ViSP: DKIST Visible Spectro-Polarimeter

• ViSP is a slit-scanning spectro-polarimeter on DKIST that measures full Stokes parameters to infer solar magnetic field vectors and related thermodynamic properties.
• It employs dual-beam polarimetry with a polychromatic modulator and high-dispersion echelle grating, enabling simultaneous observations across three spectral channels within 380–900 nm.
• ViSP supports advanced solar diagnostics including multi-line inversion techniques for magnetic field mapping, though calibration and detector limitations continue to challenge its performance.

ViSP is an acronym with multiple domain-specific meanings, but in contemporary solar-physics literature it most commonly denotes the Visible Spectro-Polarimeter of the Daniel K. Inouye Solar Telescope (DKIST): a traditional slit-scanning spectrograph with dual-beam polarimetry, a polychromatic polarization modulator, a high-dispersion echelle grating, and three automatically positionable spectral channels tunable anywhere within the $380\text{–}900~\mathrm{nm}$ range (Wijn et al., 2022). In that usage, ViSP is a first-light DKIST instrument developed to measure the full Stokes parameters across spectral line profiles and to recover the magnetic field vector as a function of height in the solar atmosphere together with associated thermodynamic structure (Wijn et al., 2012). The acronym is also used elsewhere for unrelated objects, including the Virtual Internet Service Provider project, mouse primary visual cortex denoted as VISp, and several recent machine-learning and natural-language-processing systems (0904.3634, Shi et al., 2019, Islam, 2 Sep 2025, Wang et al., 13 Jul 2025, Nguyen et al., 11 Feb 2025).

1. Referential scope and disambiguation

The literature represented here uses ViSP, VISP, and VISp in distinct ways. The most technically developed and most extensively documented referent is the DKIST spectro-polarimeter.

Term	Domain	Representative source
ViSP, Visible Spectro-Polarimeter	Solar instrumentation and solar physics	(Wijn et al., 2022)
VISP, Virtual Internet Service Provider	Workflow technologies for telecom/ISP SMEs	(0904.3634)
VISp (V1)	Mouse visual cortex in the Allen Brain Observatory	(Shi et al., 2019)
VISP / ViSP	ML and NLP methods or datasets	(Islam, 2 Sep 2025, Wang et al., 13 Jul 2025, Nguyen et al., 11 Feb 2025)

In solar physics, the capitalization ViSP is associated with DKIST and earlier ATST design documentation. In neuroscience, VISp is the Allen Brain Observatory label for mouse primary visual cortex, treated as the earliest or primary area among the recorded visual areas and analyzed through representation-matching metrics against VGG16 layers (Shi et al., 2019). In workflow technology, VISP denotes a Virtual Internet Service Provider platform intended to let several independent companies cooperate as if they were a single provider through choreography and orchestration technologies (0904.3634). Recent ML/NLP work reuses the acronym for Volatility Informed Stochastic Projection, a sarcasm-generation framework, and a Vietnamese paraphrase benchmark (Islam, 2 Sep 2025, Wang et al., 13 Jul 2025, Nguyen et al., 11 Feb 2025).

2. DKIST/ATST ViSP as an instrument

The solar-physics ViSP was introduced in ATST-era design work as an echelle spectrograph designed to measure three different regions of the solar spectrum in three separate focal planes simultaneously between 380 and 900 nm (Wijn et al., 2012). The later DKIST instrument paper describes it as a traditional slit-scanning spectrograph able to observe solar regions up to a $120\times78~\mathrm{arcsec}^2$ area, with a defining capability to tune anywhere within the $380\text{–}900~\mathrm{nm}$ range and thus realize a virtually infinite number of simultaneous three-wavelength combinations (Wijn et al., 2022).

