🔭 astrophysics

The Roman Coronagraph Community Participation Program: Observation planning and data reduction for polarimetric mode

This paper outlines the observation planning, simulations, and data reduction procedures for the Roman Coronagraph Instrument's polarimetric mode, demonstrating their efficacy through a test case involving polarimetric observations of the debris disk HD 172555.

Original authors: Ramya M Anche, Toshiyuki Mizuki, Justin Hom, Alexis Lau, Saanika Choudhary, Jaren N. Ashcraft, Clarissa Do O, Tsutsumi Nagai, Sophie Noiret, Eric Shen, Taichi Uyama, Chen Xie, Jingwen Zhang, Vanessa P

Published 2026-08-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Ramya M Anche, Toshiyuki Mizuki, Justin Hom, Alexis Lau, Saanika Choudhary, Jaren N. Ashcraft, Clarissa Do O, Tsutsumi Nagai, Sophie Noiret, Eric Shen, Taichi Uyama, Chen Xie, Jingwen Zhang, Vanessa P. Bailey, Eric Cady, Jessica Gersh-Range, Julien H. Girard, Guillermo Gonzalez, John Livingston, Bertrand Mennesson, Maxwell A. Millar-Blanchaer, Julia Milton, Naoshi Murakami, Dmitry Savransky, Motohide Tamura, Jason J. Wang, Schuyler G. Wolff, Marie Ygouf

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Deep in the quiet dark between the stars, dust swirls in vast, flat disks around distant suns. These are not the dusty rooms of a house, but cosmic graveyards of planet formation, where leftover rocks and ice clump together or get ground down into fine powder. For decades, astronomers have tried to understand what these disks are made of and how they behave, but the dust is stubborn. It scatters starlight in ways that look very similar whether the grains are large or small, smooth or jagged. This similarity creates a confusing blur, making it nearly impossible to tell the true nature of the dust just by looking at how bright the disk appears. To solve this puzzle, scientists need a new kind of eye, one that can see not just the brightness of the light, but the direction in which the light waves are vibrating. This direction is called polarization. By measuring how the light is polarized, researchers can finally separate the different properties of the dust grains, revealing the hidden architecture of these distant planetary nurseries.

A team of researchers has now laid out the blueprint for how the Nancy Grace Roman Space Telescope will use this technique to study the universe. Scheduled for launch in August 2026, the telescope carries a special instrument called the Roman Coronagraph, designed to block the blinding glare of distant stars so that faint companions and disks can be seen. While the instrument's main job is to find and study planets, it also has a "best-effort" mode dedicated to polarimetry, the measurement of polarized light. In a new paper, the team explains exactly how they plan to use this mode, how they will prepare for the observations, and how they will turn the raw data into scientific results. They have built a complete simulation of the entire process, from the moment the telescope points at a target to the final analysis of the data, demonstrating that the instrument's simulation and data reduction workflow can successfully process polarized light signals from debris disks.

The team chose a real star system, HD 172555, to test their methods. This system is home to a bright, warm disk of debris that has been studied before, making it a perfect candidate for a trial run. The researchers used a sophisticated computer program to create a fake version of what the telescope would see if it pointed at this star. They started by modeling the disk itself, calculating how light would bounce off the dust grains to create a specific pattern of polarization. Then, they fed this model into a simulation of the Roman Coronagraph's optics. The simulation included every possible source of noise and error that the real instrument might face, such as the way the telescope's mirrors slightly alter the light's polarization and the tiny imperfections in the detectors. They also simulated the necessary calibration steps, which involve observing other stars to understand the instrument's behavior and subtracting out the star's own light to reveal the faint disk.

