Mathieu Remazeilles

Residual foreground contamination in thermal Sunyaev-Zeldovich (SZ) Compton- y parameter maps ( y -maps) arises mainly from Galactic emissions — thermal dust and synchrotron radiation — on large angular scales, and from cosmic infrared background (CIB) anisotropies on small scales. Unlike the thermal SZ effect, Galactic foregrounds are strongly polarized. Exploiting this distinction, we introduce a hybrid Needlet Internal Linear Combination (Hybrid NILC) method that combines Planck total-intensity and polarization frequency maps in the component-separation pipeline, thereby improving the suppression of residual Galactic emission while preserving the unpolarized SZ signal by leveraging the intrinsic TE and TB correlations of thermal dust and synchrotron. Using Planck PR4 data, we find that the Hybrid NILC y -map exhibits about 40 % lower cross-correlation with the IRAS dust tracer than the standard temperature-only Planck y -map, indicating reduced residual Galactic contamination. Simulations further indicate that, for future high-sensitivity surveys such as LiteBIRD , the Hybrid NILC will become increasingly effective at suppressing Galactic residuals. We further address small-scale extragalactic contamination by selectively deprojecting specific moments of the CIB using a Constrained Hybrid NILC variant, achieving an improved balance between CIB suppression and noise penalty compared to previous implementations in the literature. These novel approaches — particularly the joint use of temperature and polarization in component separation — offer a powerful framework for disentangling polarized and unpolarized signals.

We present updated forecasts for the scientific performance of the degree-scale (0.5 deg FWHM at 93 GHz), deep-field survey to be conducted by the Simons Observatory (SO). By 2027, the SO Small Aperture Telescope (SAT) complement will be doubled from three to six telescopes, including a doubling of the detector count in the 93 GHz and 145 GHz channels to 48,160 detectors. Combined with a planned extension of the survey duration to 2035, this expansion will significantly enhance SO's search for a B -mode signal in the polarisation of the cosmic microwave background, a potential signature of gravitational waves produced in the very early Universe. Assuming a 1/ f noise model with knee multipole ℓ _knee = 50 and a moderately complex model for Galactic foregrounds, we forecast a 1 σ (or 68% confidence level) constraint on the tensor-to-scalar ratio r of σ _r = 1.2 × 10 ^-3 , assuming no primordial B-modes are present. This forecast assumes that 70% of the B -mode lensing signal can ultimately be removed using high resolution observations from the SO Large Aperture Telescope (LAT) and overlapping large-scale structure surveys. For more optimistic assumptions regarding foregrounds and noise, and assuming the same level of delensing, this forecast constraint improves to σ _r = 7 × 10 ^-4 . These forecasts represent a major improvement in SO's constraining power, being a factor of around 2.5 times better than what could be achieved with the originally planned campaign, which assumed the existing three SATs would conduct a five-year survey.

We introduce a novel approach to estimate the spectral index, β _s , of polarised synchrotron emission, combining the moment expansion of Cosmic Microwave Background foregrounds and the constrained Internal Linear Combination method. We reconstructed the maps of the first two synchrotron moments, combining multi-frequency data, and applied the ‘T-T plot’ technique between two moment maps to estimate the synchrotron spectral index. This approach offers a new technique for mapping the foreground spectral parameters, complementing the model-based parametric component separation methods. Applying this technique, we derived a new constraint on the spectral index of polarised synchrotron emission using QUIJOTE MFI wide-survey 11 and 13 GHz data, Wilkinson Microwave Anisotropy Probe data at K and Ka bands, and Planck LFI 30 GHz data. In the Galactic plane and North Polar Spur regions, we obtained an inverse-variance-weighted mean synchrotron index of β _s = −3.11 with a standard deviation of 0.21 due to intrinsic scatter, consistent with previous results based on parametric methods using the same dataset. We find that the inverse-variance-weighted mean spectral index, including both statistical and systematic uncertainties, is β _s ^plane = −3.05 ± 0.01 in the Galactic plane and β _s ^high-lat = −3.13 ± 0.02 at high latitudes, indicating a moderate steepening of the spectral index from low to high Galactic latitudes. Our analysis indicates that, within the current upper limit on the Anomalous Microwave Emission polarisation fraction, our results are not subject to any appreciable bias. Furthermore, we infer the spectral index over the entire QUIJOTE survey region, partitioning the sky into 21 patches. This technique can be further extended to constrain the synchrotron spectral curvature by reconstructing higher-order moments when better-quality data become available.

Intensity Mapping (IM) of the 21-cm line of the neutral hydrogen ( Hi ) has become a compelling new technique to map the large-scale structure of the Universe. One of the main challenges is the presence of strong foreground emissions of several orders of magnitude larger than the Hi signal. Here, we implement a version of the Principal Component Analysis, a blind component-separation technique, based on a kind of spherical wavelets called needlets. These functions exploit double localization both in real and in harmonic space. We test Need-PCA performances on a set of maps that simulates the SKA MID radio telescope in the AA4 configuration. We compare our results with other component separation methods such as Generalised Morphological Component Analysis (GMCA) and Generalized Needlet Internal Linear Combination (GNILC). All the methods have comparable results, recovering the Hi signal within 10% accuracy across the frequency channels, in the multipole range 30 ≲ ℓ ≲ 136. We also test our pipeline in the presence of systematics such as polarization leakage. We find that the cleaning methods are insensitive to the presence of such systematic, yielding the same results as in the leakage-free case. Finally, under the assumption of a realistic telescope beam with sidelobes, we find that standard PCA and GMCA fails to recover the Hi signal at larger scales, while the Need-PCA and Need-GMCA are less affected. GNILC tends to over-clean, yielding to a loss of the signal.

