SDSS-V: Panoptic All-Sky Spectroscopy

Updated 4 July 2026

SDSS-V is an all-sky, multi-epoch spectroscopic survey combining optical and infrared data to decode the chemo-dynamical history of the Milky Way and other galaxies.
It employs robotic multi-object spectroscopy and ultra wide-field integral-field mapping across dual hemispheres to achieve high-cadence observations.
The survey’s three mappers target the Milky Way structure, black hole growth, and galactic ecosystem enrichment, driving breakthroughs in astrophysics.

Sloan Digital Sky Survey-V (SDSS-V) is the fifth generation of the Sloan Digital Sky Survey and is described as the first all-sky, multi-epoch, optical-to-infrared spectroscopic survey and, in DR19, as the first panoptic, dual-hemisphere spectroscopic survey. It is organized around three scientific “mappers”—the Milky Way Mapper (MWM), Black Hole Mapper (BHM), and Local Volume Mapper (LVM)—and combines robotic multi-object spectroscopy in both hemispheres with ultra wide-field optical integral-field spectroscopy. The survey’s stated purpose is to decode the chemo-dynamical history of the Milky Way, trace the growth physics of supermassive black holes, and understand self-regulation and chemical enrichment in nearby galactic ecosystems (Kollmeier et al., 9 Jul 2025, Collaboration et al., 9 Jul 2025).

1. Historical position and survey identity

SDSS-V extends the SDSS lineage from the imaging-led original SDSS through SDSS-II, SDSS-III, and SDSS-IV into a survey model centered on spectroscopy, cadence, and hemispheric coverage. DR19 presents SDSS-V as having begun in 2020 and as inheriting the SDSS practice of large, public, homogeneous data products while changing the operational paradigm through robotic targeting, multi-epoch design, and coordinated optical, near-infrared, and integral-field modes (Collaboration et al., 9 Jul 2025).

A design-era overview framed SDSS-V as an all-sky, multi-epoch spectroscopic survey of over six million objects, with systematic spectroscopic monitoring on timescales from 20 minutes to 20 years and a resolved gas-mapping component that was intended to be 1000 times larger than the then state of the art in areal IFU coverage (Kollmeier et al., 2017). Later overview papers recast this program under the term “panoptic spectroscopy,” explicitly defining SDSS-V by the conjunction of all-sky access, multi-epoch cadence, optical-to-infrared spectroscopy, and ultra wide-field integral-field mapping (Kollmeier et al., 9 Jul 2025).

This identity is operational as well as scientific. DR18 emphasized that SDSS-V is not a monolithic sample but a modular survey composed of versioned target-selection units called “cartons,” grouped within the three mappers and tied to explicit selection functions, input catalogs, and realized assignments (Almeida et al., 2023). A plausible implication is that SDSS-V should be understood less as a single catalog than as a coordinated survey infrastructure with heterogeneous but formally documented subprograms.

2. Facilities and instrumental architecture

The multi-object spectroscopy (MOS) component operates at the 2.5-m Sloan Foundation Telescope at Apache Point Observatory and the 100-inch du Pont Telescope at Las Campanas Observatory. At both sites, 500 zonal robotic fiber positioners on a wide-field focal plane feed an optical spectrograph with resolving power $R \sim 2000$ and 500 fibers, together with a near-infrared spectrograph with $R \sim 22{,}000$ and 300 fibers; the standard definition used in the overview is $R \equiv \lambda/\Delta\lambda$ (Kollmeier et al., 9 Jul 2025). In the robostrategy description of the Focal Plane System, each of the 500 robotic positioners carries three fibers, but because of APOGEE slit-head capacity only 298 of the 500 actually feed APOGEE, while the remaining 202 are BOSS-only (Blanton et al., 27 May 2025).

The MOS field geometry is asymmetric between the two sites. The fully covered hexagon is $5.442~\mathrm{deg}^2$ at APO and $2.389~\mathrm{deg}^2$ at LCO, while the patrol radius reaches about 315 mm, corresponding to roughly 1.45 deg at APO and 0.96 deg at LCO (Blanton et al., 27 May 2025). These numbers matter because survey planning in SDSS-V is explicitly constrained by local sidereal time, sky brightness, cadence requirements, patrol geometry, and fiber collisions rather than by a simple fixed tiling.

LVM adds a distinct facility, LVM-I, at Las Campanas Observatory. The 2025 overview describes it as an ultra wide-field integral-field system with 1801 lenslet-coupled fibers arranged in a 0.5-degree-diameter hexagon, feeding multiple optical spectrographs at $R \sim 4000$ over $3600\text{--}9800\,\text{\AA}$ (Kollmeier et al., 9 Jul 2025). The dedicated LVM overview gives the more detailed implementation: a new facility of alt-alt mounted siderostats feeding 16 cm refractive telescopes, with a stationary fiber system designed to avoid the line-spread-function instability induced by flexure and fiber motion in conventional moving-fiber systems (Drory et al., 2024). This suggests that the “panoptic” concept in SDSS-V is inseparable from new hardware, not merely from survey scale.

