CAMB: Can't recover sigma-8 from published table

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Matthieu Schaller
Posts: 2
Joined: July 05 2020
Affiliation: Leiden Observatory

CAMB: Can't recover sigma-8 from published table

Post by Matthieu Schaller » July 05 2020

Hi all,

I am trying to get a matter power-spectrum prediction out of CAMB to test some simulation results. Before that I wanted to verify that I can run CAMB and get an output that matches the published Planck-13 values.
I am especially interested in the last column of table 5 of https://arxiv.org/abs/1303.5076.

I use as CAMB input:

Code: Select all

ombh2          = 0.022161
omch2          = 0.11889
omnuh2        = 0
omk             = 0
hubble         = 67.77

initial_power_num         = 1
pivot_scalar              = 0.05
pivot_tensor              = 0.05
scalar_amp                = 2.21381e-9   # ln(10^10 A_s) = 3.0973
scalar_spectral_index     = 0.9611

reionization         = T
re_use_optical_depth = T
re_optical_depth     = 0.0952

massless_neutrinos = 3.046
massive_neutrinos  = 0
share_delta_neff = T
(All the rest is pretty standard but happy to provide more... can't attach my full file to the post)

but when asking CAMB what sigma_8 at z=0

Code: Select all

pars = camb.read_ini('EAGLE_2013.ini')
pars.set_matter_power(redshifts=[0.], kmax=30.0)
pars.NonLinear = model.NonLinear_both
results = camb.get_results(pars)
results.calc_power_spectra(pars)
sigma_8_z0_1 = results.get_sigma8_0()
should be I get: 0.84117.

This differs from the reported value in the table (0.8288).

I am surely doing something silly here in the way I ran CAMB. I have tried changing some of the accuracy parameters or range parameters but could not see any difference in the outcome.

Any help figuring out what I am doing wrong here would be greatly appreciated.

Thanks!
Matthieu

Antony Lewis
Posts: 1941
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: CAMB: Can't recover sigma-8 from published table

Post by Antony Lewis » July 06 2020

The Planck LCDM model has one massive neutrino eigenstate with mnu=0.06eV.

Matthieu Schaller
Posts: 2
Joined: July 05 2020
Affiliation: Leiden Observatory

Re: CAMB: Can't recover sigma-8 from published table

Post by Matthieu Schaller » July 06 2020

Thanks! Indeed, I overlooked the neutrinos.
That indeed goes in the right direction. If I change my input to include

Code: Select all

omnuh2         = 0.00064

massless_neutrinos = 2.046
massive_neutrinos  = 1
share_delta_neff = T
then I obtain sigma_8 (z=0) = 0.82927. (0.05% difference)

Could I be missing something else or is this within the expected range of differences?

Antony Lewis
Posts: 1941
Joined: September 23 2004
Affiliation: University of Sussex
Contact:

Re: CAMB: Can't recover sigma-8 from published table

Post by Antony Lewis » July 07 2020

Seems close - also check helium fraction.

Post Reply