New spectral quantities (transmittance & absorption)

[erwanp]

Current https://radis.app plots the spectrum emission ("radiance"). Many spectroscopists are also looking for spectrum absorption, with either "transmittance" and absorption coefficient / opacity ("abscoeff").

Suggested implementation :

A first version should be quite easy to implement as RADIS backend already computes all of these quantities.

• We would need a button [dropdown listchoice ?] for "emission" / "transmittance" / "opacity" [which is shorter than "absorption coefficient"] that would feed a variable quantity [or whatever]

• plot_spectrum should be modified to something along the lines of :

if quantity == 'emission':
    if request.query_params.simulate_slit:
        var = 'radiance'
   else:
       var = 'radiance_noslit'
elif quantity == 'transmittance':
    if request.query_params.simulate_slit:
        var = 'transmittance'
   else:
       var = 'transmittance_noslit'
elif quantity == '
    if request.query_params.simulate_slit:
        ## this is not implemented in radis because it makes little sense from an experimentalist perspective (this quantity is never measured directly), but could be if people want to use it to "filter" the spectrum. 
   else:
       var = 'transmittance_noslit'
else:
    raise ValueError

For the moment switching from emission to transmittance would require the users to recompute the spectrum (which would be quite fast as the bottleneck is the automatic download of the files, which would be cached on the 2nd try). If #153 is implemented and the Spectrum object is stored, then switching would not even require to recompute the spectrum at all and could be made instantaneous !

[erwanp]

Hello @suzil !

Have shown the app around to some people recently, and they were impressed : well done !

Many ask about absorption features. Do you think it's easy to add a simple component to choose between transmittance / absorption coefficient / radiance ?

[suzil]

Yup! I can prioritize doing this.

[suzil]

elif quantity == 'transmittance':
    if request.query_params.simulate_slit:
        var = 'transmittance'
   else:
       var = 'transmittance_noslit'
elif quantity == '
    if request.query_params.simulate_slit:
        ## this is not implemented in radis because it makes little sense from an experimentalist perspective (this quantity is never measured directly), but could be if people want to use it to "filter" the spectrum. 
   else:
       var = 'transmittance_noslit'

What's the point of having an opacity/absorption coefficient option if it's treated the same as transmittance?

this is not implemented in radis because it makes little sense from an experimentalist perspective (this quantity is never measured directly), but could be if people want to use it to "filter" the spectrum.

Would it make sense to leave it off for now if we don't treat opacity any difference than transmittance? @erwanp

[erwanp]

Hello, seems I never answered this, sorry !

"absorption coefficient" is never measured, so won't be used by experimentalists (people measure either radiance or transmittance spectra). However, it's of interest for people doing modeling.

Btw, transmittance (without slit function) is just :

T = exp(-k L)   

where k is the absorption coefficient I'm talking about. Then transmittance is convolved with a slit function to obtain the experimental transmittance.