3. Scientific organization: the three mappers

Mapper	Stated aim	Dominant mode
Milky Way Mapper	Decode the chemo-dynamical history of the Milky Way and tackle fundamental open issues in stellar physics	APOGEE and BOSS MOS
Black Hole Mapper	Trace the growth physics of supermassive black holes	Repeat optical MOS, X-ray/optical follow-up
Local Volume Mapper	Understand self-regulation mechanisms and chemical enrichment of galactic ecosystems at the energy-injection scale	Ultra wide-field optical IFS

Within DR19, MWM is presented as a federation of thirteen overarching programs, including Galactic Genesis, Halo, White Dwarfs, Solar Neighborhood Census, Young Stellar Objects, OB Stars, Binary Systems, Compact Binaries, and Asteroseismic Red Giants (Collaboration et al., 9 Jul 2025). One concrete MWM product is the halo survey: the 2025 halo paper describes SDSS-V as conducting the first all-sky low-resolution spectroscopic survey of the Milky Way’s stellar halo and introduces the BOSS-MINESweeper pipeline, which simultaneously models spectra, broadband photometry, and parallaxes to derive stellar parameters, metallicities, alpha abundances, and distances. That catalog is validated against star clusters and high-resolution surveys and is used to identify chemically peculiar stars, map distant halo substructures, and measure all-sky Galactic dynamics (Chandra et al., 1 Aug 2025).

BHM combines time-domain quasar spectroscopy with X-ray counterpart identification. DR19 describes a reverberation mapping program targeting $>1000$ confirmed quasars over at least four dedicated fields with at least 100–150 spectral epochs, the AQMES multi-epoch quasar programs, and SPIDERS/eROSITA spectroscopy for X-ray selected sources (Collaboration et al., 9 Jul 2025). The methodological precursor is the SDSS-RM project, explicitly described as a precursor to the SDSS-V Black Hole Mapper Reverberation Mapping program; its final data release reported 23, 81, 125, and 110 lags for broad H $\alpha$ , H $\beta$ , Mg II, and C IV, respectively, and derived $R \sim 22{,}000$ 0 for the average virial factor from RMS-spectrum line dispersion (Shen et al., 2023).

LVM is the mapper that most strongly differentiates SDSS-V from earlier SDSS phases. The 2024 LVM overview defines it as a 4-year integral-field spectroscopic survey of the Milky Way, Magellanic Clouds, and a sample of local volume galaxies, covering $R \sim 22{,}000$ 1 square degrees of sky and more than 55M spectra. Its scientific purpose is to connect resolved pc-scale sources of feedback to kpc-scale ionized interstellar medium properties, thereby linking energy and metal injection to cooling, shocks, turbulence, bulk flows, and larger disk structures (Drory et al., 2024).

4. Targeting, cadence, and robotic operations

The basic organizational unit of SDSS-V targeting is the carton: a named, versioned target-selection module associated with a specific science goal. DR18 made the targeting databases public precisely so that users could reconstruct the relationship among input catalogs, cartons, calibration targets, and realized fiber assignments (Almeida et al., 2023). In the operational stack summarized by robostrategy, targets are selected into targetdb, fields and designs are generated strategically, and roboscheduler then decides what to execute night by night (Blanton et al., 27 May 2025).

Cadence is a first-class survey variable. Robostrategy defines a field as a fixed field center plus position angle, a design as one planned fiber configuration for one observation, an observation as nominally 12 or 15 minutes of exposure time plus overhead, and an epoch as a set of back-to-back observations that should occur together. Target cadences are parameterized by quantities such as nepochs, nexp[], skybrightness[], delta[], delta_min[], delta_max[], and obs_mode_pk[]. The field-allocation problem is then cast as a large linear program: for each field $R \sim 22{,}000$ 2, cadence $R \sim 22{,}000$ 3, and local-sidereal-time/sky-brightness slot $R \sim 22{,}000$ 4, robostrategy defines $R \sim 22{,}000$ 5, the number of observations allocated to that combination, and maximizes $R \sim 22{,}000$ 6 subject to non-negativity, one cadence per field, and time-availability constraints $R \sim 22{,}000$ 7 (Blanton et al., 27 May 2025).

At the mechanical level, rapid reconfiguration depends on collision-free astrobot motion. The SDSS-V astrobot coordination study models each positioner as a two-degree-of-freedom $R \sim 22{,}000$ 8- $R \sim 22{,}000$ 9 manipulator and evaluates a cooperative artificial potential field method within the completeness seeker algorithm. On a 19-astrobot bench, 974 of 1000 scenarios reached completeness with no parameter modification, corresponding to 97.4% inherent completeness; direct coordination was reported as noticeably faster than two-phase fold-then-deploy motion (Macktoobian et al., 2020). This supports the survey’s claim that data throughput is directly coupled to how many fibers safely reach their assigned targets before each observation.

