The hunt for Sirius Ab: Comparison of algorithmic sky and PSF estimation performance in deep coronagraphic thermal-IR high contrast imaging

TL;DR

This paper compares algorithmic sky and PSF estimation techniques in deep coronagraphic thermal-IR imaging of Sirius Ab, aiming to improve detection sensitivity for faint companions using advanced data analysis methods.

Contribution

It introduces a novel comparison of dimensionality reduction and PSF alignment techniques for high-contrast imaging data analysis.

Findings

Enhanced background illumination modeling using dimensionality reduction.

Improved sub-pixel alignment with a physical-optics-based PSF model.

Potential for increased sensitivity to faint planetary signals.

Abstract

Despite promising astrometric signals, to date there has been no success in direct imaging of a hypothesized third member of the Sirius system. Using the Clio instrument and MagAO adaptive optics system on the Magellan Clay 6.5 m telescope, we have obtained extensive imagery of Sirius through a vector apodizing phase plate (vAPP) coronagraph in a narrowband filter at 3.9 microns. The vAPP coronagraph and MagAO allow us to be sensitive to planets much less massive than the limits set by previous non-detections. However, analysis of these data presents challenges due to the target's brightness and unique characteristics of the instrument. We present a comparison of dimensionality reduction techniques to construct background illumination maps for the whole detector using the areas of the detector that are not dominated by starlight. Additionally, we describe a procedure for sub-pixel alignment of vAPP data using a physical-optics-based model of the coronagraphic PSF.