REVIEW 3 major objections 5 minor 26 references
Gaussian deformable-mirror probes beat Roman’s baseline sinc probes for electric-field estimation, and new vector vortex coronagraphs reach high contrast on an upgraded testbed.
Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →
T0 review · grok-4.5
2026-07-31 20:29 UTC pith:VPS7W7G4
load-bearing objection Solid facility paper with two real lab results after a genuine software upgrade; the Gaussian-probe claim is motivation for a planned Roman flight test, not a demonstrated CGI result, and the abstract oversells the VVC floors. the 3 major comments →
First high-contrast results on THD2 testbed after infrastructure upgrade
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
After the control infrastructure upgrade, the testbed recovered monochromatic dark-hole performance near 10^-8–10^-9 and then delivered two advances: Gaussian-shaped pair-wise probes reduce nonlinear contributions relative to the Roman baseline sinc–sinc–sine family under comparable conditions, improving estimation efficiency and motivating their use as an enhanced early Roman observation; and charge-6 double-grating vector vortex coronagraphs achieve mean normalized intensities of order 2–3×10^-8 in ~1% bandwidth and ~10^-7 in 20% bandwidth with active wavefront control.
What carries the argument
Pair-wise probing with electric-field conjugation: opposite deformable-mirror commands create intensity differences from which the coherent focal-plane field is estimated under a linear approximation; probe morphology (Gaussian versus sinusoidally modulated sinc) sets how large the neglected nonlinear terms become as amplitude rises.
Load-bearing premise
The operational claim for Roman assumes that a simplified in-air Roman-like coronagraph ranks probe nonlinearities the same way the real space instrument will.
What would settle it
An on-orbit Roman comparison in which Gaussian probes give no cleaner electric-field estimates and no reduction in images or time needed relative to the baseline sinc probes would refute the claimed efficiency gain.
If this is right
- Gaussian probes are prioritized for an enhanced early Roman Coronagraph observation comparing estimation accuracy and efficiency in flight.
- Lower higher-order probe error should also improve coherent differential imaging that separates residual starlight from incoherent signals.
- The tested liquid-crystal vortex designs become concrete broadband candidates for future space high-contrast instruments.
- A common open control layer lets different coronagraphs and wavefront-control methods share calibration, diagnostics, and closed-loop workflows on one facility.
- The platform remains a European in-air path to contrasts of 10^-8–10^-9 for maturing technologies aimed at deeper future missions.
Where Pith is reading between the lines
- If nonlinear probe error dominates estimation bias, reshaping the whole probe library around smooth localized commands could save more ground-in-the-loop time than incremental algorithm tweaks alone.
- The same Gaussian-versus-sinc nonlinearity ranking should appear on any system that relies on the linear pair-wise intensity model, not only hybrid Lyot layouts.
- Ghosts and polarization leakage in the vortex runs imply that mask manufacturing alone will not reach 10^-10 without concurrent control of stray light and polarization.
- Shared, version-controlled testbed software may transfer reproducibility across facilities as much as any single optical upgrade.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript reports the migration of the THD2 high-contrast testbed at LIRA/Paris Observatory to the open-source CATKit2 control framework, and presents two science cases enabled by the upgrade. Validation with the baseline FQPM configuration at 783 nm recovers a mean normalized intensity of ~9.8e-9 in a 3–11 λ/D dark hole (~5e-9 in 5–10 λ/D) after 19 EFC iterations, demonstrating continuity of optical performance. Science case I reviews Roman-like HLC work (~2e-8 in closed loop) and argues that Gaussian-shaped pair-wise probes yield a smaller fraction of non-linear error than the CGI baseline sinc–sinc–sine probes at equal focal-plane intensity, motivating a planned alternative-probe campaign on Roman. Science case II reports SUPPPPRESS liquid-crystal vortex coronagraphs: the charge-6 dgVVC reaches 2.8e-8, 6.1e-8, and 1.3e-7 mean normalized intensity at 1%, 10%, and 20% bandwidth in a one-sided scoring region that excludes ghost reflections.
