Page 1 of 1

CAMB: replacing reionization history

Posted: February 18 2009
by Shek Yeung
I'm doing research about a model with non-tanh reionization history. I already changed the function Reionization_xe in reionization.f90, but now I'm not sure about the value of ReionHist%tau_start. The comment says it's the time when xe is 'very small', how should I change it if the reionization history is not tanh anymore?

Another question is about the calculation of actual_opt_depth. I've read the subroutine inithermo but I don't understand how it's calculated. What is the theory behind this value? Any recommended readings?

Re: CAMB: replacing reionization history

Posted: February 20 2009
by Antony Lewis
Any value where your x_e is very small should be OK (e.g. 1e-4).

The "actual" optical depth is the calculated optical depth due to reionization (which may differ from input value since that is mapped approximately onto a reionization redshift). It's the integral using the difference of the total x_e to the x_e there would be with no reionization (so is well defined, not depending on an arbitrary high-z cut-off, and zero when no reionization even though there is finite scattering due to the ~2e-4 ionization fraction left from recombination).