Its design drivers are wavelength versatility, full-Stokes polarimetry, multi-line simultaneity, and high spectral resolution. The instrument implements dual-beam polarimetry, a polychromatic polarization modulator, and a high-dispersion echelle grating; the DKIST description identifies a grating with 316 lines mm$^{-1}$, blazed at 63.4°, with size $90\times340~\mathrm{mm}^2$ (Wijn et al., 2022). The ATST design report specifies a minimum spectral resolving power of 180,000 for all focal planes, 0.04 arcsec spatial resolution over a 2 arcmin field of view at $600~\mathrm{nm}$, and support for up to 4 diffraction gratings, with full automation for rapid reconfiguration (Wijn et al., 2012).

The polarimetric intent is explicit. ViSP measures the full Stokes vector $(I,Q,U,V)$ across line profiles, so that magnetically sensitive spectral lines can be used to infer the magnetic field vector and its variation with height together with associated thermodynamic properties (Wijn et al., 2012). The DKIST article further characterizes ViSP as the only wavelength-versatile spectro-polarimeter in the first-generation instrument suite and explicitly frames it as a discovery instrument for new spectroscopic and polarimetric diagnostics and for testing improved models of polarized line formation (Wijn et al., 2022).

3. Observing modes, calibration, and inversion methodology

ViSP operates in two principal modes. In polarimetric mode, the modulator spins continuously while the slit is stepped across the field and a full modulation cycle is acquired at each step. In intensity mode, the modulator is stopped and the slit is scanned continuously, sacrificing polarimetry for faster mapping (Wijn et al., 2022). The DKIST paper notes that, with the $650~\mathrm{nm}$ slit, a 2 arcmin field can be rastered in about 80 s in intensity mode, whereas a similar polarimetric map would take nearly 15 minutes at the fastest rate (Wijn et al., 2022).

Calibration and inversion pipelines vary by science case but share a common pattern: level-1 reduction, polarimetric calibration, residual cross-talk correction, and physics-based inversion. In the plage study, ViSP data are processed to level 1 with the ViSP pipeline, the full Stokes vector is recovered through polarimetric calibration, and residual cross-talk is corrected with the ad hoc method of Jaeggli et al. The inversion engine is DeSIRe, which combines the LTE inversion code SIR with the forward non-LTE radiative-transfer solver RH, solving the multilevel non-LTE problem in 1D plane-parallel geometry using analytical response functions derived in LTE (Kuridze et al., 2024).

Other studies employ alternative inversion formalisms tailored to line formation and cadence constraints. The flare pore-rotation work in AR 13293 inverts Fe I $6302.5~\text{\AA}$ with the SPIN Milne–Eddington inversion code, resolves the $180^\circ$ ambiguity with AMBIG using the minimum energy method, and transforms the magnetic vector to the local solar frame using the transformation matrix of Gary and Hagyard (1990) (Yadav et al., 2024). The quiet-Sun multiline study around $120\times78~\mathrm{arcsec}^2$0 uses SIR together with MANCHA3D MHD simulations to determine line $120\times78~\mathrm{arcsec}^2$1 values and to select a two-step strategy: first invert six intensity profiles, then invert the polarimetry of the three strongest lines (Arjona, 17 Jun 2025). The flare spectroscopy study of Ca II H and H$120\times78~\mathrm{arcsec}^2$2 uses RADYN+RH, where RADYN atmospheric snapshots and non-equilibrium densities are passed to RH, which includes a 20-level hydrogen atom and treats the overlap between Ca II H and H$120\times78~\mathrm{arcsec}^2$3 properly (Tamburri et al., 17 Feb 2026).

For magnetic-field inference under the weak-field approximation, the network-heating study gives the standard relations

$120\times78~\mathrm{arcsec}^2$4

and, for the plane-of-sky field,

$120\times78~\mathrm{arcsec}^2$5

The azimuth is obtained from

$120\times78~\mathrm{arcsec}^2$6

with the usual $120\times78~\mathrm{arcsec}^2$7 ambiguity (Judge et al., 2023). In flare back-reaction analysis, the Lorentz-force changes are computed through

$120\times78~\mathrm{arcsec}^2$8

with $120\times78~\mathrm{arcsec}^2$9 and $380\text{–}900~\mathrm{nm}$0 the radial and horizontal magnetic-field components (Yadav et al., 2024).