5. Data releases, pipelines, and public products

DR18 was the first public release of SDSS-V. Its principal content was not yet the main science yield but extensive targeting information for MWM and BHM, together with approximately $R \equiv \lambda/\Delta\lambda$ 0 new spectra, largely from the eROSITA eFEDS field, plus value-added products and previews of DR19 (Almeida et al., 2023). DR19 is described as the second and most substantial SDSS-V release and the first to contain data from all three mappers (Collaboration et al., 9 Jul 2025).

The DR19 high-level contents are 479,081 stars with optical BOSS spectra, 390,676 stars with near-infrared APOGEE spectra, 318,123 galaxies and quasars/AGN with BOSS spectra, one LVM preview tile on the Helix Nebula, and nine value-added catalogs (Collaboration et al., 9 Jul 2025). The release also makes explicit that DR19 is not yet the full dual-hemisphere realization of SDSS-V: the newest SDSS-V spectra in DR19 come only from APO, while broader southern FPS data are deferred to later releases (Collaboration et al., 9 Jul 2025).

Pipeline architecture is mapper-specific. DR19 BOSS reductions use idlspec2d v6_1_3 and distinguish daily, epoch, and allepoch coadds; APOGEE reductions use apred_vers 1.3; and MWM analysis is organized under Astra, which runs multiple stellar-parameter and abundance pipelines and supplies the convenience summary product astraMWMLite (Collaboration et al., 9 Jul 2025). Access follows standard SDSS practice through the Science Archive Server and Catalog Archive Server, but DR19 also introduced the Zora/Valis web framework, the Semaphore targeting-flag system, and the new sdss_id identifier needed because SDSS-V uses multiple crossmatch versions (Collaboration et al., 9 Jul 2025).

6. Scientific capabilities, early results, and caveats

The scientific record already shows that SDSS-V’s architecture is usable across very different regimes. In Galactic archaeology, the halo survey demonstrates simultaneous inference of stellar parameters, chemistry, and distances over the all-sky stellar halo, with concrete applications to chemically peculiar stars, distant substructure discovery, and all-sky dynamics (Chandra et al., 1 Aug 2025). In compact-object work, the cataclysmic-variable survey reports that SDSS-V is the first SDSS phase to specifically target CV candidates, and that a convolutional neural network reduced the number of spectra requiring visual inspection to $R \equiv \lambda/\Delta\lambda$ 1 per cent of over 2 million SDSS-V spectra; the resulting sample yielded 61 new CVs, spectroscopically confirmed 248 published candidates, refuted 13, reported 82 new or improved orbital periods, and re-estimated the CV space density as $R \equiv \lambda/\Delta\lambda$ 2 (Inight et al., 2024).

In time-domain AGN physics, the changing-look AGN J162829.17+432948.5 at $R \equiv \lambda/\Delta\lambda$ 3 was identified from first-year SDSS-V repeat spectroscopy, with a $R \equiv \lambda/\Delta\lambda$ 4 month rebrightening interval constrained by SDSS-V and follow-up data. The discovery paper explicitly presents this as an early result of the Black Hole Mapper program, which is spectroscopically monitoring tens of 1000s of AGNs on timescales of days to years (Zeltyn et al., 2022). Together with the SDSS-RM precursor results, this shows that BHM is intended not as a single-epoch quasar redshift survey but as a cadence-driven accretion-physics program (Shen et al., 2023).

Several caveats recur across the SDSS-V documentation. First, SDSS-V is not a simple magnitude-limited survey: cartons, cadence compatibility, field assignment, and calibration requirements are all part of the selection function (Almeida et al., 2023). Second, early operations were affected by the COVID-era plate program before full robotic operations matured, so public data combine legacy-style and FPS-style observing histories (Collaboration et al., 9 Jul 2025). Third, mapper maturity is uneven: DR19 contains substantial MWM and BHM products but only an LVM preview tile, and the LVM data-analysis pipeline outputs are not yet included in that release (Collaboration et al., 9 Jul 2025). These limitations do not diminish the scope of SDSS-V, but they define the conditions under which its public data should be interpreted.

In aggregate, SDSS-V represents a shift in survey astronomy from static spectroscopic catalog production toward a formally planned, cadence-aware, dual-hemisphere spectroscopic system. Its distinctiveness lies in the simultaneous presence of robotic optical and near-infrared MOS, carton-based targeting, long-baseline scheduling, and half-degree-class integral-field mapping. The survey’s broader legacy will depend not only on the number of spectra collected, but on whether its panoptic design succeeds in linking stars, black holes, and ionized gas within a common public data framework (Kollmeier et al., 9 Jul 2025)