The point-spread function (PSF) of the Extreme Ultraviolet High-Resolution Imager (HRIEUV) on board ESA/NASA's Solar Orbiter, derived by Hofmeister et al. (2026) from 126 images of the 2023 January 3 Mercury transit, shows that a combined 56.6% of the incoming light is redistributed across the detector — by diffraction from the meshes and mounting supporting the entrance and filter-wheel filters and by diffuse scattering from mirror microroughness (the two processes overlap, since light can be both diffracted and scattered, so the combined redistributed fraction is less than the sum of the two component fractions). The derivation, however, is explicitly incomplete: the mesh diffraction parameters could not be refined beyond mechanical specifications (no flare showed multiple diffraction orders), the uniqueness of the diffuse-scattering inversion from Mercury's small disk is not guaranteed, a possible ~0.5% anisotropic component could neither be confirmed nor excluded, scattered-light removal shows position-dependent errors up to 13% of unexplained origin (spatially variant PSF versus residual light leak), and the PSFs of the other two EUI channels — the Lyman-alpha high-resolution imager and the Full Sun Imager — remain uncharacterized. This problem asks to complete the instrumental scattered-light characterization of EUI: refine and validate the HRIEUV PSF on all four open points and extend PSF characterization to the remaining channels.

Background

Solar Orbiter, launched in 2020, carries the Extreme Ultraviolet Imager (EUI), a suite of three telescopes observing the Sun in extreme ultraviolet (EUV) and ultraviolet wavelengths (Rochus et al. 2020): HRIEUV, a high-resolution imager of the 174 Å coronal line; HRILya, a high-resolution imager of the hydrogen Lyman-alpha line at 121.6 nm; and FSI, the Full Sun Imager, which observes 174 Å and 304 Å with a wide field of view for context and off-limb coronal studies. A point-spread function (PSF) describes how an instrument redistributes the light from a point-like source over the detector; in EUV solar telescopes it has two dominant components. First, the thin-film entrance filters must be supported by a periodic wire mesh and by a spider mounting (a radially armed mechanical support structure at the entrance aperture), and this periodic structure diffracts a substantial fraction of the light into a pattern whose peak spacing is set by the mesh pitch dd (the wire center-to-center spacing) and whose envelope is controlled by the mesh window width ww (pitch minus wire width). Second, residual microroughness of the mirror surfaces — of order 5 Å root-mean-square for HRIEUV, comparable to the 174 Å wavelength — diffusely scatters light over the whole detector. Together these effects suppress contrast and corrupt photometry of both bright and faint structures, so accurate science requires deconvolving images with a well-determined PSF. Hofmeister, Hahn and Savin (2022) developed a semi-blind inversion method that recovers a PSF from partially occulted images: pixels occulted by a moving body contain, a priori, only instrumental scattered light, so the observed intensities there constrain the PSF directly. Applying it to the 2023 January 3 Mercury transit (126 images, on-disk and off-limb), Hofmeister et al. (2026) found that HRIEUV diffracts about 26.4% of the incoming light (from the meshes supporting the entrance and filter-wheel filters and from the entrance spider mount), diffusely scatters about 41.1% (well described by a softened power law a (b+Δr)−ca\,(b+\Delta r)^{-c} in angular distance Δr\Delta r), and that deconvolution brightens bright structures by up to 40% and darkens dark ones by up to 85%. The same team had earlier derived revised PSFs for the Atmospheric Imaging Assembly (AIA) on board NASA's Solar Dynamics Observatory (Hofmeister, Savin and Hahn 2025), where the analysis was stronger on exactly the points that remain open for EUI: AIA's mesh diffraction was calibrated from flare images with multiple visible diffraction orders, and its diffuse scattering was fitted using large-scale lunar occultations, which cleanly separate long-distance (PSF tail) from short-distance (PSF core) scattering. For Solar Orbiter no such data exist: no observed flare was strong enough to show several diffraction orders in HRIEUV, and no large-scale (lunar) occultation is available from the spacecraft's vantage point, so the small angular disk of Mercury had to serve, entangling core and tail. The EUI analysis is therefore a detector-averaged, isotropic, and possibly non-unique approximation, and the authors list its unresolved issues explicitly.

Problem Statement

Complete the instrumental scattered-light (PSF) characterization of the Solar Orbiter/EUI telescopes left open by Hofmeister et al. (2026), whose common objective is a validated PSF description for each EUI channel. The five numbered strands below are independently standing problems, each resolvable on its own, and the characterization is complete only once all five have been addressed (either branch of strand (2) discharges that strand). Specifically: (1) refine the HRIEUV diffraction PSF beyond the mechanical drawings — the mesh window width and pitch, and the orientations and spacings of the diffraction patterns; (2) settle the uniqueness of the recovered diffuse-scattering function: either prove uniqueness (or quantify the degenerate solution manifold) of the semi-blind inversion from partially occulted images under the realistic noise and dark-current model, or determine the scattering function unambiguously; (3) confirm or rule out a small anisotropic component of the HRIEUV PSF, currently bounded only at the level of roughly 0.5% elliptical stretching; (4) determine whether the up-to-13% position-dependent error in scattered-light removal arises from a spatially variant PSF or from an instrumental artifact such as a residual light leak — that is, whether a detector-averaged PSF suffices or a spatially resolved PSF is required; and (5) characterize the PSFs of the remaining EUI channels, HRILya (whose own instrumental limitations block the Mercury-transit analysis) and FSI (whose spatial resolution is insufficient for it, though it shares HRIEUV's mesh geometry).

References

  1. Hofmeister, S. J., Kraaikamp, E., Shestov, S., Teriaca, L., Koukras, A., Verbeeck, C., Auchère, F., Savin, D. W., Hahn, M., & Berghmans, D. (2026). A Point-spread Function for the Extreme Ultraviolet High-resolution Imager on Board Solar Orbiter. The Astrophysical Journal Supplement Series, 286, 3. https://doi.org/10.3847/1538-4365/ae8996 (preprint: arXiv:2605.01607, https://arxiv.org/abs/2605.01607)
  2. Hofmeister, S., Savin, D. W., & Hahn, M. (2025). Revised Point-spread Functions for the Atmospheric Imaging Assembly on board the Solar Dynamics Observatory. The Astrophysical Journal Supplement Series, 278, 8. https://doi.org/10.3847/1538-4365/adbaed (preprint: arXiv:2410.08967, https://arxiv.org/abs/2410.08967)
  3. Hofmeister, S. J., Hahn, M., & Savin, D. W. (2022). Deriving instrumental point spread functions from partially occulted images. Journal of the Optical Society of America A, 39(12), 2153. https://doi.org/10.1364/josaa.471477 (preprint: arXiv:2207.03487, https://arxiv.org/abs/2207.03487)
  4. Hofmeister, S. J. (2024). The Basic Iterative Deconvolution: A Fast Instrumental Point-Spread Function Deconvolution Method That Corrects for Light That Is Scattered Out of the Field of View of a Detector. Solar Physics, 299, 77. https://doi.org/10.1007/s11207-024-02321-y (preprint: arXiv:2312.11784, https://arxiv.org/abs/2312.11784)
  5. Rochus, P., Auchère, F., Berghmans, D., et al. (2020). The Solar Orbiter EUI instrument: The Extreme Ultraviolet Imager. Astronomy & Astrophysics, 642, A8. https://doi.org/10.1051/0004-6361/201936663