Moriond Talks - Day 1

CMB Blind component separation - Alessandro¶

• ILC component separation $\to$ remove foreground

• Problems:

  • Complex spatial structure

  • Complex spectral structure

• Need a component separation with minimal assumptions.

  • ILC: minimize the variance, need the covariance of the data

    $$\mathbf{w} = [A_{CMB}^T C^{-1}A_{CMB}]^{-1} A_{CMB}^T C^{-1}$$

    (1)

  • MultiClustering-ILC (MC-ILC) $\to$ improve the fit on the large scales.

  • Generalized-ILC $\to$ improve on the noise, fit for $F$ instead of A

    $$A = F + N$$

    (2)

• Check the biases on different $r$ values and foreground models

• Include the template of foregrounds into likelihood.

• Broom packages

• Blind detection of systematic residuals after component separation.

• Frontier?

  • Apply to SO data $\to$ transfer function. (some preliminary results)

  • Apply Broom to Planck data.

Questions¶

• Why using templates ? Can not model statistically ?

  • Alessandro: even using the templates it wouldn’t leak into the variance ?

Spectral Distortion, Bayronic feedback, FIRAS - Fabbian Guillio¶

Spectral Distortion from perfect blackbody of CMB. Distortions when energy is injected or redistributed and the photon do not have enough time to thermolize.

• Compton scattering (before recombination)

• SZ effect (after recombination)

The last experiment is COBE (1990s) - FIRAS. Can we do better with the modern techniques? (170 channels)

• Original efforts

  • Correlations between different components $\to$ need to be careful when do separation.

  • Original: template fitting $\to$ prone to error because they use very high-frequency extrapolation

  • Refit: adding the distortion.

• Current efforts

  • Foregrounds 100x distortions

  • Monopole modeling, MCMC

Results

• $\langle y \rangle$-distortions measurements (new measurement after 30 years)

• reduce the impact of foreground residuals.

Why it is important?

• probe of gas thermodynamics (AGN/SN feedback)

• physics of the gas in the dark matter halo

• Compare with BAO, weak lensing

• Modeling with hydrodynamic simulation

• Constraint on galaxy formation models

  • Direct constrains are weak ? $\to$ likelihood-free inference

• Galaxy lensing ????

Spectral distortion future missions, FOSSIL - Bruno Maffei¶

Status

• Proposal: PIXIE (NASA), PRISTINE (ESA), FOSSIL (ESA)

Goals of FOSSIL

• $\mu$-distortion: probing density perturbations + dark sectors

  • Little red dots, BH mergers

• $y$-distortion: ....

• Star formation + dust

• Instruments: more frequency bands (30-2000 GHz) $\to$ understand the foreground much better ?

Plans

• Launch 2041

Questions¶

The Simons Observatory - Daniel Dutcher¶

• LAT, SAT

• LF 30/40 GHz, MF 90/150 GHz, UHF 220/280

Science goals

• Tensor to scalar ratio $\to$ primordial GWs, inflation

• Non-gaussianity

• Neutrino mass

• Cluster catalog

• LAT $\to$ start to take data

  • 6m telescope

  • LAT receiver - the largest cryogenic mm-wave camera.

  • Optics Tube is much powerful than ACT

• SAT $\to$ already have data

  • 3 instruments with 2 MF, 1 UHF

  • 12000 detectors in each instruments

  • Half-wave plate

  • TES detector (go-to for 20 years) $\to$ improve on the sensitivity ?

Initial data: 90/150 GHz SATs

• Few months of data $\to$ but very good agreement with other experiments

LATs

• detector performance >= sensitivity targets in all bands

• Lensing 600 deg$^2$

Results

• Got the temperature map in 180 hours comparable with ACT in 5 years

  • see the points sources (good resolution)

Upgrades

• 13 optic tubes: 30000 $\to$ 60000 (NSF grant) (Jan 2026)

• 3 more SATs: 1 LF, 2 MF: 24000 $\to$ 48000

Commander4, End-to-end Joint Analysis - Ana Isabel¶

• Cosmoglobe: multi-experiments (break degeneracy thanks to same physics, different instrument setting) + loop analysis (use the constraints to prior for analysis)

• BeyondPlanck: Gibbs sampling $\to$ cosmo + nuisance parameters (explore the whole parameter space)

  • WMAP + Planck: DR1

  • DR2: IR

• Commander3 (Fortran) single node $\to$ Commander4 (Python) multiple nodes DRX:

• AKARI (very high frequency, high resolution) $\to$ improve the calibration

• COBE-FIRAX: recalibration the CMB temperature

• OpenHFI

There are infinite possibilities.

Questions:¶

• Can it combine with LSS ?

• For two different experiments: the nuisance parameters will be different and the cosmological parameters will be shared ?

• Parametric models ???

Cosmic birefringence, SPT-3G¶

Cosmic birefringence = EB correlation in CMB polarizations

• Chern-Simons coupling (axion)

• Primordial magnetic fields

Many constraints in the past few years

• Two ways to measure anisotropic cosmic birefringence

  • B-mode modeling

  • Quadratic estimator

Results:

• Consistent with no cosmic birefringence for now

• Put a new upper limit: 0.017