>The dose response is not linear, there is no “instantaneous damage” below some threshold. Your argument assumes something that isn’t true.

Source? All things being equal, I'd expect half the UV damage by going from 98% UV protection to 99% UV protection. That's significant even though the protection only went up by only 1%. Moreover as I mentioned in my previous comment, even if you assume that 2% UV exposure (from 98% protection) basically never results in skin cancer, that figure is only achieved if you use sunscreen perfectly, which no one really does.

Melanin production is stimulated by UV so over time the response is absolutely non linear.