Hi Daniel,
Thanks for reaching out. It does indeed seem odd that the velocity is shown as being negative when you set it to 0 km/s, but I have an idea for this. I assume you're getting these results by using the `SpectrumGenerator` class with `bin_space=velocity` keyword? Can you check what the redshift is for the elements of your `LightRay` that you used to generate the spectrum? When Trident generates a `LightRay` through a simulation volume, it usually starts with the redshift of the simulation volume at the *back* of the LightRay and then decrements in redshift as the LightRay gets closer to the observer. This means that usually, you have lower values at the front of the LightRay than at the back, usually lower than the redshift of the simulation snapshot by a little bit. I imagine this might be happening here, in that you've set the velocities to 0, representing the peculiar velocity of the gas, but if the redshift is slightly lower than you expect, then you'll get cosmological velocities that are slightly negative, which might give you the behavior you're showing.
I'm on vacation this week and return to "my office" on Tuesday. What I'd recommend doing is :
(1) checking these redshift values of the LightRay that generated these spectra to confirm that the redshift is slightly lower than you anticipated on the near side. This may be the answer.
(2) checking to see if you get the same behavior if you change to `bin_space=wavelength`.
I can help with this next week if you don't make any progress before then.
Cameron