Document setting hostebv from extinction law

[geotaylo]

Apologies if there is some obvious way to do this, I couldn't find anything in the documentation.

I'm wanting to fit a 'realistic' SN with host and mw dust. When simulating photometry using realize_lcs(), I can include mwebv in the list of params by generating it from some coordinates (and set this in the model before fitting the other parameters) - this is well documented. How can I also generate a sensible hostebv value to be included in the list of params when simulating photometry? There is no point in me fitting for it if the effect hasn't been applied to the photometry.

[kbarbary]

In a simulation, you would typically randomly select the hostebv value from a distribution. Which distribution depends a lot on the purpose, and what kind of SNe you're simulating. Core collapse SNe are typically in higher-dust environments than Ia's, for example, so if you were simulating CC SNe, your choice of distribution should reflect that.

Here are some "possible" distributions to use for SNe Ia in wide-field surveys:

This is in terms of A_V = 3.1 * E(B-V). This is Figure 5 of this paper. There might be better references now, but it at least discusses the uncertainty in A_V a bit.

[geotaylo]

Perfect, that clears it up. Thanks.