Significance. If the reported numbers stand, the paper is a useful and timely facility record: THD2 is currently the only operable Roman-like HLC testbed after the JPL decommissioning, so its results carry real weight for Roman CGI observing preparation. Strengths worth crediting: empirical closed-loop contrasts with loop diagnostics rather than simulations; an open-source, version-controlled control stack (CATKit2 / thd-controls) that improves reproducibility; broadband coronagraph measurements up to 20% bandwidth, which remain rare; and a concrete, falsifiable comparison plan (in-flight probe comparison after commissioning). The Gaussian-probe nonlinearity result, once backed by inspectable data, would be of direct operational interest to the Roman CPP. The dgVVC broadband numbers are honestly reported, including the ghost-contamination caveat, which is good practice.
major comments (3)
- [§3.3.2 / abstract] §3.3.2 and abstract: the paper's strongest scientific claim — Gaussian probes yield a lower non-linear-error fraction than the baseline sinc–sinc–sine probes at equal focal-plane probe intensity — rests entirely on Eq. 13 and Fig. 2 of Delaye et al. 2026 (ref. 20), an 'in these proceedings' item that is not inspectable. This manuscript itself presents no figure, table, or number for the comparison, and the in-text statement is qualified ('lower ... than two of the three baseline ... probes over the investigated range'), with the amplitude range and uncertainties unstated. Since the abstract asserts that these results 'led to their prioritization' for Roman, the manuscript should either reproduce the key comparison figure/quantitative summary here (including amplitude range and error bars, and which baseline probe was not outperformed), or explicitly downgrade the abstract claim to a poin
- [Abstract vs. §4.2] Abstract vs. §4.2: the abstract states the VVCs are designed 'to high-contrast performance approaching 10^-10 over a >20% bandwidth', but the best reported dgVVC measurement at 20% bandwidth is 1.3e-7 (§4.2, Fig. 4c), and the best overall result is 2.8e-8 at 1% bandwidth. As written, a reader could take 10^-10-class performance at 20% bandwidth as a demonstrated THD2 result. Please rephrase the abstract to distinguish the component design goal (described in ref. 16) from the system-level contrasts actually measured here.
- [§3.3.2–3.3.3] §3.3.2–3.3.3: the probe-nonlinearity ranking is obtained on THD2's 32x32 Boston Micromachines DM, with sinc commands 'adapted from the Roman DM geometry' (§3.3.1). The higher-order terms being compared depend on actuator count, influence functions, and surface-fitting error of the probe commands; CGI's 48x48 Xinetics DMs differ on all three, so a Gaussian smoothness margin measured here does not automatically transfer to flight. The planned in-flight comparison (§3.3.3) is the right eventual test, but the manuscript should state this DM-dependence caveat explicitly — ideally with a simulation check using Xinetics influence functions, or at minimum a sentence acknowledging that the ranking could change with DM surface-fitting behavior — before citing the result as motivation for the Roman observing priority.
minor comments (5)
- [Fig. 2 / §4.2] Fig. 2 caption gives the FQPM DH as 3–11 λ/D while §4.2 and Fig. 3 use 3–10 λ/D; please make the dark-hole definitions consistent or note that they are configuration-dependent.
- [Figs. 1, 3; §2.3] Fig. 3 caption: 'with it convergence plit' — presumably 'with its convergence plot'. Also 'testdbed' in Fig. 1 and 'Boston Michromachines' in §2.3 (Micromachines); 'SuperK Fianium' capitalization.
- [§4.2] §4.2: the one-sided scoring region is well justified (fixed ghosts), but please state what fraction of the full annulus it covers, so readers can gauge how the quoted 2.8e-8/6.1e-8/1.3e-7 would shift under a full-annulus metric.
- [§2.4 / §4.2] Reported contrasts carry no stability information. A brief statement of temporal hold time or repeatability of the DH floor (e.g., drift of the 9.8e-9 FQPM level over minutes/hours) would substantially strengthen the 'improved stability' framing of the upgrade.
- [§3.3.1] §1.3: ref. 13 (Potier et al. 2020) is cited for both the <1e-8 monochromatic result and the single-actuator probe claim in §3.3.1; please verify the latter attribution, since Laginja et al. (2025) is the more natural source for probe-family comparisons.