Once the fake data was generated, the team ran it through their data reduction pipeline, a set of software tools designed to clean up the images and extract the scientific information. This process involves stripping away the noise, correcting for the telescope's quirks, and combining the different views of the light to calculate the final polarization values. The result was a clean, three-dimensional map of the light's properties, showing not just how bright the disk is, but how the light is vibrating at every point. The simulation showed that the instrument's workflow could successfully separate the four different components of polarized light needed to build this map. It also demonstrated that the team could accurately measure the polarization fraction, which is a key number describing how much of the light is polarized, even in the presence of the instrument's own noise.

The paper details the specific steps the team will take when the telescope is actually in space. The observation plan is intricate, requiring the telescope to look at the target star and a nearby reference star from different angles. By rotating the telescope and taking pictures through special prisms that split the light, they can capture four distinct views of the same scene. These views are then combined to cancel out the overwhelming glare of the star and isolate the faint signal from the dust. The team also calculated how long they would need to stare at a target to get a good result. Their estimates show that for a typical debris disk, the telescope could gather enough data to make a clear measurement in a reasonable amount of time, ranging from a few hours to a few days depending on the brightness of the target and the strength of the polarization signal.

This work is a crucial step forward because it validates the observation planning and data reduction pipeline for the Roman Coronagraph's polarimetric mode. Before this paper, the polarimetric mode was a theoretical possibility; now, the team has shown that the software and the observation plans work together smoothly. They have identified the specific calibration products needed, such as maps of the instrument's internal polarization effects, and they have shown how to generate them. While the current simulations used placeholder data for some of these calibrations, the team has a clear path to replacing them with real data once the instrument is tested on the ground. The ultimate goal is to use these tools to map the linear polarization of debris disks around other stars, providing a level of detail that has never been achieved before. By doing so, they hope to unlock the secrets of how planets form and what these distant worlds are made of, turning the faint, scattered light of cosmic dust into a clear story of planetary evolution.

Technical Summary: The Roman Coronagraph Community Participation Program – Observation Planning and Data Reduction for Polarimetric Mode

Problem Statement
The detection and characterization of exoplanets and circumstellar debris disks via reflected light polarimetry offer critical constraints on atmospheric properties and dust-grain characteristics that are often degenerate in total intensity observations alone. While the Nancy Grace Roman Space Telescope's Coronagraph Instrument (RCI) is primarily designed to meet Technology Threshold Requirement 5 (TTR5) for high-contrast imaging in Band 1 using the Hybrid Lyot Coronagraph (HLC), it possesses a "best-effort" polarimetric mode. This mode utilizes two Wollaston prisms (POL0 and POL45) to split incoming light into four orthogonal components ( $I_0, I_{90}, I_{45}, I_{135}$ ). However, effectively utilizing this mode requires robust observation planning, accurate exposure time estimation, and a specialized data reduction pipeline capable of handling complex polarization effects, including instrument-induced Mueller matrix crosstalk and reference star differential imaging (RDI). Prior to this work, the community lacked a comprehensive framework for simulating and reducing polarimetric data for the RCI, specifically for extended sources like debris disks.

Methodology
The paper outlines a complete workflow for the polarimetric mode, encompassing observation sequencing, exposure time estimation, end-to-end simulation, and data reduction.

