search

xzackli / Bolt.jl

differentiable boltzmann code
MIT License
43 stars 5 forks source link

zero polarization in RSA #56

Closed jmsull closed 2 years ago

jmsull commented 2 years ago

Right now there is no implementation of photon polarization multipoles during RSA: https://github.com/xzackli/Bolt.jl/blob/de08e4d5e46ee45008b6c6c344cc6e3aabf4e8df/src/perturbations.jl#L231

This is definitely not right - these should at least be zeroed out so that we neglect them rather than preserving whatever the final values were before turning on RSA.

If not zeroing them, we can somehow relate them to the photon RSA expressions? It is not obvious to me if they do this in the CLASS 2 paper, but we should be able to figure something out.

jmsull commented 2 years ago

Well, I guess in RSA the photon quadrupole is zero anyway, so we are probably justified in just setting all the polarization multipoles to zero, since they are sourced by the quadrupole. Will add this

jmsull commented 2 years ago

This is in the new draft PR #57