search

moustakas / impro

A wide-ranging suite of IDL routines which observational extragalactic astronomers may find useful.
GNU General Public License v2.0
14 stars 11 forks source link

Add FSPS model based on the MIST isochrone #7

Open dr-guangtou opened 5 years ago

dr-guangtou commented 5 years ago

FSPS has updated to the new MIST isochrone: http://waps.cfa.harvard.edu/MIST/

It would be great to have it in iSEDFit! Let me know if there is anything I can help.

moustakas commented 5 years ago

@dr-guangtou

I'm just taking a look at this--it's been a long time since I've added models to iSEDfit. One thing I realized is I need a grid of SSPs with these isochrones. To match my existing models, it would be useful if these were generated with both the MILES and BaSeL stellar libraries and for the Chabrier, Kroupa, and Salpeter IMFs.

The output files (for each available metallicity) should be in the FSPS .spec format with the filename, e.g., SSP_Padova_CKC14z_Kroupa_Z0.0003.out.spec.

Are these spectra something you would be able to generate yourself using either FSPS or python-fsps?

Some more notes on metallicity:

It looks like the MIST isochrones exist for 

['m2.50','m2.00','m1.75','m1.50','m1.25','m1.00','m0.75',
  'm0.50','m0.25','p0.00','p0.25','p0.50']

or

Z = 10^[-2.5,-2,-1.75,-1.25,-1,-0.75,-0.5,-0.25,0.0,0.25,0.5]*0.019
  6.0083272e-05   0.00018999999   0.00033787309    0.0010684485    
  0.0018999999    0.0033787307 0.0060083275     0.010684485     0.018999999     
  0.033787306     0.060083274

assuming Zsun=0.019.

For MILES my notes say that the only metallicities available are Z=0.0008, 0.0031, 0.0096, 0.0190, 0.0300 so we'll need to be a little careful. The manual may include some useful information or we can reach out to Charlie or Ben.

dr-guangtou commented 5 years ago

Yes, I can generate the SSP spectra using fsps. It should be very straightforward.

The MIST isochrone does allow more metallicity, but I am not sure all of them are "safe". I can take a look of the MIST paper and check with Ben on this.

If it is not too much trouble, it would be interesting to include the new Parsec isochrones too. I can provide all the .spec files.

dr-guangtou commented 5 years ago

And, BTW, MIST uses a updated Zsun=0.0142 based on Asplund et al. 2009; Parsec assumes Zsun=0.01524.

These can be found at line 166 of sps_vars.f90.

dr-guangtou commented 5 years ago

I just asked Ben Johnson. Apparently for SSP with metallicity lower than the spectral library limit, you will basically get the closest metallicity (lowest Z) one.

In this case, I think we should ignore all SSPs with Z lower than the lower boundary of the spectral library (when using MILES).