search

IQcollaboratory / galpopFM

forward modeling galaxy populations
MIT License
1 stars 0 forks source link

physical motivation for EDA model #55

Open changhoonhahn opened 3 years ago

changhoonhahn commented 3 years ago

One of the comments on the draft from Rachel was that we need to physically motivate the EDA model. She suggested that one way to do this would be to look at the Av vs M* and SFR relations using e.g. the GSWLC catalog (see notebook).

There's a correlation, so this might be sufficient for justifying our parameterization in the EDA: image

TStarkenburg commented 3 years ago

@changhoonhahn Can you upload (or send me) the GSWLC data you use? I'd like to see contour/2dhist plots as well as the dependence on sSFR because we changed the DEM to have sSFR instead of SFR, but I'd be happy to change the notebook myself if you get me the data used.

changhoonhahn commented 3 years ago

@TStarkenburg I used GSWLC-M2 from https://salims.pages.iu.edu/gswlc/#catalog-download

TStarkenburg commented 3 years ago
Screen Shot 2021-01-07 at 9 19 10 PM Screen Shot 2021-01-07 at 9 19 16 PM

The correlation of A_V is actually strongest with sSFR. There is not a very clear dependence of A_V on mass in general, but at fixed sSFR there is a dependence for higher sSFR galaxies (sSFR ~> -10.5 yr^-1), not as much below that.

changhoonhahn commented 3 years ago

@TStarkenburg I've marked where I think a paragraph that describes the motivation for our parameterization (or refers to the appendix) should go in Section 3.

TStarkenburg commented 3 years ago

Added a figure in the paper (not sure we do want it because it is a bit confusing that it uses a different dataset (GSWLC); could move it to an appendix or remove later), working on text in Sect. 3.