[Paper Review] Simulations and Observations of the Microwave Universe

M. Peel|arXiv (Cornell University)|Jun 14, 2010

Galaxies: Formation, Evolution, Phenomena167 references3 citations

TL;DR

This thesis presents end-to-end simulations and observational analyses of the microwave sky, focusing on the Sunyaev-Zel'dovich (SZ) effect from galaxy clusters and point source contamination. Using N-body and Pinocchio simulations, it models the SZ power spectrum and validates with OCRA-p receiver observations, demonstrating that point sources significantly contribute to high-multipole power excess and that accurate modeling of cluster physics and source distributions is essential for CMB experiments.

ABSTRACT

[Abridged] Simulations and observations of the microwave sky are of great importance for understanding the Universe that we reside in. Specifically, knowledge of the CMB and its foregrounds - including the SZ effect from clusters of galaxies and radio point sources - tell us about the Universe on its very largest scales, and also what the Universe is made of. We describe the creation of software to carry out large numbers of virtual sky simulations. The simulations include the CMB, SZ effect and point sources, and are designed to examine the effects of point sources and the SZ effect on present and recent observations of the CMB. Utilizing sets of 1,000 simulations, we find that the power spectrum resulting from the SZ effect is expected to have a larger standard deviation by a factor of 3 than would be expected from purely Gaussian realizations, and is significantly skewed towards increased values for the power spectrum. The effects of the clustering of galaxy clusters, residual point sources and uncertainties in the gas physics are also investigated, as are the implications for the excess power measured in the CMB power spectrum by the CBI and BIMA. We carry out end-to-end simulations for OCRA-p observations of point sources. The introduction of simulated 1/ f noise significantly reduces the predicted ability of the instruments to observe weak sources by measuring the sources for long periods of time. The OCRA-p receiver has been used to observe point sources in the VSA fields so that they can be subtracted from observations of the CMB power spectrum. We find that these point sources are split between steep and flat spectrum sources. We have also observed 550 CRATES flat spectrum radio sources, which will be useful for comparison to Planck satellite observations. Finally, the assembly and commissioning of the OCRA-F receiver is outlined. [Abridged]

Motivation & Objective

• To model the microwave sky's statistical properties, particularly the Sunyaev-Zel'dovich (SZ) effect from galaxy clusters and point source contamination.
• To develop and validate a virtual sky simulation framework using N-body and Pinocchio simulations to predict power spectra and source distributions.
• To analyze real OCRA-p receiver data to calibrate point source flux densities and spectral indices, improving source confusion modeling.
• To quantify the impact of cluster mass function, gas profiles, and cosmological parameters (e.g., σ₈) on the SZ power spectrum.
• To assess the role of point sources in the high-multipole CMB power spectrum excess, especially at 30 GHz.

Proposed method

• Employed the Pinocchio algorithm to generate light-cone catalogs of galaxy clusters with mass and redshift distributions consistent with cosmological simulations.
• Simulated the thermal SZ effect using analytical models of cluster gas profiles and the Y-M relation, incorporating redshift and mass-dependent corrections.
• Generated full-sky realizations of the CMB and SZ effect power spectra using theoretical and simulated transfer functions.
• Modeled low- and high-frequency point sources using differential source counts, spectral indices, and polarization statistics from surveys (e.g., CRATES, NVSS, WMAP).
• Conducted end-to-end simulations of the One Centimetre Receiver Array (OCRA) using the UMBRELLA software to simulate cross-scan and on-off observations.
• Validated simulated noise and signal power spectra against real OCRA-p data, including flux density measurements and spectral index distributions.

Experimental results

Research questions

• RQ1 How do the statistical properties of the SZ power spectrum vary with cosmological parameters such as σ₈ and cluster mass function?
• RQ2 To what extent do point sources at 30 GHz contribute to the observed high-multipole excess in CMB power spectra?
• RQ3 How do clustering and spatial distribution of point sources affect the measured power spectrum in small-scale microwave maps?
• RQ4 What is the accuracy of flux density measurements and spectral index distributions for radio sources using the OCRA-p receiver?
• RQ5 How do simulated OCRA observations compare with real data in terms of noise, calibration, and source detection?

Key findings

• The SZ power spectrum exhibits significant skewness and non-Gaussianity at high multipoles (ℓ ≈ 2000–4000), especially in small 1°×1° maps, with skewness values up to 0.3 for σ₈ = 0.825.
• Point source contamination at 30 GHz contributes substantially to the high-multipole power excess, with the BB polarization power spectrum (Cℓ^BB) reaching ~10⁻⁵ μK² at S_cut = 0.1 Jy.
• The OCRA-p receiver successfully measured flux densities for CRATES sources with a median uncertainty of ~10%, and spectral index distributions were consistent with previous surveys.
• The 1.4–30 GHz spectral index distribution for sources shows a median of α ≈ -0.8, with a significant tail toward flatter spectra (α > -0.5) at higher flux densities.
• Simulated OCRA-p maps reproduced real data noise levels within 10%, validating the UMBRELLA simulation framework for future CMB experiments.
• The correlation matrix of the CMB+SZ power spectrum shows strong mode coupling at high ℓ, indicating that cluster clustering and non-Gaussianity must be modeled for accurate cosmological inference.