Circularity Check
Empirical testbed paper; no by-construction circularity. Minor companion-paper dependence for the Gaussian-probe nonlinearity numbers does not force the measured contrasts.
specific steps
-
self citation load bearing
[§3.3.2; abstract claim on Gaussian probes]
"At equal focal-plane probe intensity, the Gaussian probes produce a lower fraction of non-linear error than two of the three baseline sinc–sinc–sine probes over the investigated range. This comparison suggests that the additional phase modulation by a sine/cosine on the sinc probes introduces higher-order terms without providing a corresponding increase in useful estimated signal. (see also Fig. 2 in Delaye et al. 2026; Eq. 13 in Delaye et al. 2026)"
The paper’s strongest scientific claim (Gaussians reduce non-linearities vs Roman baseline sinc probes, motivating Roman prioritization) is quantitatively justified only by citing Delaye et al. 2026—a same-proceedings companion with overlapping authors—for the PW expansion and the linear/non-linear fraction. This text gives no independent figure, table, or equation for that comparison. This is mild load-bearing self-citation of uninspectable companion detail, not a by-construction identity; the underlying THD2 measurements remain empirical.
full rationale
This is a facility/instrumentation paper reporting laboratory closed-loop contrasts and hardware comparisons, not a first-principles derivation. The FQPM validation (mean ~9.8e-9 in the DH; ~5e-9 between 5–10 λ/D), the Roman-like HLC DH (~2e-8), and the dgVVC broadband intensities (2.8e-8 / 6.1e-8 / 1.3e-7 at 1%/10%/20% bandwidth) are direct measurements on THD2; they are not fitted parameters renamed as predictions, nor quantities defined in terms of themselves. The Gaussian-vs-sinc claim cites Delaye et al. 2026 (overlapping authors, ‘in these proceedings’) for the linear/non-linear fraction comparison (their Eq. 13) and does not reproduce that expansion or a figure here—so the quantitative ranking is not self-contained in this text—but that is ordinary companion-paper dependence for an overview, not a reduction of the result to its inputs by construction. Roman baseline probe shapes are an external mission definition, not an author-fitted ansatz. No uniqueness theorem, no self-definitional scale, and no renaming of a known empirical law appear. Score 1 reflects only light load-bearing self-citation on the probe-nonlinearity detail; the experimental content stands independently.
Axiom & Free-Parameter Ledger
free parameters (4)
- Probe amplitude / focal-plane probe intensity =
Not numerically fixed in this text; set per experiment
- One-sided DH scoring region (ghost exclusion) =
Dashed contour in Fig. 4; exact pixels not tabulated
- Dark-hole annulus (approx. 3–10 or 3–11 λ/D) =
3–11 λ/D (FQPM); ~3–10 λ/D (HLC/VVC)
- Central wavelength and fractional bandwidths =
λ≈780–783 nm; BW ∈ {1%,5%,10%,20%}
axioms (4)
- domain assumption Pair-wise probing difference intensities estimate the coherent focal-plane field under a first-order expansion; residual higher-order terms quantify probe quality.
- domain assumption Mean normalized intensity inside a defined dark hole is the primary performance metric for coronagraph + WFS&C success.
- ad hoc to paper THD2 in-air Roman-like HLC and VVC setups are adequate proxies for ranking flight-relevant probe shapes and mask behavior.
- domain assumption Fixed-location ghosts from polarizers/QWPs can be excluded from the scoring region without biasing the coronagraph leakage claim.
read the original abstract
We present the first scientific results delivered by the upgraded THD2 high-contrast imaging testbed. We report two advances enabled by its improved stability and broadband performance. First, for the Roman Space Telescope, we demonstrate that Gaussian-shaped diversity probes outperform the baseline sinc probes by reducing non-linearities, supporting higher probe amplitudes, and improving electric field estimation efficiency. These results have led to their prioritization as an enhanced early observation for Roman. Second, within ESA's SUPPPPRESS project, we test new polarization-independent Vector Vortex Coronagraphs and design them to high-contrast performance approaching 1e-10 over a 20% bandwidth. We assess their behavior in narrow- and broadband light with active focal-plane wavefront control. Together, these results show how THD2 strengthens Europe's capability in high-contrast imaging, providing a unique platform reaching contrasts of 1e-8 to 1e-9 for developing next-generation coronagraphic technologies.
