Lensed Spectra from Lenspix and CAMB

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Aparajita Sen
Posts: 1
Joined: August 05 2020
Affiliation: Indian Institute of Science Education and Research

Lensed Spectra from Lenspix and CAMB

Post by Aparajita Sen » August 05 2020

Hi Antony,

I am generating lensed maps at NSIDE = 512 using lenspix. I used Cl s generated by CAMB as input and have set lmax 1535 in the lenspix parameter file. I find that the spectra from lenspix do not match with CAMB at l>1000. I have also used the lensed maps from lenspix and generated the spectra (green lines in the figures) using anafast from healpy. All three do not match with each other.
I find a very small disagreement even for the unlensed spectra at l>1200.

Could you please tell me if this is expected? Or if I am missing something? I have attached the plots for TT and BB spectra for your reference.

Thanks in advance,

Cosmo_lens_BB.png (30.91 KiB) Viewed 270 times

Antony Lewis
Posts: 1696
Joined: September 23 2004
Affiliation: University of Sussex

Re: Lensed Spectra from Lenspix and CAMB

Post by Antony Lewis » August 06 2020

Lensing is nonlinear and non-local in L, so you need quite high L max for stable results (e.g 4000). I've not tested at such low resolution, you may need to increase NSIDE (as well as making sure the interp factor is high enough - for low Nside output needs to be much larger).

Post Reply