When ASTERIA accesses spectra from snewpy models, it is pulling out the meanE and pinch attributes of the model classes and then computing energy spectra from those. Instead, it should be using the models' native function call
spectra = model.get_initial_spectra(t, E)
Is there a good reason we are not doing this? As written the code is brittle and doesn't respect the encapsulation provided by the model classes.