jul 2025). Both are fed by robotic fiber positioners (“FPS”), enabling rapid reconfiguration and dual-hemisphere coverage (Collaboration et al., 9 Jul 2025). In the APOGEE open-cluster context, the spectrographs are described as being fed by 300 fibers on the 2.5 m Sloan Foundation telescope at Apache Point Observatory and on the 2.5 m du Pont telescope at Las Campanas Observatory, with nearly continuous H-band coverage and typical S/N≈100 per resolution element for H≈12 mag stars (Otto et al., 9 Jul 2025).

Targeting is organized through “cartons” defined by color, magnitude, and astrometric cuts, with field assignment by robostrategy (Collaboration et al., 9 Jul 2025). The release documentation gives representative examples. The Galactic Genesis carton for APOGEE red giants selects stars with H\<11, (G–H)\>3.5 mag OR no G detection, and 2MASS quality flags: ph_qual∈{A,B}, rd_flag≤3, cc_flag=0 (Collaboration et al., 9 Jul 2025). The Solar Neighborhood Census uses Gaia [parallax](https://www.emergentmind.com/topics/parallax) ϖ\>10 mas, G\<12 mag, explicitly probing within 100 pc (Collaboration et al., 9 Jul 2025). Halo metal-poor giants are pre-selected by (u–g) vs (g–r) photometric metallicity cuts with [Fe/H]<–1 and |b|\>20° (Collaboration et al., 9 Jul 2025).

The sky coverage is similarly heterogeneous but survey-wide. The full MWM footprint places red-giant targeting in the Galactic plane at |b|≲15°, halo targeting at |b|≳20°, and achieves ≈4π sr total sky coverage through dual-hemisphere operations (Collaboration et al., 9 Jul 2025). For APOGEE M dwarfs, the DR19 Milky Way Mapper footprint is described as covering essentially the entire Galactic plane and intermediate latitudes, (|b|≲30°), across both hemispheres and reaching into the solar neighborhood at d≲1 kpc (Behmard et al., 24 Jan 2025).

The release documentation also states the nominal radial-velocity scaling

$\sigma_v \approx \frac{c}{R\sqrt{N_{\rm pix}}},$

with N_pix≈1024 pixels per resolution element (Collaboration et al., 9 Jul 2025). For APOGEE, this gives a theoretical minimum σ_{v,\min}≈0.42 km/s, while DR19 achieves a floor ≈41 m/s at SNR≳100; for BOSS, the corresponding estimate is ≈4.7 km/s, consistent with pipeline performance (Collaboration et al., 9 Jul 2025). This establishes from the outset that DR19 spans two very different precision regimes: high-resolution H-band spectroscopy for detailed stellar labels and low-resolution optical spectroscopy for large-scale classification and complementary parameter inference.

3. Reduction, analysis framework, and core data products

The raw and reduced data products in MWM DR19 are organized around separate reduction pipelines for APOGEE and BOSS, and a common analysis framework for derived quantities. Raw frames and one-dimensional spectra are archived in the Science Archive Server under the environmental variables APOGEE_DATA_N and BOSS_SPECTRO_DATA_N (Collaboration et al., 9 Jul 2025). Reduced spectra are distributed as FITS files with [FLUX](https://www.emergentmind.com/topics/flux), ERR, and LAMBDA extensions at both the per-exposure and coadd levels (Collaboration et al., 9 Jul 2025).

For APOGEE, the Data Reduction Pipeline is apred_vers=1.3, with the sequence overscan → bias → flat → [PSF](https://www.emergentmind.com/topics/poisson-safety-function-psf) trace → arclamp+FSI calibration → extraction → sky subtraction → flux calibration (Collaboration et al., 9 Jul 2025). The more detailed ASPCAP documentation describes the APOGEE DRP as performing detector calibration, fiber tracing and optimal extraction, flat-field correction, wavelength calibration via ThAr/Ne lamps + sky lines, sky subtraction with dedicated sky fibers, telluric correction using telluric standard star fibers and model fitting, barycentric correction, radial-velocity derivation by template cross-correlation, and visit combination into coadded spectra (Mészáros et al., 9 Jun 2025). For BOSS, DR19 uses idlspec2d v6_1_3 with overscan → [CCD](https://www.emergentmind.com/topics/cross-modal-consistency-distortion-ccd) flat → extraction of red+blue arms → wavelength calibration with updated dust maps (Green 2015 at |b|\<15°) → co-adds per MJD and per field epoch (Collaboration et al., 9 Jul 2025).

