Can you find the name of that photodiode? Maybe we can look it up to find the spectrals for it.
I notice they use a ultraviolet filter on top of the photodiode.
I can't see the actual diode for the filter and a case below it. Here is another photo:
It looks like it would be pretty easy to remove the whole sensor assembly to put it into a different, and waterproof case. I think the white plastic dome, the cosine filter? - has to be the only optical thing between the light and the sensor, so I couldn't have that part inside the watertight case. Is that correct? The only effect I can figure out from having a layer of acrylic covering the dome is reflection off the flat layer, but there is reflection off the dome too, so I can't understand why that does much harm.
The 60.6 conversion factor could change if there is acrylic over the dome, I think, but that's not a problem either. And, if the conversion factor is accurate to +/- 5% that would be outstanding.