Camb AxionEffectiveFluid dark energy model

Use of Cobaya. camb, CLASS, cosmomc, compilers, etc.
Post Reply
Adrien La Posta
Posts: 3
Joined: May 01 2020
Affiliation: CNRS LAL IJClab

Camb AxionEffectiveFluid dark energy model

Post by Adrien La Posta » May 01 2020

Hi !

I'm trying to use the AxionEffectiveFluid dark energy model from CAMB to generate spectra.

class camb.dark_energy.AxionEffectiveFluid[source]
Bases: camb.dark_energy.DarkEnergyModel
Variables:
w_n – (float64)
om – (float64)
a_c – (float64)
theta_i – (float64)

I noticed that varying the om parameter seems to change the value of [math]. But I fixed this parameter while declaring the cosmology parameters. Do you know why the spectra are modified this way ?

Thanks for your future answsers

Adrien
difference_om.png
difference_om.png (86.41 KiB) Viewed 1429 times

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

Re: Camb AxionEffectiveFluid dark energy model

Post by Antony Lewis » May 01 2020

Theta_MC is an approximation for standard cosmologies. Better to fix thetastar in this case if you want the sound horizon to actually stay constant (see https://camb.readthedocs.io/en/latest/model.html#camb.model.CAMBparams.set_cosmology).

Adrien La Posta
Posts: 3
Joined: May 01 2020
Affiliation: CNRS LAL IJClab

Re: Camb AxionEffectiveFluid dark energy model

Post by Adrien La Posta » May 04 2020

Hi !

I tried to keep thetastar constant, but the two spectra are still moving the same way. I put the script I used in attached files (sorry for the .zip, i could not attach a .py file)
axion_camb.zip
(853 Bytes) Downloaded 148 times

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

Re: Camb AxionEffectiveFluid dark energy model

Post by Antony Lewis » May 04 2020

If you change the dark energy model, you need to do so *before* calling set_cosmology if you are using cosmomc_theta or thetastar (because the solution for H0 to give required thetastar depends on the dark energy model). Alternatively use camb.set_params, e.g.

Code: Select all

pars = camb.set_params(dark_energy_model='AxionEffectiveFluid', thetastar=...)

Also your fluid om values are probably far too large to get a valid H0 solution, since this model decays rapidly at late times.

Adrien La Posta
Posts: 3
Joined: May 01 2020
Affiliation: CNRS LAL IJClab

Re: Camb AxionEffectiveFluid dark energy model

Post by Adrien La Posta » May 04 2020

Thanks for this answer. I tried and it seems to be working, it was very helpful !

Post Reply