Derived stellar quantities are produced within the Astra framework (Collaboration et al., 9 Jul 2025). On the APOGEE side, the relevant pipelines listed for DR19 are ASPCAP, The Payne, AstroNN, and APOGEENet; on the BOSS side, they are BOSSNet, SLAM (M dwarfs), SnowWhite & corv (white dwarfs), and LineForest (line indices) (Collaboration et al., 9 Jul 2025). The release therefore does not rely on a single inference formalism across all stellar classes and spectral domains.

For APOGEE H-band spectra, ASPCAP uses FERRE as its core engine and derives parameters by $\chi^2$-minimization against grids of synthetic spectra (Mészáros et al., 9 Jun 2025). DR19 provides both raw and calibrated values for T_eff and log g, in the columns raw_teff, teff, raw_logg, and logg (Mészáros et al., 9 Jun 2025). Chemical analysis proceeds in two stages: ASPCAP first fits globally for [M/H], [α/M], [C/M], and [N/M], then fixes T_eff, log g, [M/H], and micro/macro-turbulence while fitting individual elements in dedicated wavelength windows (Mészáros et al., 9 Jun 2025). The release includes 24 abundance flags for 21 distinct elements, with multiple measures for C, N, and O (Mészáros et al., 9 Jun 2025).

An important technical distinction in DR19 is that, although MWM observes both APOGEE and BOSS spectra, ASPCAP parameters in DR19 are derived from the H-band only (Mészáros et al., 9 Jun 2025). This corrects a common misunderstanding that all MWM stellar parameters in DR19 form a single spectroscopic product family. In practice, DR19 is a federation of products, some APOGEE-based, some BOSS-based, and some VAC-specific.

4. Calibration, verification, and measurement precision

The APOGEE-based DR19 parameters were validated against external reference scales. For effective temperatures, a low-reddening sample defined by E(B–V)\<0.02, S/N\>50, and flag_bad=false was compared to InfraRed Flux Method relations from González Hernández & Bonifacio (2009) and Casagrande et al. (2021) (Mészáros et al., 9 Jun 2025). For giants with log g\<3.8, the median offset is approximately

$$\langle T_{\rm raw} - T_{\rm IRFM}\rangle_{\rm giants}\approx -70\ {\rm K},$$

while for dwarfs it is approximately

$$\langle T_{\rm raw} - T_{\rm IRFM}\rangle_{\rm dwarfs}\approx 0\ {\rm K}.$$

The inferred precision is 50–70 K for giants and 70–100 K for dwarfs (Mészáros et al., 9 Jun 2025).

For surface gravities, raw $\log g$ values were checked against APOKASC3 and TESS asteroseismic reference samples (Mészáros et al., 9 Jun 2025). The reported offsets are +0.09 [dex](https://www.emergentmind.com/topics/differential-extension-dex) for RGB stars, +0.18 dex for RC stars, and +0.14 dex for the TESS sample overall, with a spectroscopic precision of 0.07–0.09 dex for RGB stars and 0.11 dex for the TESS comparison (Mészáros et al., 9 Jun 2025). Calibrated gravities in DR19 apply a T_eff- and [M/H]-dependent zero-point correction of the form

$$\log g_{\rm calib}=\log g_{\rm raw}+\Delta_{\log g}(T_{\rm eff},[M/H]).$$

Abundance calibration in DR19 is based on a Solar-Neighborhood Sample with |[M/H]|\<0.05, distance\<500 pc, S/N\>50, and no bad flags, under the assumption that $\langle[X/M]\rangle\approx 0$ for X≠C,N,O (Mészáros et al., 9 Jun 2025). Constant offsets were derived separately for dwarfs and giants, and users obtain calibrated abundances through

$[X/H]_{\rm cal} = [X/H]_{\rm raw} - \Delta_X.$

Three empirical approaches were used to bound the abundance precision: scatter in the solar-neighborhood sample, internal dispersions in 12 open clusters, and half the star-to-star difference in 348 wide binaries (Mészáros et al., 9 Jun 2025). Representative precision ranges for giants include σ(α)≈0.012–0.027 dex, σ(O)≈0.056–0.079 dex, σ(Mg)≈0.041–0.053 dex, and σ(Si)≈0.023–0.032 dex, with at least ten species below 0.10 dex (Mészáros et al., 9 Jun 2025).