Figures
Reference graph
Works this paper leans on
-
[1]
An Overview of Exoplanet Biosignatures,
Schwieterman, E. W. and Leung, M., “An Overview of Exoplanet Biosignatures,”Reviews in Mineralogy and Geochemistry90, 465–514 (July 2024)
2024
-
[2]
The Roman Space Telescope coronagraph technology demonstration: current status and relevance to future missions,
Mennesson, B., Bailey, V. P., Zellem, R., Hildebrandt, S., Ygouf, M., Rhodes, J., Zimmerman, N., Nemati, B., Gonzalez, G., Cady, E., Kern, B., Koch, T., Krist, J., Heydorff, K., Luchik, T., Mok, F., Morrissey, P., Poberezhskiy, I., Riggs, A. J., Shi, F., Zhao, F., Akeson, R., Armus, L., Greenbaum, A., Ingalls, J., and Lowrance, P., “The Roman Space Telesc...
2022
-
[3]
Feinberg, L. D., Sitarski, B. N., McElwain, M. W., Arney, G., Baker, C., Bolcar, M. R., Levine, M., Liu, A., Mennesson, B., Roberge, A., Smith, J. S., Zhao, F., and Ziemer, J., “Habitable Worlds Observatory’s Concept and Technology Maturation: Initial Feasibility and Trade Space Exploration,”arXiv e-prints, arXiv:2601.11803 (Jan. 2026)
arXiv 2026
-
[4]
Polarization effects on high contrast imaging: measurements on THD2 bench,
Baudoz, P., Desgrange, C., Galicher, R., and Laginja, I., “Polarization effects on high contrast imaging: measurements on THD2 bench,” in [Proc. SPIE],13092, SPIE (2024). In these proceedings
2024
-
[5]
Optimization and performance of multi-deformable mirror correction on the THD2 bench,
Baudoz, P., Galicher, R., Potier, A., Dupuis, O., Thijs, S., and Patru, F., “Optimization and performance of multi-deformable mirror correction on the THD2 bench,” in [Advances in Optical and Mechanical Technolo- gies for Telescopes and Instrumentation III], Navarro, R. and Geyl, R., eds.,10706, 784 – 793, International Society for Optics and Photonics, S...
2018
-
[6]
Closed loop, dm diversity-based, wavefront correction algorithm for high contrast imaging systems,
Give’on, A., Belikov, R., Shaklan, S., and Kasdin, J., “Closed loop, dm diversity-based, wavefront correction algorithm for high contrast imaging systems,”Opt. Express15, 12338–12343 (Sep 2007)
2007
-
[7]
The self-coherent camera: a new tool for planet detection,
Baudoz, P., Boccaletti, A., Baudrand, J., and Rouan, D., “The self-coherent camera: a new tool for planet detection,”Proceedings of the International Astronomical Union1(C200), 553–558 (2005)
2005
-
[8]
Implicit electric field conjugation: Data-driven focal plane control,
Haffert, S. Y., Males, J. R., Ahn, K., Van Gorkom, K., Guyon, O., Close, L. M., Long, J. D., Hedglen, A. D., Schatz, L., Kautz, M., Lumbres, J., Rodack, A., Knight, J. M., and Miller, K., “Implicit electric field conjugation: Data-driven focal plane control,”A&A673, A28 (2023)
2023
-
[9]
Development and characterization of Four-Quadrant Phase Mask coronagraph (FQPM),
Bonafous, M., Galicher, R., Baudoz, P., Firminy, J., and Boussaha, F., “Development and characterization of Four-Quadrant Phase Mask coronagraph (FQPM),” in [Advances in Optical and Mechanical Technologies for Telescopes and Instrumentation II], Navarro, R. and Burge, J. H., eds.,9912, 99126J, International Society for Optics and Photonics, SPIE (2016)
2016
-
[10]
Complex apodization Lyot coronagraphy for the direct imaging of exoplanet systems: design, fabrication, and laboratory demonstration,
Trauger, J., Moody, D., Gordon, B., Krist, J., and Mawet, D., “Complex apodization Lyot coronagraphy for the direct imaging of exoplanet systems: design, fabrication, and laboratory demonstration,” in [Space Telescopes and Instrumentation 2012: Optical, Infrared, and Millimeter Wave], Clampin, M. C., Fazio, G. G., MacEwen, H. A., and Jr., J. M. O., eds.,8...