4. Scientific diagnostics enabled by ViSP

ViSP’s multi-line spectro-polarimetry has been used to reconstruct coupled magnetic, thermodynamic, and radiative structure from the photosphere into the chromosphere. In DKIST plage observations, ViSP simultaneously sampled Fe I $380\text{–}900~\mathrm{nm}$1 and Ca II $380\text{–}900~\mathrm{nm}$2, and the inversions were used to construct semiempirical models of the plage atmosphere. The results indicate dense fibrils in the Ca line, show that the morphological characteristics such as orientation, inclination, and length of fibrils are defined by the topology of the magnetic field in the photosphere, and reveal a prominent magnetic canopy where fibrils are directed toward the observer (Kuridze et al., 2024). The same work introduces the analytical relation $380\text{–}900~\mathrm{nm}$3 for the height at which neighboring canopies merge, linking fibril geometry to the large-scale filling factor of magnetic flux concentrations (Kuridze et al., 2024).

In quiet-Sun magnetism, ViSP’s $380\text{–}900~\mathrm{nm}$4 window supports multiline, depth-resolved inversions beyond the traditional Fe I $380\text{–}900~\mathrm{nm}$5 pair. The reported solution shows that, in the internetwork, the averaged magnetic field strength decreases with height, fields become more horizontal around $380\text{–}900~\mathrm{nm}$6, and the network exhibits a flux-tube/canopy morphology with localized temperature enhancements near tube walls and opposing mass motions inside and around flux tubes (Arjona, 17 Jun 2025). This suggests that ViSP’s broader visible window can recover not only field strength and inclination but also fine-scale thermodynamic structure as a function of optical depth.

In studies of magnetic connectivity and coronal heating, ViSP has been used in a “low” resolution mode with $380\text{–}900~\mathrm{nm}$7 slit width and $380\text{–}900~\mathrm{nm}$8 over a common area of $380\text{–}900~\mathrm{nm}$9. Those observations provided longitudinal Zeeman sensitivities of 0.25 Mx/cm$^{-1}$0 in Fe I $^{-1}$1 and 0.5 Mx/cm$^{-1}$2 in Ca II $^{-1}$3 (Judge et al., 2023). The principal result was negative in a physically important sense: no meaningful correlations were found between heated plasma and chromospheric magnetic-field properties derived from the weak-field approximation on scales below supergranules, and the authors argued that processes hidden from the observations control plasma heating (Judge et al., 2023).

Flare studies use ViSP both for vector magnetometry and for high-resolution spectroscopy. In the C4.1-class flare SOL2023-05-03T20:53, ViSP and VBI data showed two oppositely polarized pores undergoing rotational motion before and during the flare; the magnetic analysis found a 30% relative change in the horizontal component of Lorentz force at the flare peak time and roughly no change in the radial component, supporting the interpretation that $^{-1}$4 drove the observed pore rotation (Yadav et al., 2024). In the C6.7 flare SOL2022-08-19T20:31, ViSP delivered the first flare-time DKIST spectra of Ca II H 396.8 nm and H$^{-1}$5 397.0 nm. Comparison with RADYN+RH simulations showed that the models reproduce the width of H$^{-1}$6 reasonably well but significantly underestimate the red-wing width of Ca II H and fail to reproduce the exact Ca II H/H$^{-1}$7 relative intensity ratio (Tamburri et al., 17 Feb 2026).

ViSP has also extended into precision scattering-polarization diagnostics. Quiet-Sun observations in the Sr I $^{-1}$8 line show that, at $^{-1}$9, the signal-to-noise ratio is low but sufficient in the total linear polarization map to directly reveal sub-arcsec structures in the Sr I line for the first time, attributable to scattering polarization (Zeuner et al., 2 Mar 2026). At disk center, however, the observations remain dominated by noise related to current setup limitations (Zeuner et al., 2 Mar 2026).