Radial-velocity zero points were compared against Gaia DR3, GALAH DR4, and Gaia-ESO DR5 on non-variable comparison subsets (Mészáros et al., 9 Jun 2025). The median offsets are +0.19±0.96 km/s relative to Gaia DR3, +0.07±0.94 km/s relative to GALAH DR4, and +0.06±0.87 km/s relative to Gaia-ESO DR5 (Mészáros et al., 9 Jun 2025). The release notes explicitly state that No global re-zeroing has been applied in DR19, although the measured offsets may be subtracted if desired (Mészáros et al., 9 Jun 2025). The internal Doppler precision is reported as ≲70 m/s for S/N\>50 H-band spectra after SDSS-V fiber upgrades and ≲100 m/s for cool giants with T_eff\<4000 K (Mészáros et al., 9 Jun 2025).

DR19 also contains population-specific validation exercises. For APOGEE M dwarfs analyzed with The Cannon, the principal training set comprises 79 wide binaries with FGK primaries, spanning −0.56<[Fe/H]<+0.31 dex (Behmard et al., 24 Jan 2025). Leave-one-out cross-validation yields rms scatters of 0.09–0.17 dex in A(X) and 68 K in T_eff, with reduced χ²≈1.4 (Behmard et al., 24 Jan 2025). Repeat-visit spectra for ≈500 stars were used to inflate label covariance errors so that label differences follow unit-Gaussian scatter, giving median uncertainties σ_Teff≈13 K and σ_A(X)≈0.018–0.029 dex for X∈{Fe, C, N, O, Mg, Si, Ca, Ti, Cr, Ni} (Behmard et al., 24 Jan 2025). The same model recovers Hyades [M/H] and A(O) to ±0.05 dex and reproduces Fe, C, O, Mg, Al, Ca for the 21 M dwarfs in Souto et al. (2022) to ≲0.1 dex, aside from edge-of-parameter-space cases (Behmard et al., 24 Jan 2025).

5. Catalog architecture, access modes, and value-added products

DR19 products are distributed through the Science Archive Server ([SAS](https://www.emergentmind.com/topics/shortest-absent-substring-sas)), the Catalog Archive Server ([CAS](https://www.emergentmind.com/topics/cartridges-at-scale-cas)), and SciServer interfaces (Mészáros et al., 9 Jun 2025). On the file side, SAS supports HTTP download as well as rsync and Globus, and the file-system hierarchy is documented by the public data model (Collaboration et al., 9 Jul 2025). On the catalog side, CAS and CasJobs provide SQL and ADQL access, including asynchronous queries and MyDB table sharing (Collaboration et al., 9 Jul 2025). SciServer Compute & Jupyter exposes a pre-mounted $sdss_sas/dr19 environment, and the web interfaces include Zora for interactive target search and spectra visualization and the Valis REST API for programmatic access to spectra and metadata (Collaboration et al., 9 Jul 2025).

The principal APOGEE database tables include apogeeStar, which stores per-object parameters and abundances such as raw_teff, teff, raw_logg, logg, raw_x_h, x_h, e_x_h, and x_h_flags; apogeeVisit, which stores per-visit radial velocities, S/N, and fiber information; and apogeeField and apogeePlate for observational metadata (Mészáros et al., 9 Jun 2025). Value-added tables include allStar and allVisit with reduced spectra, parameter fits, and bitmask flags (Mészáros et al., 9 Jun 2025).