2012
-
[11]
A coronagraph with a band-limited mask for finding terrestrial planets,
Kuchner, M. J. and Traub, W. A., “A coronagraph with a band-limited mask for finding terrestrial planets,” The Astrophysical Journal570(2), 900 (2002)
2002
-
[12]
Optical vectorial vortex coronagraphs using liquid crystal polymers: theory, manufacturing and laboratory demon- stration,
Mawet, D., Serabyn, E., Liewer, K., Hanot, C., McEldowney, S., Shemo, D., and O’Brien, N., “Optical vectorial vortex coronagraphs using liquid crystal polymers: theory, manufacturing and laboratory demon- stration,”Opt. Express17, 1902–1918 (Feb 2009)
1902
-
[13]
Comparing focal plane wavefront control techniques: Numerical simulations and laboratory experiments,
Potier, A., Baudoz, P., Galicher, R., Singh, G., and Boccaletti, A., “Comparing focal plane wavefront control techniques: Numerical simulations and laboratory experiments,”Astron. & Astrophys.635, A192 (Mar. 2020)
2020
-
[14]
Extended linearity in the high-order wavefront sensor for the roman coronagraph,
Laginja, Iva, Baudoz, Pierre, Mazoyer, Johan, Potier, Axel, Galicher, Rapha¨ el, and Boussaha, Faouzi, “Extended linearity in the high-order wavefront sensor for the roman coronagraph,”A&A698, A130 (2025)
2025
-
[15]
spacetelescope/catkit2: v0.7.0,
Por, E. H., Laginja, I., Steiger, S., Soummer, R., Meier, L., Pougheon, E., Pourcelot, R., Philbin, M. P., Mazoyer, J., Sevin, A., Nguyen, M., Sahoo, A., Lau, A., Paviot, C., Egger, L., Delaye, L., Meredith, A., Lim, P. L., linarphy, and Fowler, J., “spacetelescope/catkit2: v0.7.0,” (June 2026)
2026
-
[16]
SUPPPPRESS: prototyping and testing liquid crystal vector vortex coronagraphs with reduced polarization leakage,
Landman, R., Doelman, D., Rietjens, J., Laginja, I., Baudoz, P., Peeters, K., van Dijk, C., Nishie, Y., Watanabe, Y., Eigenraam, A., Vretenar, M., van den Born, J., Galicher, R., Mazoyer, J., Potier, A., Krasteva, M., Taccola, M., Bettonvil, F., and Snik, F., “SUPPPPRESS: prototyping and testing liquid crystal vector vortex coronagraphs with reduced polar...
2026
-
[17]
CATKit2- HCI: a collaborative framework for advancing high-contrast coronagraph testbeds,
Soummer, R., Laginja, I., Por, E. H., Pourcelot, R., Steiger, S., Nickson, B., Lau, A., Baudoz, P., Do ´O, C., Fowler, J., Galicher, R., Groff, T. D., Jensen-Clem, R., Mawet, D. P., Mazoyer, J., McElwain, M. W., Meier, L., Potier, A., Pueyo, L. A., Redmond, S. F., Will, S. D., and Zimmerman, N. T., “CATKit2- HCI: a collaborative framework for advancing hi...
