I am attempting generate a velocity from the molecular hydrogen content in each gas cell for multiple subhalos in the TNG100-1 (z=0) simulation. While the velocity and several metallicities are included on a cell-by-cell basis, I see, however, that the neutral and molecular hydrogen data is provided via the metadata as a low-res radial profile. I wanted to inquire if there is another method by which I could ascertain the molecular hydrogen content per cell. Ultimately, I would like to generate the velocity field for both atomic and neutral hydrogen, but these will essentially look quite similar since they would both be generated from the same profile, just different fractional masses. Thank you for your time.
(1) For a given cell, the velocity of the different components of hydrogen would all be the same (by definition).
(2) If you construct a velocity field, i.e. projected along a given line of sight, then the differing amounts of different hydrogen components in different gas cells could (would) produce different velocity fields. This is because each pixel of the velocity field map would be averaging over (the same) cells, but with different weights.
(3) You definitely can compute, cell by cell, the amount of different hydrogen components (e.g. molecular). This is exactly how the low-res radial profiles were made. If interested, I would suggest to follow Popping+2019 as a start. There a few different H2 models are described (these have been computed for cosmological hydrodynamical simulations like TNG in many papers over the past several years).