The Atacama Cosmology Telescope: Temperature and Gravitational Lensing Power Spectrum Measurements from Three Seasons of Data

TL;DR

ACT delivers high-resolution measurements of the CMB temperature power spectrum from three seasons across 148 and 218 GHz, including cross-frequency and cross-season spectra. The analysis combines equatorial and southern sky patches with careful calibration to WMAP, sophisticated beam and foreground modeling, and extensive simulations to produce robust spectra and a significant lensing detection. The results reproduce the damping tail, constrain SZ and extragalactic foregrounds, and yield a lensing amplitude in agreement with LCDM and SPT, validating ACT's cosmological potency and paving the way for polarization-based lensing studies. Overall, the work demonstrates precise, multi-frequency CMB power spectrum estimation with thorough systematic validation and cross-consistency with other major CMB experiments.

Abstract

We present the temperature power spectra of the cosmic microwave background (CMB) derived from the three seasons of data from the Atacama Cosmology Telescope (ACT) at 148 GHz and 218 GHz, as well as the cross-frequency spectrum between the two channels. We detect and correct for contamination due to the Galactic cirrus in our equatorial maps. We present the results of a number of tests for possible systematic error and conclude that any effects are not significant compared to the statistical errors we quote. Where they overlap, we cross-correlate the ACT and the South Pole Telescope (SPT) maps and show they are consistent. The measurements of higher-order peaks in the CMB power spectrum provide an additional test of the Lambda CDM cosmological model, and help constrain extensions beyond the standard model. The small angular scale power spectrum also provides constraining power on the Sunyaev-Zel'dovich effects and extragalactic foregrounds. We also present a measurement of the CMB gravitational lensing convergence power spectrum at 4.6-sigma detection significance.

Figures (20)

• Figure 1: Equatorial maps (ACT-E) made from 2009 (upper panel) and 2010 (lower panel) 148 GHz observations filtered to emphasize modes in the range $\ell=500-2500$. The four data splits in either season were co-added to make this plot. Also delineated are the patches used for computing power spectra.
• Figure 2: Noise spectra for each season for the ACT-E maps for 148 GHz (upper panel) and 218 GHz (lower panel). The red solid line shows the CMB-only spectrum. At 148 GHz the power spectrum is sample variance limited at $\ell<2500$, while at 218 GHz detector and atmospheric noise dominate on most scales.
• Figure 3: Southern maps (ACT-S) made from 2008 (top panel) 2009 (middle panel) and 2010 (bottom panel) 148 GHz observations filtered to emphasize modes in the range $\ell=500-2500$. The four data splits were co-added to make this plot. Also delineated are the patches used for computing power spectra. The smaller two patches common between the three maps are used to compute cross-season cross-power spectra. The four larger patches for season 2sf are used to compute the full footprint 2008-only cross-power spectrum. Areas of large noise or stripes are heavily down weighted in the analysis. The color scale is the same as Fig \ref{['fig:equMaps']}.
• Figure 4: Noise spectra for each seasons for the ACT-S maps for 148 GHz (upper panel) an 218 GHz (lower panel). The red solid line shows the CMB-only spectrum. Season 3s is significantly noisier than the other two seasons. Note that the combination of seasons 3s and 4s is more sensitive than season 2s which was used in das/etal:2011 and dunkley/etal:2011.
• Figure 5: Comparison of a sky patch from the WMAP 7-year 94 GHz map jarosik/etal:2011 (top) with the map of the same region made from ACT 148 GHz (bottom) observations (co-added across seasons). All maps have been high-pass filtered with a $\cos^2 \ell$-like filter that goes from 0 to 1 for $100 < \ell < 300$. Agreement between the CMB features in the two maps is clear by eye.