Cascade adaptive optics with a second stage based on a Zernike wavefront sensor for exoplanet observations II. Validation in broadband light on the ESO/GHOST testbed

A. Rahim, M. N’Diaye, A. Vigan, M. Kasper, J. Nousiainen, B. Engler, K. Dohlen, S. Leveratto, J. Floriot, M. Marcos, C. Bailet, P. Bristow, E. S. Douglas · 2026

Read Cascade adaptive optics with a second stage based on a Zernike wavefront sensor for exoplanet observations II. Validation in broadband light on the ESO/GHOST testbed by A. Rahim, M. N’Diaye, A. Vigan, M. Kasper, J. Nousiainen, B. Engler, K. Dohlen, S. Leveratto, J. Floriot, M. Marcos, C. Bailet, P. Bristow, E. S. Douglas (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-26895.txt.

Introduction

A. Rahim and twelve colleagues validate a two-stage adaptive optics system for photographing exoplanets. Adaptive optics cancels atmospheric blur with deformable mirrors. Even extreme systems leave residuals because their control loops lag the changing air. A fast second stage can clean up what the first stage misses. The team's second stage uses a Zernike wavefront sensor, a device that reads phase errors from a focused spot.

Earlier validation used single-colour light. Real sensors work in broadband light to gather more photons. Experiments on a European Southern Observatory testbed simulated first-stage leftovers. The loop was tested in narrow and broad bands across seeing conditions, wind speeds, and star brightness levels. The work is the second part of a series aimed at sharper planet images from the ground.

Catalog record

CREATORA. Rahim, M. N’Diaye, A. Vigan, M. Kasper, J. Nousiainen, B. Engler, K. Dohlen, S. Leveratto, J. Floriot, M. Marcos, C. Bailet, P. Bristow, E. S. Douglas
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE66 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.26895

