Open iancze opened 8 years ago
It seems like there's a resonance feature that reorders the wavelengths at one index location:
Not sure how to handle this, though these are exceptionally short wavelengths, so it's unlikely that you have data there. Still it's problematic if you're trying to integrate the function or something.
Ah, your plot makes the issue clearer. I think the problem is that the vacuum to air conversion we are using is not valid outside of optical wavelengths.
The Ciddor 1996 formula used is valid for the wavelength range between 0.23 and 1.7μm. Found in the conclusion here as the actual paper is behind a paywall.
Steps to reproduce:
Output
Obviously, the wavelength grid should be monotonic. I'm not really sure what's causing this, but it seems to be related to applying the
vacuum_to_air
method to a larger range of wavelengths than just the optical.