Format for simulated 1-D spectra

We need to define a format for the spectra that are input to the simulations. These should come from the science collaborations so need to be flexible enough to capture a lot of different science cases (e.g. stars from GA; z ~ 1.5 galaxies with strong O[III] from cosmology).

The spectra should be noise free scaled to the observed brightness (and redshifted as appropriate) at sufficiently high sampling spatial that we can interpolate. The units are TBD, but probably Jy.

Questions: Should the lines be listed separately (lambda, EW, width)? If we don't do this we are in danger of convolving twice (forcing us to generate simulations at very high resolution).