Selected Milky Way Mapper VACs listed for DR19 are as follows (Collaboration et al., 9 Jul 2025):

VAC name	Contents	CAS table
MWM_MDWARF	detailed M dwarf abundances	mos_mdwabund
MWM_MINESWEEPER	halo Stellar Params & distances from MINESweeper	minesweeper
APOGEE_STARHORSE	Bayesian distances, extinctions for 308 860 giants	apogee_starhorse
APOGEE_OCCAM	open cluster membership, [Fe/H], RVs, orbits for 164 clusters	occam_cluster/member
MWM_WD	DA white dwarf $\log g$0, $\log g$1, M, R, RVs for 8 500 stars	wd_da_df
APMADGICS	MADGICS component spectra for APOGEE visits	allVisit_MADGICS_th/dd
MWM_STARFLOW	stellar ages for 378 000 red giants with full posterior	StarFlow_summary
DL1_SDSS_EROSITA	spectroscopic+X-ray eROSITA All-Sky counterparts	DL1_eROSITA_eRASS1

The inclusion of both core tables and VAC tables is central to how DR19 is used. The release is not limited to generic stellar parameters; it explicitly exposes program-specific inference layers such as STARHORSE distances, MINESweeper halo distances, [OCCAM](https://www.emergentmind.com/topics/optimal-cross-correlation-analysis-for-multiplets-occam) cluster membership and chemistry, and specialized white-dwarf and M-dwarf products (Collaboration et al., 9 Jul 2025). This suggests that DR19 is designed as an extensible research platform rather than only as a uniform parameter catalog.

6. Population-specific DR19 products

One of the most developed DR19 subproducts is the APOGEE M-dwarf abundance catalog. In the Milky Way Mapper program, high-resolution near-infrared spectra of low-mass stars are collected with the twin APOGEE spectrographs, and over the first two years of SDSS-V they amassed ≃50 000 M-dwarf exposures (Behmard et al., 24 Jan 2025). After quality cuts of SNR≥50, single-star astrometry, 3088 K< T_eff\<4085 K, 4< log g\<5.5, and χ²_fit\<10⁵, the DR19 M-dwarf catalog contains 16 590 unique M-dwarf spectra (Behmard et al., 24 Jan 2025). The labels are inferred with The Cannon, using the normalized-flux model

$\log g$2

and test-step minimization of

$\log g$3

over 13 labels (Behmard et al., 24 Jan 2025). The catalog provides identifiers, T_eff, and detailed abundances A(Fe), A(C), A(N), A(O), A(Mg), A(Si), A(Ca), A(Ti), A(Cr), A(Ni), together with covariance-matrix uncertainties, χ²_fit, and the temp_agree flag (Behmard et al., 24 Jan 2025).

DR19 also includes a BOSS-based M-dwarf parameter product. "Stellar Parameters of BOSS M dwarfs in SDSS-V DR19" reports the use of the Stellar LAbel Machine (SLAM), a data-driven model based on Support Vector Regression, to derive [Fe/H], T_{\rm eff}, and \log g for SDSS-V M dwarfs from low-resolution optical spectra with R\sim2000 obtained using the BOSS spectrographs (Qiu et al., 25 Nov 2025). These parameters are calibrated using LAMOST F, G or K dwarf companions for [Fe/H] and APOGEE Net for T_{\rm eff} and \log g; the abstract reports comparison scatters and biases for metallicity, temperature, and gravity, and notes a correction formula for APOGEE ASPCAP metallicities (Qiu et al., 25 Nov 2025). In DR19, therefore, M dwarfs are represented by both high-resolution APOGEE abundance inference and low-resolution BOSS parameter estimation.

For the stellar halo, "Mapping the Distant and Metal-Poor Milky Way with SDSS-V" describes the first all-sky low-resolution spectroscopic survey of the Milky Way’s stellar halo and a stellar parameter pipeline that simultaneously models spectra, broadband photometry, and parallaxes to derive stellar parameters, metallicities, alpha abundances, and distances (Chandra et al., 1 Aug 2025). The resulting BOSS-MINESweeper catalog is stated to be validated across a wide range of stellar parameters and metallicities using star clusters and comparison to high-resolution spectroscopic surveys, and the abstract highlights capabilities that include identifying the most chemically peculiar stars in the Galaxy, discovering and mapping distant halo substructures, and measuring the all-sky dynamics of the Milky Way on the largest scales (Chandra et al., 1 Aug 2025). The abstract further states that the BOSS-MINESweeper catalog for SDSS DR19 is publicly available and will be updated for future data releases (Chandra et al., 1 Aug 2025).

