CMB Diagnostics

This repo contains code to derive calibration information for SO cmb maps.

Dependencies

• namaster
• scipy
• numpy
• healpy
• pixell
• pygsm(https://github.com/liuyiqiandrew/pygsm)

Installation

Clone the repo, add the path of the parental directory of the repo to PYTHONPATH

Functions

At the moment, the repo supports:

• Estimating transfer function from EE SO x Planck cross spectra.
• Estimating transfer function from TE SO x Planck cross spectra (Planck T, SO E).
• Estimating polarization angle from SO EB spectra.

Example

In test/example.py. If you're on Della, you can run the codes directly. Resources are hard coded. You will derive the results for 20240714 SAT-P3 maps. Otherwise, you'll need to prepare stuffs below and modify attributes in each class accordingly. See next paragraph for what you need to prepare.

If cmb_util(https://github.com/liuyiqiandrew/cmb_utils) is available on git. Feel free to grab functions directly from repo or add it to your PYTHONPATH.

What's about the code

Apologies for the horrible user interface, I may consider refining it if I have more spare time. I would strongly suggest skim through PSContainer to get a sense in how the code works. In particular, understanding the constructor and initalizer for namaster field init_planck_f2(self) will give you a good sense on how to format file name and frequency information.

Currently, all estimator class inherits from PSContainer which handles all power spectra information. Estimator is only responsible for the final estimator specific computation (i.e. transfer function, polarization angle etc.) and saving the result.

What you need to prepare

• Planck maps in equatorial coordinates and HealPix Pixellization (if you're on della, they are availale at /home/yl9946/projects/tp_leakage/planck_equatorial/planck_*_equatorial_rm_mnp_dp.fits). You need to modify the path to these maps by, for example

  tf_ee_est = cmb_diagnostics.TransferFuncEE()
  tf_ee_est.planck_fname = '/home/yl9946/projects/tp_leakage/planck_equatorial/planck_{}_equatorial_rm_mnp_dp.fits'
  

  The file name is generic, {} is where you distinguish frequencies (but you should specify frequncy in tf_ee_est.planck_freqs, add string formating info in {} as it fits.). You will also need the effective planck frequency for dust and put them in tf_ee_est.planck_eff_freqs
• SO maps, in carr pixelization. Support the file name as follows

  tf_ee_est = cmb_diagnostics.TransferFuncEE()
  tf_ee_est.so_fname = '/scratch/gpfs/sa5705/shared/SO_SAT/satp3_maps/cmb_maps_satp3_20240714/map_f{freq:03d}_muKcmb.fits'
  

• Beams for SO and planck. Assume gaussian beams, the values are stored as numpy array in tf_ee_est.so_beams and tf_ee_est.planck_beams.
• Camb model, something like the 'nobb' template from BBPower(https://github.com/simonsobs/BBPower/blob/main/examples/data/camb_lens_nobb.dat). Supply the path at tf_ee_est.camb_dl_path.

Inquiries