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Cascade adaptive optics with a second stage based on a Zernike wavefront sensor for exoplanet observations II. Validation in broadband light on the ESO/GHOST testbed A. Rahim, M. N’Diaye, A. Vigan, M. Kasper, J. Nousiainen, B. Engler, K. Dohlen, S. Leveratto, J. Floriot, M. Marcos, C. Bailet, P. Bristow, E. S. Douglas Abstract Context. Current high-contrast facilities on the ground use extreme adaptive optics (XAO) systems to achieve contrasts down to 10 − 6 10^{-6} at 200 mas for exoplanet observations. This performance is mainly limited by the XAO residuals due to the temporal errors in the XAO control loop. To overcome this issue, a promising solution consists in using cascade adaptive optics with a fast second stage. This approach was recently validated for a control loop based on a Zernike wavefront sensor (ZWFS) in monochromatic light. Aims. As wavefront sensors operate in broadband light to maximise photon sensitivity, this work aims to validate the ZWFS-based control loop in polychromatic light and assess its performance over a wide range of seeings, wind speeds, and stellar fluxes. Methods. Experiments were conducted on the ESO’s GPU-based High-order adaptive OpticS Testbench (GHOST) testbed to probe our scheme in polychromatic light. Residual aberrations from a first-stage XAO system were simulated and our approach was evaluated in narrowband and broadband light through contrast in Lyot coronagraphic images. Results. Closing the ZWFS-based control loop consistently improves contrast within its correction region in most tested conditions. After subtraction of quasi-static aberrations, our loop reaches a contrast gain up to one order of magnitude, independently of bandwidth and turbulence strength. The broadband and narrowband cases match in performance for bright sources, while narrowband remains slightly preferable for faint targets. Conclusions. These results demonstrate the feasibility of broadband ZWFS-based control loop and underline promising avenues with achromatic masks and an accurate calibration of quasi-static aberrations for future high-contrast imaging on Extremely Large Telescopes. 1 Introduction Understanding the formation and evolution of planetary systems is nowadays one of the key challenges in astrophysics. Direct imaging is a promising approach to provide clues and address this topic, as it both enables the detection of exoplanets and the characterisation of their atmospheres. This technique is inherently limited by the combination of low planet-to-star flux ratios at small angular separations, with contrasts typically ranging from 10^{-4} to 10^{-10} at distances ranging from a few hundreds to a few tens of milliarcseconds in the visible and near infrared ( 12 ; 9 , e.g.,) . These constraints impose severe requirements on high-contrast instrumentation for two complementary classes of facilities: space missions with coronagraphic capabilities and ground-based extremely large telescopes (ELTs). Space-based observatories such as the Nancy Grace Roman Space Telescope (RST) and the envisioned Habitable Worlds Observatory (HWO) will operate in an exceptionally stable environment free from atmospheric turbulence. These missions target planet-to-star contrasts smaller than 10 -8 through a combination of ultra-stable telescopes and coronagraphs with high-precision wavefront control ( 2 ; 30 ) . In parallel, ground-based facilities such as the Extremely Large Telescope (ELT), the Giant Magellan Telescope (GMT), and the Thirty Meter Telescope (TMT) with high-contrast capabilities will benefit from large primary mirrors of 25 to 39 m diameter, providing unprecedented angular resolution and photon collecting power ( 26 ; 5 ; 17 ) . Their apertures typically exceed those of space telescopes (typically 2 to 8 m) since ground-based facilities are not constrained by launch fairing dimensions, mass limits, or the extreme mechanical and thermal stability required for space operation. However, telescopes on the ground suffer from the impact of the atmospheric turbulence on the image quality. Equipped with extreme adaptive optics (XAO) systems to correct for these effects, these telescopes aim to deliver images with Strehl ratios larger than 90% in the near infrared ( 22 , e.g.,) . However, even state-of-the-art XAO instruments on 8 m-class telescopes such as Gemini/GPI, Subaru/SCExAO, and VLT/SPHERE remain limited by residual temporal wavefront errors, linked to correction speed, restricting achievable contrasts to the 10 -4 -10 -6 range at 200 mas ( 29 ; 25 ; 3 ) . Given these limitations, several strategies have been proposed to further reduce the residual aberrations left by the XAO loop. One promising approach is to cascade adaptive optics (AO) systems, so that a second, more sensitive control stage operates downstream of the main XAO system to correct for the remaining small-amplitude wavefront errors ( 6 ; 35 , e.g.,) . To measure these residual aberrations, the second stage requires a wavefront sensor with excellent sensitivity. Among the possible options, the Pyramid Wavefront Sensor ( 37 , PWFS,) and the Zernike Wavefront Sensor ( 42 , ZWFS,) are two of the leading candidates for this role ( 21 ) . The PWFS can be operated in two distinct regimes: modulated or unmodulated. Modulation increases the dynamic range of the sensor at the expense of sensitivity ( 36 ) . For this reason, many current PWFS implementations rely on modulated operation, which facilitates its use under operational conditions but introduces additional complexity, both from an opto-mechanical and a calibration standpoint. This approach is currently being considered in the context of SAXO+, the second-stage adaptive optics system under development for SPHERE ( 4 ) . While modulation is effective in extending the dynamic range, it may prove less relevant for a second-stage correction loop in the presence of low amplitude residual aberrations. The unmodulated PWFS is theoretically very sensitive, but in its standard four-face implementation, it requires the acquisition and processing of four pupil images. This increases the complexity of the data processing chain, as it demands a larger detector area and longer processing times compared to approaches based on a single pupil image. In addition, the unmodulated PWFS presents several limitations, including a light distribution that strongly depends on the incoming phase, a limited intrinsic capture range in the absence of modulation and an optical architecture involving multiple paths and tight alignment tolerances ( 7 ; 1 ) . In contrast, the ZWFS operates with a single pupil image, enabling simple and fast reconstruction. Its response is highly linear in the small aberration regime, and its moderate capture range is well-suited for the measurement of low-amplitude residuals expected after a first stage XAO correction. From an implementation standpoint, the ZWFS relies only on a simple focal-plane phase mask, making a control loop based on the ZWFS compact, stable, and easy to integrate. These characteristics combined with its optimal photon-noise sensitivity ( 21 ; 34 ; 8 , e.g.,) explain the emergence of the ZWFS as a particularly attractive solution for driving the second-stage control loop in cascade adaptive optics architectures. In this context, 35 , hereafter 35 , proposed a second-stage adaptive optics control loop based on a ZWFS and carried out its experimental validation on Graphics Processing Unit (GPU)-based High-order adaptive OpticS Testbench (GHOST), the ESO testbed at Garching, Germany ( 15 ) . This study successfully demonstrated the operation of the control loop in the small aberration regime, thereby experimentally confirming the potential of the ZWFS-based second stage for fine wavefront error correction in cascade adaptive optics architectures. However, this demonstration was performed in monochromatic light at the wavelength \lambda=770 nm, chosen as

Read the complete text