White-dwarf products form another distinct DR19 component. "Double White Dwarf Binaries in SDSS-V DR19" presents a search for hydrogen-atmosphere (DA) double white dwarf binaries using the multiple-exposure structure of SDSS-V to quantify radial-velocity variations between sub-exposures (Pallathadka et al., 3 Sep 2025). The study reports 63 DWD binary candidates, of which 43 are new discoveries, with tentative orbital periods for 10 systems (Pallathadka et al., 3 Sep 2025). It also derives constraints on the white-dwarf binary population, including a Galactic white-dwarf binary fraction with < 0.4 AU separations of 9%, a power-law index \alpha = -0.62, an expectation of ≤ 10 super-Chandrasekhar binaries that merge within a Hubble time in the sample, and a prediction that ≤ 5 systems in the sample should be detectable by LISA, one of which has already been identified as a verification source (Pallathadka et al., 3 Sep 2025).

7. Scientific results enabled by DR19

The open-cluster analysis from the OCCAM survey is one of the clearest examples of DR19 as a Galactic-structure dataset. Using SDSS-V/MWM DR19, the OCCAM team established a sample of 164 high quality open clusters and 1 083 red-giant members selected from Gaia-based membership, spectroscopic membership probabilities, and visual inspection against PARSEC isochrones (Otto et al., 9 Jul 2025). They find an overall linear Galactic radial [Fe/H] gradient of -0.075 \pm 0.006 dex kpc^{-1} using present-day Galactocentric radius and -0.068 \pm 0.005 dex kpc^{-1} using the guiding-center radius (Otto et al., 9 Jul 2025). They also fit a bilinear model with a knee, but an AIC comparison favors the simpler single-line model (Otto et al., 9 Jul 2025). For azimuthal structure, five azimuthal wedges show negative radial slopes from -0.055 to -0.093 dex kpc^{-1}, while direct azimuthal gradients are nearly flat, -0.000 ± 0.007 to -0.003 ± 0.007 dex deg^{-1} (Otto et al., 9 Jul 2025). This combination argues against strong global azimuthal abundance gradients while leaving room for localized non-axisymmetric structure.

DR19 has also been used to reconstruct the Galaxy’s chemical history with global and radially resolved Galactic chemical evolution modeling. "Reconstructing the Milky Way chemical map with Galactic Chemical Evolution tool OMEGA+ from SDSS-MWM" defines a golden sample of 393 743 disk stars selected from raw DR19 after cuts in flags, T_{\rm eff}, \log g, S/N, carbon abundance, cluster membership, and Galactic position (Hegedűs et al., 31 May 2025). The work uses a two-infall model in OMEGA+, with thin/thick-disk separation defined in the [Mg/M]–[M/H] plane by

$\log g$4

(Hegedűs et al., 31 May 2025). The best-fit global parameters include τ₁=0.32±0.02 Gyr, τ₂=2.86±0.70 Gyr, t_{\rm peak}=4.13±0.19 Gyr, τ_{\rm up}=0.55±0.06 Gyr, and σ₂/σ₁=7.61±0.23, with radial trends interpreted as an inside-out disk assembly (Hegedűs et al., 31 May 2025). A plausible implication is that DR19 moved Milky Way chemical-evolution analysis from a mainly local inference problem to a spatially resolved one.

Other DR19 products extend the science beyond disk chemistry. The halo survey abstract emphasizes distant halo substructure, chemically peculiar stars, and all-sky halo dynamics (Chandra et al., 1 Aug 2025). The APOGEE M-dwarf catalog detects the canonical radial metallicity gradient in the solar neighborhood, ∂[Fe/H]/∂R≈−0.06 dex kpc⁻¹, and mild [α/Fe] elevation at |b|\>20°, consistent with an increasing thick-disk contribution (Behmard et al., 24 Jan 2025). The white-dwarf binary catalog connects DR19 to compact-binary demographics and low-frequency gravitational-wave source forecasts (Pallathadka et al., 3 Sep 2025). Taken together, these studies show that SDSS-V/MWM DR19 is not a single-topic release but a common observational substrate for open-cluster chemistry, disk chemical evolution, halo cartography, low-mass stellar abundances, and compact-binary population studies.