xpol 1.6.0

XPOL cross power-spectrum estimator

Table of Contents

• Xpol

• Download

• Install Python

• Install C

• Equations

• Usage

Xpol

a Power Spectrum estimator based on cross-correlation between maps.

This is the generalisation to polarisation of Tristram M. et al., 2005 MNRAS 358 833

Xpol have been used to derive power spectra for:

• Archeops

  • CMB power spectra [Tristram+2005, A&A 436 785]

  • Dust power spectra [Ponthieu+2005, A&A 444 327]

• Planck:

  • CMB power spectra [Planck 2015 results. XI, Couchot+2015]

  • Dust power spectra [Planck Intermediate Paper XXX]

  • CIB power spectra [Planck 2013 results. XXX]

  • SZ power spectra [Planck 2013 results. XXI, Planck 2015 results XXII]

  • SZ-CIB cross-power spectra [Planck 2015 results. XXIII]

Download

git clone https://gitlab.in2p3.fr/tristram/Xpol.git /where/to/clone

Install Python

Install the python stand-alone module xpol using

pip install -e /where/to/clone

The -e option allow the developer to make changes within the Xpol directory without having to reinstall at every changes. If you plan to just use the code and do not develop it, you can remove the -e option.

You can find a example of the module usage in test/script_xpol.py

Install C

The code is in C, fully MPI parallelized in CPU and memory (using spherical transform by s2hat).

Librairies needed are :

• HEALPix (http://healpix.jpl.nasa.gov/)

• cfitsio (http://heasarc.gsfc.nasa.gov/fitsio/fitsio.html)

• s2hat (http://www.apc.univ-paris7.fr/APC_CS/Recherche/Adamis/MIDAS09/software/s2hat/s2hat.html)

In addition you will need some BLAS/LAPACK optimized routines (MKL, NAG, …)

Equations

wiki is here

Usage

You can use the python class (Xpol) to compute spectra based on a data.

import xpol
binning = xpol.Bins( lmins, lmaxs)
xp = xpol.Xpol( mask, bins=binning)
pcl,cl = xp.get_spectra( dT)

see the file examples/script_xpol.py for more detailed usage.

For intensive computation and simulation you can alternatively use the MPI C version with a parameter file detailed here