Public repository of the Cosmic Linear Anisotropy Solving System (master for the most recent version of the standard code; GW_CLASS to include Cosmic Gravitational Wave Background anisotropies; classnet branch for acceleration with neutral networks; ExoCLASS branch for exotic energy injection; class_matter branch for FFTlog)
230
stars
285
forks
source link
Discontinuities in the ratio of non-linear to linear power spectra #388
We are calculating the ratio of the non-linear matter power spectrum (using halofit) to the linear matter power spectrum. The linear matter power spectrum is always calculated at a reference redshift, z_0=0.0. A test example is:
k=1.5
z=4.66
and all cosmological parameters (the best fit parameters, constrained by the KiDS-450 data) are kept fixed, except omega_cdm. This ratio (non-linear power spectrum to linear power spectrum), as a function of omega_cdm has some discontinuities even around the most likely values for the parameter. However, for small values of the wavenumber, for example, k=10^-4, the ratio is a smooth function of omega_cdm. The results from the two test cases are shown below
We have made a Jupyter notebook to show the inputs to CLASS and the calculations.
A somewhat related issue is raised in https://github.com/lesgourg/class_public/issues/370, but, unfortunately no one has provided any comments. Is there any information regarding the source of the discontinuity, which must be unphysical? Or is it possible that we are misinterpreting either the inputs to CLASS, or the output?
We are calculating the ratio of the non-linear matter power spectrum (using halofit) to the linear matter power spectrum. The linear matter power spectrum is always calculated at a reference redshift, z_0=0.0. A test example is:
and all cosmological parameters (the best fit parameters, constrained by the KiDS-450 data) are kept fixed, except omega_cdm. This ratio (non-linear power spectrum to linear power spectrum), as a function of omega_cdm has some discontinuities even around the most likely values for the parameter. However, for small values of the wavenumber, for example, k=10^-4, the ratio is a smooth function of omega_cdm. The results from the two test cases are shown below
We have made a Jupyter notebook to show the inputs to CLASS and the calculations.
https://github.com/Harry45/class_codes/blob/main/issue_1.ipynb
A somewhat related issue is raised in https://github.com/lesgourg/class_public/issues/370, but, unfortunately no one has provided any comments. Is there any information regarding the source of the discontinuity, which must be unphysical? Or is it possible that we are misinterpreting either the inputs to CLASS, or the output?
Thank you.