A critical evaluation of PCA detection of polarized signatures using real stellar data

TL;DR

This study evaluates the effectiveness of PCA and other analysis techniques in detecting polarized signatures in stellar spectropolarimetric data, using real observations to compare methods for improved post-processing.

Contribution

It provides a comparative analysis of PCA, line addition, and least-squares deconvolution methods for polarized signature detection in stellar spectra, guiding future data processing.

Findings

PCA-based denoising is effective for polarized signature detection.

Comparison clarifies the strengths and limitations of each method.

Results inform optimal techniques for real-time data analysis.

Abstract

The general context of this study concerns the post-processing of multiline spectropolarimetric observations of stars, and in particular these numerical analysis techniques aiming at the detection and the characterization of polarized signatures. Hereafter, using real observational data, we compare and clarify a number of points concerning various methods of analysis. Indeed, simple line addition, least-squares deconvolution and denoising by principal component analysis have been applied, and compared to each other, to polarized stellar spectra available from the TBLegacy database of the Narval spectropolarimeter. Such a comparison between various approaches of distinct sophistication levels allows us to make a safe choice for the next implementation of on-line post-processing of our unique database for the stellar physics community.