spacetelescope / synphot_refactor

Synthetic photometry using Astropy
http://synphot.readthedocs.io/en/latest/
BSD 3-Clause "New" or "Revised" License
38 stars 25 forks source link

Interest for synthetic infrared molecular spectra? #256

Open erwanp opened 4 years ago

erwanp commented 4 years ago

Hello!

We developed the Radis package to calculate synthetic infrared molecular spectra. Radis was initially used for Plasma & Combustion, but we're now finding new use cases, in particular in Astronomy for Exoplanet detection.

I'm not an astronomer, but reading the Synphot documentation I saw that you sometimes generate Source spectra with certain lineshapes (e.g: Gaussian), or Blackbody continuum (although, mostly in the visible spectrum).

Is so, it may be interesting (and very easy) to feed Synphot with some spectra calculated in Radis. Let me know if you see potential use cases, here or in other packages.

pllim commented 4 years ago

Hello and thank you for re-posting this on GitHub as requested!

In a nutshell, a synphot spectrum is an extension of astropy models (doc) that handles spectrum-specific things like fluxes.

What are the use of these source spectra?

Mostly for instrumentation and observation planning. For example, "I want to observe this star. I can approximate it as a black body. Now I want to know the count rate if I use this telescope." Or... whatever people were doing with them back in IRAF synphot task days.

Do you ever need synthetic data from molecules ?

I am not an astronomer either, so I cannot speak for the users. I am sure some of them need molecular data, but whether they need to use the data with synphot or not, I do not have that information, unfortunately.

Do HST and JWST also provide infrared spectra?

Yes, to the extend that is needed by the telescopes, which do have IR detectors. All the available data are documented at https://stsynphot.readthedocs.io/en/latest/stsynphot/appendixa.html (stsynphot is a package that extends synphot).

feed Synphot with some spectra calculated in Radis

If you have a use case that needs this feature, I don't see why not as long as your Spectrum object can be cast into an astropy model. Sampled data usually ends up as Empirical1D model (think tabular model in astropy).

There is already a bridge to specutils implemented in #243 . If you are curious on how to interface with synphot, maybe that PR will give you some ideas, particularly in spectrum.py.

Hope this helps!