Reference frequency for s5 model in intensity

[erussier]

While reading the presets.cfg file, I noticed that the s5 model has 23 GHz as a reference frequency for both intensity and polarization. In the PySM documentation, it says that the reference frequency for intensity is supposed to be Haslam (408MHz). Susan (@seclark) tells me that we want the reference frequency for intensity to be 408MHz to avoid free-free. Could somebody perhaps clarify this?

[zonca]

@giuspugl when preparing the templates has scaled the Haslam map to 23 GHz using a fixed spectral index. See https://pysm3.readthedocs.io/en/latest/preprocess-templates/synchrotron_template_logpoltens.html#Synchrotron-maps.

If this choice needs further discussion let's plan to talk about it on one of the next calls. Or add more details to this issue.

[giuspugl]

@erussier i confirm that we have scaled the haslam intensity template to WMAP K band by assuming a constant spectral index -3.1 for s5.

[seclark]

Thanks both. Looking at this now I think it's clear, but happy to discuss further as needed. It may lead to some user confusion that reference frequency and template frequency are not the same thing, though they are often the same. But I think the documentation you link to is clear.

[zonca]

as I think about it, we are forced to have the templates at the same frequency when we use the polarization fraction formalism, as we are mixing the components together.