Straylight 3.0 cache (JWST ETC 5.1) and correction for 1.2 micron backgrounds

[ariedel]

There is now a fix (empirically derived correction, based on NIRCam results, for values below 1.2 microns (see Issue #25)
It has been implemented in the code that generated straylight cache 3.0 (now available at https://archive.stsci.edu/missions/jwst/simulations/straylight/sl_cache_3.0/) which means the actions that need to be taken are:

• Switch the location of the cache to query https://archive.stsci.edu/missions/jwst/simulations/straylight/sl_cache_3.0/
• Add an additional function to multiply the thermal background by this correction factor before using it

The ETC Team can provide the correction factor array and code to apply it (to mirror what is being done in the ETC's BMG)

[ariedel]

The correction factor: correction1p2micron.csv

code to apply the correction (once it's loaded from that file as correction_1p2_wave and correction_1p2_corr):

    def modify_background(self, thermal_int, wave_array):
        '''
        E. Han's modification to incorporate NIRCam measurements
        1.2 micron background correction (currently stored as precomputed arrays)

        Already included in the cache itself; only needs to be added to the thermal background

        thermal_int: thermal background flux to modify
        wave_array: the corresponding wavelength array
        '''

        bg_wvl_to_mod = np.asarray(wave_array)
        bg_flux_to_mod = np.asarray(thermal_int)

        # This correction comes in two precomputed arrays
        corr_factor = np.interp(bg_wvl_to_mod, correction_1p2_wave, correction_1p2_corr)
        thermal_flux = bg_flux_to_mod * corr_factor

        return thermal_flux

[mperrin]

As discussed in Slack, but copied here for reference. There is no need and no benefit to applying this correction to the thermal flux term. At the short wavelengths relevant for this correction (< 1.6 microns), the thermal background is a completely negligible term and can be considered zero for all practical purposes. In general the thermal term is << 1e-6 of the zodiacal light term at these wavelengths. Example plot:

So it's fine to not modify the thermal flux term. The only correction that's needed is to switch the JBT to use the updated cache 3.0.

[ariedel]

The work of this ticket is therefore only my first bullet point.