Observation Sequence: The proposed observing campaign involves alternating visits to a target star (e.g., HD 172555) and a reference star (e.g., $\alpha$ Ara) at multiple roll angles. Both stars are observed in both Wollaston configurations (POL0 and POL45). The sequence includes the acquisition of satellite spots for alignment and unocculted observations of both stars to estimate and subtract host star polarization.
Exposure Time Estimation: The authors incorporated polarimetric capabilities into the corgietc exposure time calculator. The Signal-to-Noise Ratio (SNR) calculation accounts for shot noise, detector noise, residual speckle noise, and a polarization speckle contrast term ( $C_{pab}$ ). The required integration time is derived as a function of the target's polarization fraction ( $p$ ) and the desired polarization SNR ( $SNR_{pol}$ ).
Simulation (corgisim): A high-fidelity end-to-end simulation package, corgisim, was used to generate mock Level 1 (L1) data. For debris disks, the input is a 3D Stokes cube ( $I, Q, U, V$ ) generated by the radiative transfer code MCFOST. This cube is scaled to the target's magnitude, multiplied by modeled instrument Mueller matrices to simulate polarization effects, convolved with the instrument point spread function (PSF), and finally processed through the Wollaston prism matrices to generate the four orthogonal detector images.
Data Reduction (corgidrp): The corgidrp pipeline processes L1 data through a sequence of levels (L1 $\to$ $\to$ L2a $\to$ $\to$ L2b $\to$ $\to$ L3 $\to$ $\to$ L4 $\to$ $\to$ TDA). Key steps include:
- L1 to L2a/b: Standard corrections (bias, nonlinearity, cosmic rays) followed by division by a polarimetric flatfield (currently a mock uniform array) and conversion to electrons.
- L2b to L3: Separation of POL0 and POL45 images into the four Stokes components ( $I_0, I_{90}, I_{45}, I_{135}$ ), extraction of subarrays, and normalization by exposure time.
- L3 to L4: Alignment of frames, subtraction of host star polarization (derived from unocculted observations), and construction of the Stokes cube. A Mueller matrix calibration is applied to correct for instrument-induced crosstalk.
- L4 to TDA: PSF subtraction is applied to the calibrated Stokes cube to derive final science products, including polarization fractions, position angles, and azimuthal components ( $Q_\phi, U_\phi$ ) for disks.

Key Contributions and Results
The paper presents the first integrated overview of the polarimetric mode for the Roman Coronagraph, specifically demonstrating the workflow using the debris disk HD 172555 as a test case.

Observation Planning Framework: The authors defined a specific observing sequence that integrates reference star vetting, multiple roll angles, and specific calibration steps (flatfields, Mueller matrix, absolute flux, and host star polarization) necessary for high-precision polarimetry.
Exposure Time Models: New SNR equations were formulated to estimate integration times for various polarization fractions, demonstrating the trade-offs between target brightness, contrast, and required integration time for both HLC (Narrow FOV, Band 1) and SPC (Wide FOV, Band 4) modes.
End-to-End Demonstration: The authors successfully generated a mock dataset for HD 172555 using corgisim and processed it through corgidrp.
- The simulation successfully modeled the generation of the four orthogonal components and the subsequent reconstruction of the Stokes parameters.
- The reduction pipeline demonstrated the ability to handle the separation of polarization states and the application of Mueller matrix corrections.
- A specific result noted is that while the input MCFOST model for HD 172555 had zero circular polarization ( $V=0$ ), the reduced output showed non-zero $V$ . The paper attributes this to crosstalk between $U$ and $V$ introduced by the modeled instrument Mueller matrices, validating the necessity of the calibration step.
Calibration Strategy: The paper details the four distinct calibrations required for the polarimetric mode, distinguishing them from standard imaging calibrations. It notes that while the current demonstration uses mock calibration products (uniform flatfields and synthetic Mueller matrices), future updates will replace these with simulated products derived from corgisim and eventually real data from Thermal Vacuum (TVAC) tests.

Significance
The paper establishes the foundational infrastructure for the Roman Coronagraph Community Participation Program (CPP) regarding polarimetry. It provides the necessary tools (corgisim and corgidrp) and methodologies for the community to plan observations and process data for the instrument's best-effort polarimetric mode. By validating the simulation and reduction pipeline with a representative debris disk target, the authors demonstrate that the RCI can effectively characterize dust grain properties and planetary system architectures through multi-wavelength polarimetry. This work supports the scientific goal of mapping linear polarization in circumstellar debris disks with high precision, a capability that extends the mission's utility beyond its primary Technology Threshold Requirement. The paper modestly claims that these initial tests serve as a proof-of-concept, with future iterations expected to incorporate more realistic calibration products as the mission approaches its 2026 launch readiness.

More like this