[Paper Review] Simulation of the Far-Infrared Polarimetry Approach Envisioned for the PRIMA Mission

C. D. Dowell, Brandon S. Hensley+1 more|arXiv (Cornell University)|Apr 25, 2024

Geophysics and Gravity MeasurementsEarth and Planetary Sciences3 citations

TL;DR

This paper proposes a novel far-infrared polarimetric imager (PPI) for the PRIMA space mission, using single-polarization Kinetic Inductance Detectors oriented at 120° intervals to measure Stokes I, Q, and U in a single scan, without a half-wave plate. Using an end-to-end simulator with realistic noise and beam effects, the method achieves near-fundamental-limit sensitivity, recovering astrophysical maps with only a 30–40% increase in noise over Gaussian expectations, and exceeds mission sensitivity requirements by 1.5–5×.

ABSTRACT

Interest in the study of magnetic fields and the properties of interstellar dust, explored through increasingly capable far-IR/submillimeter polarimetry, along with maturing detector technology, have set the stage for a transformative leap in polarization mapping capability using a cryogenic space telescope. We describe the approach pursued by the proposed Probe far-Infrared Mission for Astrophysics (PRIMA) to make ultra-deep maps of intensity and polarization in four bands in the 91-232 micron range. A simple, polarimetry-optimized PRIMA Polarimetric Imager (PPI) is designed for this purpose, consisting of arrays of single-polarization Kinetic Inductance Detectors oriented with three angles which allow measurement of Stokes I, Q, and U in single scans. In this study, we develop an end-to-end observation simulator to perform a realistic test of the approach for the case of mapping a nearby galaxy. The observations take advantage of a beam-steering mirror to perform efficient, two-dimensional, crossing scans. Map making is based on 'destriping' approaches demonstrated for Herschel/SPIRE and Planck. Taking worst-case assumptions for detector sensitivity including 1/f noise, we find excellent recovery of simulated input astrophysical maps, with I, Q, and U detected at near fundamental limits. We describe how PPI performs detector relative calibration and mitigates the key systematic effects to accomplish PRIMA polarization science goals.

Motivation & Objective

• To develop and test a novel, detector-optimized polarimetric imaging approach for the PRIMA space mission using single-polarization Kinetic Inductance Detectors.
• To evaluate the performance of a beam-steering-based scanning strategy for efficient two-dimensional mapping of Stokes parameters.
• To assess the robustness of the method against key systematic effects such as detector 1/f noise, beam shape variations, and polarization cross-talk.
• To demonstrate that the approach can achieve sensitivity close to the fundamental noise limit despite pessimistic assumptions on detector performance.
• To validate that the method satisfies or exceeds PRIMA’s science requirements for deep, high-fidelity polarization mapping of nearby galaxies and star-forming regions.

Proposed method

• An end-to-end observation simulator is developed to model the full signal chain from sky input to reconstructed maps, incorporating realistic detector noise, including 1/f noise.
• The PPI instrument uses three arrays of single-polarization detectors oriented at 120° to one another, enabling simultaneous measurement of Stokes I, Q, and U in a single scan.
• A beam-steering mirror enables efficient, two-dimensional crossing scans, improving mapping speed and signal-to-noise efficiency.
• Map-making is performed using a 'destriping' approach adapted from Herschel and Planck, which removes low-frequency signal drifts and baseline variations.
• Systematic effects such as beam shape differences (width and ellipticity) and cross-polar response are modeled and mitigated via focal plane design and a convolution-based data pipeline step.
• The reconstruction uses a least-squares fitting method to solve for I, Q, and U maps, with beam-matching corrections applied before final solution.

Figure 1: Illustration of detector $1/f$ noise model (Section 4.1 ) and baseline subtraction by segments. The lower curve shows a noise model realization for one detector, for the first 16 seconds of an observation. The detectors are sampled at $f_{samp}=$ 350 Hz. The gaps in the curve are the turn-

Experimental results

Research questions

• RQ1 Can a single-polarization detector array with 120° orientation sampling achieve high-fidelity polarization mapping without a half-wave plate, under realistic noise conditions?
• RQ2 How well does the beam-steering scanning strategy perform in recovering Stokes parameters with minimal noise amplification and systematic errors?
• RQ3 To what extent do beam shape variations (width, ellipticity) and cross-polar response affect polarization fidelity, and can they be mitigated effectively?
• RQ4 Does the least-squares reconstruction method with destriping achieve sensitivity close to the fundamental noise limit despite 1/f noise and detector non-idealities?
• RQ5 Can the method exceed PRIMA’s required sensitivity levels for deep polarization mapping of nearby galaxies and star-forming regions?

Key findings

• The simulated PPI approach recovers input astrophysical maps with excellent fidelity, achieving near-fundamental-limit sensitivity even under pessimistic noise assumptions.
• The reconstructed maps show only a 30–40% increase in noise level compared to the pure Gaussian expectation, indicating high efficiency and robustness.
• For a 1.5-hour observation of NGC 6946, polarized emission is detected with high signal-to-noise in both the galaxy and cirrus foreground, demonstrating feasibility for deep surveys.
• The method exceeds PRIMA’s sensitivity requirements by a factor of 1.5 to 5, depending on instrument and scan pattern assumptions, indicating strong performance headroom.
• Systematic effects such as beam shape differences and cross-polar response are shown to produce small residuals (≤0.5% peak in polarized intensity) and are effectively mitigated via focal plane design and pipeline corrections.
• The simulation confirms that the absence of a half-wave plate does not compromise performance, and the method remains efficient and accurate under realistic conditions.

Figure 2: Focal plane layout of PPI4. 132 microlens-fed kinetic inductance detectors are arranged in a hexagonal-close-packed configuration with nearest-neighbor spacing of 25″ on the sky. The arrows show the direction of sensing linear polarization for each detector, dictated by the KID antenna des