Handling uncertainties in synphot

[pllim]

http://www.astrobetter.com/python-tip-re-sampling-spectra-with-pysynphot/

"This works well, but how would you resample an associated uncertainty array? You would want to add the uncertainties in each bin in quadrature but how could you include that in this snippet using pysynphot?"

Note: This can be tricky because instrument can have higher uncertainty over time (e.g., from increased CTE loss or noise) or different uncertainty depending on readout (e.g., with or without post-flash).

[pllim]

I do not think pysynphot handles uncertainties at all, although I have noticed in the past that some *_syn.fits files do have a third column to store uncertainties. Can @vglaidler or @perrygreenfield confirm? Are we going to try supporting this in the future? Or should I mark this as wontfix?

[vglaidler]

Indeed, pysynphot does not presently handle uncertainties at all. I don't think SYNPHOT does either, though I'm not sure.

I've had at least one conversation with John MacKenty expressing an interest in proper support for uncertainties, but it didn't get into the technical details of how that could be implemented or even clearly defined.

I might mark it "wishlist" or "maybe someday".

[pllim]

cc @tddesjardins @hover2pi