I agree with the theory that the headlight housing erosion is UV light based. One of the big wins for my old eyes with HIDs is that the extra UV pumps up reflective signage and road markings at night. After nearly 20k miles of HID, the erosion in my housing is much less than those pictures. Two dime sized areas of slight clouding in the plastic chrome above the bulb.
While the erosion
may be caused by UV light, and the HIDs
may emit more UV light than stock Halogens (which should be zero), if your eyes can pick up UV light bouncing off of signs then your eyeballs are significantly more advanced than mine.
Hint: UV (
Ultra Violet) is above the visible light spectrum. Any UV light being broadcast by a lamp would be completely wasted energy, therefore not desirable.
Furthermore, if the erosion is actually caused by light emission (UV or visible), then why is the erosion limited to just the area immediately above the capsules? The light intensity broadcast by the capsule should be about the same in a half circle arc around the top half of the reflector when in low beam position, and a full circle when in high beam.
The heat, however, would be more intense directly above the capsules, as the heated air rises in the otherwise still air confined in the headlamp bucket, hence my reason for thinking it might be heat related.
FWIW, I had very similar amount of reflector erosion (~ dime sized) after a similar amount of mileage (~20k) on my old headlamp buckets with 4300k HIDs.