I had been thinking that we could only include the new Draine models for simulations that have a grain size distribution in them (like Smuggle; some flavors of Gizmo etc.). But I don't think this is right -- there are also files in the dataverse that are based on Draine & Hensley precomputed size distributions. We can use those and still take advantage of the local ISRF in computing the PAH emissivities which will be more accurate than the current Robitaille/MW models.
Steps in order to do this:
[1] first and foremost the ISRF fork of Hyperion needs to be PR'd and merged into main so that it's more easily accessible for everyone by default.
[2] I think really the only thing that will change after that from the full PAH (size distribution) model Is that, pending some different parameters keywords, we can have a different file/io routine that reads into a different kind of PAH object that doesn't have sizes. The the pah luminosity function won't need to dot product into the size distribution but just take fractional contributions from beta_nnls.
Anyways, a feature not a bug, and something that should be doable fairly easily once the more complex PAH model is stabilized.
I had been thinking that we could only include the new Draine models for simulations that have a grain size distribution in them (like Smuggle; some flavors of Gizmo etc.). But I don't think this is right -- there are also files in the dataverse that are based on Draine & Hensley precomputed size distributions. We can use those and still take advantage of the local ISRF in computing the PAH emissivities which will be more accurate than the current Robitaille/MW models.
Steps in order to do this:
[1] first and foremost the ISRF fork of Hyperion needs to be PR'd and merged into main so that it's more easily accessible for everyone by default.
[2] I think really the only thing that will change after that from the full PAH (size distribution) model Is that, pending some different parameters keywords, we can have a different file/io routine that reads into a different kind of PAH object that doesn't have sizes. The the pah luminosity function won't need to dot product into the size distribution but just take fractional contributions from beta_nnls.
Anyways, a feature not a bug, and something that should be doable fairly easily once the more complex PAH model is stabilized.