A control variate technique using differenced estimates from realistic masked and isotropic simulations reduces the computational cost of CMB lensing bias calculations by a factor of three to five.
E/B decomposition of finite pixelized CMB maps
2 Pith papers cite this work. Polarity classification is still indexing.
abstract
Separation of the E and B components of a microwave background polarization map or a weak lensing map is an essential step in extracting science from it, but when the map covers only part of the sky and/or is pixelized, this decomposition cannot be done perfectly. We present a method for decomposing an arbitrary sky map into a sum of three orthogonal components that we term ``pure E'', ``pure B'' and ``ambiguous''. This method is useful both for providing intuition for experimental design and for analyzing data sets in practice. We show how to find orthonormal bases for all three components in terms of bilaplacian eigenfunctions. The number of ambiguous modes is proportional to the length of the map boundary so fairly round maps are preferred. For real-world data analysis, we present a simple matrix eigenvalue method for calculating nearly pure E and B modes in pixelized maps. We find that the dominant source of leakage between E and B is aliasing of small-scale power caused by the pixelization. This problem can be eliminated by heavily oversampling the map, but is exacerbated by the fact that the E power spectrum is expected to be much larger than the B power spectrum and extremely blue. We found that a factor of 2 to 3 more pixels are needed in a polarization map to achieve the same level of contamination by aliased power than in a temperature map.
citation-role summary
citation-polarity summary
fields
astro-ph.CO 2years
2026 2verdicts
UNVERDICTED 2roles
method 1polarities
use method 1representative citing papers
BROOM is a Python package that applies ILC and GILC techniques for model-independent separation of CMB, SZ, and foreground signals in microwave data along with diagnostic and simulation utilities.
citing papers explorer
-
Faster CMB lensing with control variates
A control variate technique using differenced estimates from realistic masked and isotropic simulations reduces the computational cost of CMB lensing bias calculations by a factor of three to five.
-
BROOM: a python package for model-independent analysis of microwave astronomical data
BROOM is a Python package that applies ILC and GILC techniques for model-independent separation of CMB, SZ, and foreground signals in microwave data along with diagnostic and simulation utilities.