5. Performance envelope and present limitations

The instrument literature states ambitious performance targets: wavelength coverage $90\times340~\mathrm{mm}^2$0, simultaneous observation of three lines, spectral resolving power 180,000, polarimetric sensitivity $90\times340~\mathrm{mm}^2$1, polarimetric accuracy $90\times340~\mathrm{mm}^2$2, and temporal resolution of 10 s to reach $90\times340~\mathrm{mm}^2$3 for $90\times340~\mathrm{mm}^2$4 (Wijn et al., 2022). Commissioning on NOAA AR 12822 reported a continuum noise of about $90\times340~\mathrm{mm}^2$5 for 1.5 s integration, though 7.5 s were required because of camera duty-cycle limitations (Wijn et al., 2022). The ATST design paper likewise emphasizes that full automation, multiple gratings, and rapid reconfiguration were integral to the concept (Wijn et al., 2012).

Operational studies show that performance depends strongly on observing mode and science target. The network-heating observations deliberately traded resolution for areal coverage, using the $90\times340~\mathrm{mm}^2$6 slit and obtaining an effective angular resolution along the slit of about $90\times340~\mathrm{mm}^2$7 after processing (Judge et al., 2023). The flare Ca II H/H$90\times340~\mathrm{mm}^2$8 study achieved spectral sampling of 12.7 m$90\times340~\mathrm{mm}^2$9 in the Fe I arm and 7.7 m$600~\mathrm{nm}$0 in the Ca II H arm, but the achieved slit-direction spatial resolution was only about $600~\mathrm{nm}$1 km because the median Fried parameter was low (Tamburri et al., 17 Feb 2026). In the Sr I scattering-polarization campaign, each slit position was held for 30 s, yet only 13.8 s corresponded to actual exposure because the camera duty cycle was 46%; a dual-beam intensity imbalance of about a factor of 1.7 and known internal ghosting further constrained effective polarimetric precision (Zeuner et al., 2 Mar 2026).

These reports define the current observational boundary conditions. ViSP can already deliver high-spatio-temporal and spectral resolution with exceptional polarimetric precision, and it can resolve sub-arcsecond scattering-polarization structure and fine magnetic morphology (Zeuner et al., 2 Mar 2026, Yadav et al., 2024). At the same time, disk-center scattering polarization in Sr I remains beyond robust direct detection in the present setup, flare ribbon substructure may remain unresolved under poor seeing, and camera duty-cycle limits, residual $600~\mathrm{nm}$2 cross-talk, beam imbalance, and internal ghosts remain practically relevant systematics (Zeuner et al., 2 Mar 2026, Tamburri et al., 17 Feb 2026). A plausible implication is that future gains will depend as much on calibration and detector performance as on raw aperture.

6. Other uses of VISP/VISp

Outside solar physics, VISP appears in several technically unrelated research programs. In workflow technology, VISP stands for Virtual Internet Service Provider, an IST-STREP project aimed at a software platform for telecom/ISP SMEs. Its technology-selection study distinguishes choreography from orchestration, recommends BPMN for early business-process design, BPEL4WS 1.1 and XPDL as low-level workflow/orchestration languages, and treats ebXML BPSS/CPPA as useful within an ebXML-based interoperability stack (0904.3634).

In systems neuroscience, VISp is the Allen Brain Observatory label for mouse primary visual cortex, i.e. V1. Population representations from VISp were compared with VGG16 layers using SSM and SVCCA, and VISp was found to map to an earlier or lower pseudo-depth than other recorded mouse visual areas, while the broader organization appeared more parallel than strictly hierarchical (Shi et al., 2019).

Recent ML/NLP literature reuses the acronym for at least three additional objects: VISP: Volatility Informed Stochastic Projection, an adaptive regularization method guided by gradient volatility; ViSP, a multimodal sarcasm-generation framework using PPO and contrastive learning; and ViSP, a large-scale Vietnamese sentence-paraphrase benchmark (Islam, 2 Sep 2025, Wang et al., 13 Jul 2025, Nguyen et al., 11 Feb 2025). The recurrence of the acronym across these domains is purely nominal rather than conceptual.