Cosmic birefringence, arising from a potential parity-violating interaction between cosmic microwave background (CMB) photons and evolving pseudo-scalar fields such as axion-like particles, can rotate the CMB polarization plane and induce an effective correlation between the CMB E - and B -mode polarization. In this work, we introduce a hybrid internal linear combination (ILC) method that combines both E - and B -mode frequency maps into the component separation pipeline, enabling the disentanglement of correlated and uncorrelated components of CMB polarization in the presence of cosmic birefringence and instrumental polarization angle miscalibration. We derive an analytic linear relation connecting the birefringence-induced correlated component of the CMB E - (or B -) mode field to the full CMB B - (or E -) mode field convolved with a modulating field. By performing linear regression between these fields across multiple sky patches, we directly estimate the birefringence angle at the field level. This allows us to distinguish cosmic birefringence from polarization angle miscalibration and foreground contamination, as the ILC responds differently to achromatic cosmic birefringence and chromatic systematic effects, with its weights projecting spatial or harmonic dependence only onto the latter. This non-parametric, field-level approach provides a novel way to probe cosmic birefringence directly in real space. When applied to realistic simulations of the forthcoming LiteBIRD satellite mission, our method yields constraints that are competitive with, and complementary to, existing power spectrum-based analyses. When applied to Planck Release 4 (PR4) data, we find a birefringence angle of β = 0.32° ± 0.12°, a 2.7 σ detection that remains robust against varying sky fractions.

We investigate how dust foreground complexity can affect measurements of the tensor-to-scalar ratio, r , in the context of the Simons Observatory, using a cross-spectrum component separation analysis. Employing a suite of simulations with realistic Galactic dust emission, we find that spatial variation in the dust frequency spectrum, parametrized by β _d , can bias the estimate for r when modeled using a low-order moment expansion to capture this spatial variation. While this approach performs well across a broad range of dust complexity, the bias increases with more extreme spatial variation in dust frequency spectrum, reaching as high as r ∼ 0.03 for simulations with no primordial tensors and a spatial dispersion of σ ( β _d ) ≃ 0.3 — the most extreme case considered, yet still consistent with current observational constraints. This bias is driven by changes in the ℓ-dependence of the dust power spectrum as a function of frequency that can mimic a primordial B -mode tensor signal. Although low-order moment expansions fail to capture the full effect when the spatial variations of β _d become large and highly non-Gaussian, our results show that extended parametric methods can still recover unbiased estimates of r under a wide range of dust complexities. We further find that the bias in r , at the highest degrees of dust complexity, is largely insensitive to the spatial structure of the dust amplitude and is instead dominated by spatial correlations between β _d and dust amplitude, particularly at higher orders. If β _d does spatially vary at the highest levels investigated here, we would expect to use more flexible foreground models to achieve an unbiased constraint on r for the noise levels anticipated from the Simons Observatory.

Polarized foreground emission from the Galaxy is one of the biggest challenges facing current and upcoming cosmic microwave background (CMB) polarization experiments. We develop new models of polarized Galactic dust and synchrotron emission at CMB frequencies that draw on the latest observational constraints; that employ the “polarization fraction tensor” framework to couple intensity and polarization in a physically motivated way; and that allow for stochastic realizations of small-scale structure at subarcminute angular scales currently unconstrained by full-sky data. We implement these models into the publicly available Python Sky Model (PySM) software and additionally provide PySM interfaces to select models of dust and CO emission from the literature. We characterize the behavior of each model by quantitatively comparing it to observational constraints in both maps and power spectra, demonstrating an overall improvement over previous PySM models. Finally, we synthesize models of the various Galactic foreground components into a coherent suite of three plausible microwave skies that span a range of astrophysical complexity allowed by current data. Author contributions to this paper can be found at the end of this work.

The variation of the physical conditions across the three dimensions of our Galaxy is a major source of complexity for the modelling of the foreground signal facing the cosmic microwave background (CMB). In the present work, we demonstrate that the spin-moment expansion formalism provides a powerful framework to model and understand this complexity, and we put special focus on the effects that arise from variations of the physical conditions along each line of sight on the sky. We performed the first application of the moment expansion to reproduce a thermal dust model largely used by the CMB community, demonstrating its power as a minimal tool to compress, understand, and model the information contained within any foreground model. Furthermore, we used this framework to produce new models of thermal dust emission containing the maximal amount of complexity allowed by the current data while remaining compatible with the observed angular power spectra by the Planck mission. By assessing the impact of these models on the performance of component separation methodologies, we conclude that the additional complexity contained within the third dimension could represent a significant challenge for future CMB experiments and that different component separation approaches are sensitive to different properties of the moments.