abundance of radiating species

[migueldvb]

It is now possible to define the fractional abundance of a radiating species with respect to a weighted sum of the densities supplied for collisional partners. If this fractional abundance is not constant througout the grid one needs to call the density function in the definition, for example:

void
abundance(double x, double y, double z, double *abundance){
  double *val;
  density(x,y,z,val);
  abundance[0] = f(x,y,z)/val[0];
}

Why not defining the density of the radiating species in the model file rather than a fractional value?

[allegroLeiden]

Personally I have no strong opinion about this either way. I would say it should be decided by consensus among Lime users. Unfortunately I suspect rather few of them read these pages... your idea of a mailing list would help here.

[tlunttil]

Isn't it much more common (at least in my field ISM/star formation) to talk about abundances relative to H_2 rather than absolute densities? So to me defining the abundance makes more sense

[tlunttil]

Neither @allegroLeiden nor me really use Lime, you see...

Maybe @smaret can comment.

[allegroLeiden]

Also @attilajuhasz, and my boss Michiel Hogerheijde, who I think is not signed up here.

[cleevesli]

As @tlunttil said, the most common convention in astrochemical models is providing abundances relative to H2 (or sometimes total H). The abundances are either prescribed by some kind of toy abundance model (power law within some range) or interpolated over a pre-calculated chemical model, which is what I do (or also calculated in LIME if I understand correctly, @smaret?). I could just as easily provide total space density of the radiating species but the abundance more directly tells you how much of your total available O, C, N, etc go into a particular species. So if I see a water abundance of around 2e-4 then I know that in normal circumstances that is almost all of the available oxygen, while a space density doesn't directly tell you that without knowing the H2 density.

L. Ilsedore Cleeves Hubble Fellow, Harvard-Smithsonian Center for Astrophysics 60 Garden St, MS-51 Cambridge, MA 02138 1-706-237-8398 ilse.cleeves@cfa.harvard.edu https://www.cfa.harvard.edu/~lcleeves/index.html

On Thu, Nov 10, 2016 at 7:56 AM, allegroLeiden notifications@github.com wrote:

Also @attilajuhasz https://github.com/attilajuhasz, and my boss Michiel Hogerheijde, who I think is not signed up here.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/lime-rt/lime/issues/186#issuecomment-259683186, or mute the thread https://github.com/notifications/unsubscribe-auth/AGXPcD7wmeRgXY4CBASkUqXSJbqlTOC0ks5q8xRzgaJpZM4Ktqkl .

[allegroLeiden]

We could possibly provide the option - have it that if the user provides the present abundance function, this is used, but if not, then Lime looks for a molDensity function or similar. There are other ways to implement this, although I rather shrink from suggesting yet another user parameter to control the choice.

[migueldvb]

Thank you for you comments. Maybe it is not worth to add another function for defining the density function since it looks like it will be not be very useful in general. For comet models there is no H2 and every molecule has its own spatial distribution depending on their photodisociation rates. However it is not a big issue to specify the fractional abundance with respect to the collisional partners.

[allegroLeiden]

I asked Michiel Hogerheijde here for his opinion on several of these issues. Since he is not on github I have to cut-and-paste his comments.

MH: Allowing definition of MolDensity or Abundance makes sense. Sometimes abundances are what you have. Sometime you may have abundances relative to some tracer (e.g. H) that is not equal to the collision partners that LIME uses (o-H2 and p-H2, e.g.). Expressing your abundances relative to what LIME expects may get the user confused, while calculating a MolDensity is what the user can easily do (since they understand their own standard well).

[migueldvb]

Having an optional abundance function, and otherwise using molDensity supplied by the user sounds very good.