Inferring Magnetic Field Morphology and Dust Scattering Geometry from Mid-IR Polarimetry: the Extended Aitken Method

TL;DR

The paper extends the Aitken method for mid-IR polarimetry by incorporating a graphite-scattering polarization template, enabling a three-component decomposition into absorptive, emissive, and scattering contributions. Applying this to CanariCam data for the Egg Nebula, W3 IRS5, and W51 IRS2 shows that scattering can dominate in some sightlines and that including it yields more accurate emissive/absorptive components and PA relations, particularly producing a PA difference between absorptive and emissive components that peaks near perpendicular, consistent with magnetic-field-aligned elongated grains. Where scattering is negligible, the extended method converges to the original Aitken results, underscoring its compatibility and robustness. The approach provides a practical framework for disentangling polarization mechanisms across at least three wavelengths bracketing the silicate feature, enabling improved magnetic field morphology inferences in diverse dust environments.

Abstract

The Aitken method is a useful approach for decomposing mid-IR polarimetry of silicates in astronomical sources into emissive and absorptive components. Here we extend this method to include the effects of polarization caused by scattering from graphite or similar particles along the same sightlines. To demonstrate the extended method, we apply it in the analysis of CanariCam multi-wavelength imaging polarimetry observations of the Egg Nebula, W3 IRS5, and W51 IRS2, and also spectropolarimetry of W3 IRS5. We compare these results with those obtained with the original Aitken method and show that the Egg Nebula observations are fit better when this third component is incorporated into the analysis. Polarimetry observations of W3 IRS5 are also fit better with the extended Aitken method, but the original method suffices to fit many sightlines. Observations of W51 IRS2 are fit well by either the original or extended Aitken method. Including scattering by dust in the decomposition of polarimetry observations of the Egg Nebula and W3 IRS5 produces better results for the emissive and absorptive components, and in particular for the position angle (PA) of those components. The distribution of the difference between absorptive and emissive PA is then found to be more peaked at a single angle, nearly perpendicular. This supports the theory that mid-IR polarization arises from elongated dust grains aligned along magnetic field lines, since then the PA of emissive and absorptive polarization would be perpendicular. When significant scattering is not present the extended method produces the same results as the original method.

Figures (27)

• Figure 1: The dashed red curve is the absorptive profile, dot-dashed blue curve is the emissive profile, and the short-dashed violet curve is the scattering polarization profile of the extended Aitken method.
• Figure 2: (a) Observed intensity of the Egg Nebula with the N-band filter, in units of Jy/arcsec$^2$, indicated by the logarithmic color scale and over-plotted with N-band polarization vectors. (b) polarization image over-plotted with same vectors indicating polarization and PA, but with linear color scale indicating percent polarization. Images of Stokes parameters are first smoothed with $3 \times 3$ pixel boxcar to reduce noise before computing polarization and PA. Data is shown where SNR$>$10. Note tilt of field of view (FOV) with respect to north.
• Figure 3: Results of the extended Aitken method fits at 8.7 $\mu$m for the Egg Nebula: (a) observed polarization; (b) scattering component; (c) absorptive component; (d) emissive component rotated by 90 to show alignment with possible B-field. Linear color scales indicate the polarization magnitudes p(%), and over-plotted vectors show the PA and polarization. Images of Stokes parameters are first smoothed with $3 \times 3$-pixel boxcar, followed by computation of the polarization (%) and PA. Data are shown where SNR$>$10 for all wavelengths. Note the tilt of FOV with respect to north. Circled cross symbols are sightlines at which details of fits are presented in Figures \ref{['Egg-AE+EAS-PeakPolo']} and \ref{['Egg-AE+EAS-SPL']}.
• Figure 4: Results of the original Aitken method fits at 8.7 $\mu$m for the Egg Nebula: (a) absorptive component; (b) emissive component rotated by 90 to show alignment with possible B-field. Linear color scales indicate the polarization magnitudes p(%), and over-plotted vectors show the PA and polarization. Images of Stokes parameters are first smoothed with $3 \times 3$-pixel boxcar, followed by computation of the polarization (%) and PA. Data are shown where SNR$>$10 for all wavelengths. Note the tilt of FOV with respect to north. Circled cross symbols are sightlines for which details of fits and comparison with the extended method are presented in Figures \ref{['Egg-AE+EAS-PeakPolo']} and \ref{['Egg-AE+EAS-SPL']}.
• Figure 5: Original and extended Aitken method fits to multi-wavelength imaging polarimetry of Egg Nebula at the sightline marked with circled cross symbol to the right of origin in images, with offset (+21, -07). Observed Stokes $q$ and $u$ are plotted as diamonds with error bars (averages and uncertainty in 024$\times$024 apertures). Bottom row of plots show the % polarization and PA computed from the Stokes spectra. Left column of plots show the fits with two components of the original Aitken method: absorptive (red dashed) and emissive (blue dot-dashed). Right column of plots show the extended Aitken method fits that also include scattering (violet short-dashed curves). Solid black curves are the sum of components, with gray shading indicating the range of 1-sigma uncertainty.