BUG: Overgeneration of galaxies #578
Conversation
Add velocity dispersion function sampler to skypy (skypyproject#573)
| @@ -32,7 +32,7 @@ def test_schechter_lf_redshift(): | |||
| x_min = 10.**(-0.4*(M_lim - M_star)) | |||
|
|
|||
| # density with factor from upper incomplete gamma function | |||
| density = phi_star*gamma(alpha+1)*gammaincc(alpha+1, x_min) | |||
| density = phi_star*gamma(alpha+1)*gammaincc(alpha+1, x_min)*np.power(cosmo.h, 3) | |||
There was a problem hiding this comment.
I don't think this is the right place to account for h. The parameter phi_star has the same dimension [unit] as the luminosity function, which is 1/V/luminosity [commonly Mpc^-3 mag^-1]. That means phi_star must agree with these dimensions/units.
In your example, you are using the parameters of Tortorelli2020. In this work, they are giving phi_star in units of h_70 = H_0/70. So the input parameter must be corrected by this unit. However, as your defined cosmology includes H_0 = 70, h_70 is equal to 1. Therefore, the number that you have in your config files corresponds to units of Mpc^-3 mag^-1, which are as I said the physical units.
If we, however, want to enable phi_star to be given in units of h_0, we would also need to specify this in the documentation. But in this case, the parameter in your config file has the wrong value.
|
The current implementation is correct but we will add clarification about the units in the docs with a new PR. Issue #576 will be changed accordingly. |
Description
This fixes the overgeneration of galaxies caused by the lack of a$h_0^3$ factor in the number density calculation (required due to the units of $phi_*$ . Attached are a PDF of a notebook showing this result and text files for the generation of the galaxies.
This will close #576 .
Checklist
red_galaxy_text.txt
blue_galaxy_text.txt
Galaxy_Overgeneration_(Git).pdf