Banding strategies - one photon generation and on creating crude spectra

[kslong]

Banding or something similar occurs in two places.

• We create photons in energy bands to ensure we have a good representation of the spectrum in wavelength ranges that may not carry much energy, but which affect the ionization

• We construct crude spectra of the photons that pass through each cell, which is used to calculate spectra with a cell for ionization and recombination calculations.

Although the choices we have made may be reasonable, we have not studied the effects of banding systematically. The purpose of this page is to begin a discussion on how this might be carried out.

The first steps for this studied probably should be to select a couple of systems we would like to simulate and to test whether our choices of banding matter when the banding is the only thing that is changed in the calculation. For photon generation this is straightforward since a number of options have been implemented, and in addition it is possible to define the banding on the fly. For crude spectrum generation, the bands are hardwired, and so one will like need either to add a more flexible capability to the program or just hard code a few different tests.

Once these tests are done, we should think about whether there are ways to make the banding more sensible, and in the "best of all possible worlds" to automate it.

(Note that banding is also related to questions about convergence, which is something which also begs study. )