Layered dust / B-modes

Preserved original model output. Check claims against the independent audit →

summary · 2.5 KB

Literature anchors (orchestrator notes, not a model essay)

Used later to fact-check overlap/unique claims. Not shown to the three models during independent or drive waves.

Decorrelation / LOS mixing

  • Tassis & Pavlidou 2015: misaligned magnetic fields + different cloud SEDs along the LOS → frequency-dependent polarization angle.
  • Planck Int. L 2017: evidence for spatial variation of polarized dust SED; toy models with Gaussian β_d or angle variations; possible bias on r.
  • Planck 2018 XI: mean polarized dust β_d^P ≈ 1.53; no strong evidence for frequency decorrelation on large sky fractions; sub-orbital patches less constrained.
  • Sheehy & Slosar 2018; later Planck XI: no significant BB decorrelation in some analyses (tension with Int. L).
  • Pelgrims et al. 2021: detected LOS frequency decorrelation toward multi-velocity HI sightlines; ~10% cloud-to-cloud polarized SED variation 217–353 GHz.
  • Ritacco et al. 2023 (A&A 672, A21): residual maps after mean SED subtraction; first-order moment expansion of MBB fits EE residuals.
  • Hensley, Clark, et al. 2022/2023 (arXiv:2207.14213): superposition of MBBs; CMB Q,U biased only if both SED and polarization angle change along the LOS; fitted T_d, β_d unrepresentative of distribution means.

3D dust / B-field models

  • Gaia/stellar-extinction 3D dust maps: Green/Bayestar, Leike, Edenhofer, Zucker, DECaPS southern plane, etc. These map extinction/density, not directly polarized microwave emission.
  • Martínez-Solaeche, Karakci, Huffenberger 2017 (arXiv:1706.04162): 3D polarized dust model with LOS variation of density, β_d, T_d.
  • Vansyngel et al. 2017: layered GMF + turbulence to match Planck EE/BB/TE.
  • DustFilaments (arXiv / ApJ 2022): full-sky filament model imperfectly aligned with B, Jansson & Farrar large-scale field.

Grain physics

  • Polarization from aspherical grains aligned with B (RAT alignment; polarization orthogonal to B_POS).
  • Polarized SED can differ from total-intensity SED (size-dependent alignment; Draine & Fraisse 2009 and later Hensley–Draine models).
  • Magnetic dipole emission / composition (Hensley & Bull 2018) as extra SED complexity.

Component separation implications

  • Single-pixel MBB fails when parameters vary in the beam or along the LOS → effective SED is a superposition, not an MBB (moment expansion: Chluba, Azzoni et al. 2022 for LiteBIRD).
  • LiteBIRD-scale r ~ 10^{-3}: Azzoni et al. find β and T moments both needed or r is biased.
  • HI velocity-coherent clouds as a proxy for LOS layers (Clark, Pelgrims).