The (photo)chemistry and physics of sunscreens is no different than what occurs in UV absorbers in plastics (except that they are applied non-uniformly over the body, whereas UV absorbers are either mixed into the entire polymeric matrix or applied as a uniform coating. Given that and knowing we do a tremendous a tremendous amount of work with here at Aspen Research on UV degradation of polymer and prevention of that degradation using UV absorbers, I'm surprised that I never looked into SPF's before. Digging into the matter, I was pleasantly surprised by how well the science behind the SPF has been worked out. A key part of the calculation is that the activation spectra of the skin [*] is included in the calculation of the UVB absorption. From Wikipedia (and confirmed elsewhere in the open literature), here is the activation spectra of skin for sunburn, along with the intensity spectra of sunlight, and the product of the two: This is then used in the following formula for calculation of the SPF:
SPF is only calculated for the UVB spectra (280 nm - 315 nm), not UV A (315 nm - 400 nm). I imagine trying to patch together an activation spectra for UVA light would be challenging - not so much in preparing the spectra, but in trying to assess the relative impact of it compared to the SPF calculated for the UV B range. Bottles could be labeled with 2 SPF's, but that would be too confusing, wouldn't it?
[*] The activation spectra of a reaction is a description of how the kinetics change with wavelength. This is an extremely important concept that is often overlooked in most UV exposure testing. Most people assume that exposing a sample to shorter wavelengths will be fine for mimicking degradation in the natural environment. That only works for materials that absorb at the lower wavelengths. Unabsorbed photons do not lead to degradation, so in some cases, exposure to the wrong wavelengths will lead to completely erroneous conclusions about how well a part will survive. Always know your activation spectra!