2026
-
[18]
Roman space telescope coronagraph instrument: software architecture, development, and verification,
Fluckiger, K., Koch, T., Cady, E., Mazer, A., DeMartino, A., Schoolcraft, J., Schneider, R., Miller, D., Hauss, J., Bowman, N., Umana, M., Sarkissian, T., Milton, J., and Soliman, A., “Roman space telescope coronagraph instrument: software architecture, development, and verification,”Journal of Astronomical Telescopes, Instruments, and Systems11(2), 021412 (2025)
2025
-
[19]
High-order wavefront sensing and control for the Roman Coronagraph Instrument (CGI): architecture and measured performance,
Cady, E., Bowman, N., Greenbaum, A. Z., Ingalls, J. G., Kern, B., Krist, J., Marx, D., Poberezhskiy, I., Eldorado Riggs, A. J., Ruane, G., Seo, B.-J., Shi, F., and Zhou, H., “High-order wavefront sensing and control for the Roman Coronagraph Instrument (CGI): architecture and measured performance,”Journal of Astronomical Telescopes, Instruments, and Syste...
2025
-
[20]
Enhanced wavefront sensing for the Ro- man Coronagraph Instrument: Gaussian probes and compact model validation,
Delaye, L., Laginja, I., Baudoz, P., Potier, A., Mazoyer, J., Galicher, R., Redmond, S. F., Lau, A., Riggs, A. J. E., Sirbu, D., Por, E. H., Soummer, R., and Pueyo, L., “Enhanced wavefront sensing for the Ro- man Coronagraph Instrument: Gaussian probes and compact model validation,” in [Proceedings of SPIE Astronomical Telescopes + Instrumentation], 14145...
2026
-
[21]
Pair-wise, deformable mirror, image plane-based diversity electric field estimation for high contrast coronagraphy,
Give’on, A., Kern, B. D., and Shaklan, S., “Pair-wise, deformable mirror, image plane-based diversity electric field estimation for high contrast coronagraphy,” in [Techniques and Instrumentation for Detection of Exoplanets V], Shaklan, S., ed.,8151, 376 – 385, International Society for Optics and Photonics, SPIE (2011)
2011
-
[22]
Methods and limitations of focal plane sensing, estimation, and control in high-contrast imaging,
Groff, T. D., Riggs, A. J. E., Kern, B., and Kasdin, N. J., “Methods and limitations of focal plane sensing, estimation, and control in high-contrast imaging,”Journal of Astronomical Telescopes, Instruments, and Systems2(1), 1 – 15 (2015)
2015
-
[23]
Coherent differential imaging on VLT/SPHERE: first results and beyond ,
Potier, A., Desgrange, C., Galicher, R., Mazoyer, J., Baudoz, P., and Wahhaj, Z., “Coherent differential imaging on VLT/SPHERE: first results and beyond ,” in [Proceedings of SPIE Astronomical Telescopes + Instrumentation], 14150–54, SPIE (2026). In these proceedings
2026
-
[24]
SUPPPPRESS: developing low-leakage liquid-crystal vector vortex coronagraphs for space-based exoplanet imaging,
Landman, R., Doelman, D., Rietjens, J., Laginja, I., Baudoz, P., Peeters, K., van Dijk, C., Nishie, Y., Watanabe, Y., van den Born, J., Eigenraam, A., Vretenar, M., Krasteva, M., Taccola, M., and Snik, F., “SUPPPPRESS: developing low-leakage liquid-crystal vector vortex coronagraphs for space-based exoplanet imaging,” in [Proceedings of SPIE Astronomical ...
2026
-
[25]
Advancing European High-Contrast Imaging R&D towards the Habitable Worlds Observatory,
Laginja, I., Carri´ on-Gonz´ alez,´O., Laugier, R., et al., “Advancing European High-Contrast Imaging R&D towards the Habitable Worlds Observatory,”Astrophysics and Space Science370, 29 (2025)
2025
-
[26]
Continuing to Advance European High Contrast Imaging Research and Development towards HWO and LIFE,
Chauvin, G., Carrion Gonzalez, O., Laginja, I., Dicken, D., Haffert, S., Kasper, M., Absil, O., Kammerer, J., Potier, A., Le Coroller, H., Matthews, E., Laugier, R., Defrere, D., Hansen, J., Krause, O., de Visser, P., Karlsson, M., Henning, T., Ferrari, M., Kuehn, J., Janson, M., Zhao, F., and Desgrange, C., “Continuing to Advance European High Contrast I...
Pith/arXiv arXiv 2026
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.