rppvt - Right. The living tissue there is very thin over the white skeleton and lacking zooxanthellae, hence a pale color.
As for which colors are protective, Dunlap and Schick (1998), "Ultraviolet radiation-absorbing mycosporine-like amino acids in coral reef organisms: a biochemical and environmental perspective" goes into it pretty well. Mycosporine-like amino acids (named so because they were first identified in a fungal fruiting body) absorb light very strongly in the 310-360nm range, which covers both ultraviolet A and ultraviolet B radiation. They're found not only in corals but also in sponges, anemones, algae, flatworms, fish and others. There are 19 distinct MAA compounds. They also state that blue light and UV induce MAA production but red or green light does not. (As a side note that I found really cool, they note that a possible explanation for nighttime spawning in corals is that the coral sperm are too small to contain enough MAAs to protect them from UV). The source of MAAs, whether the coral or the zooxanthellae, seems unclear from this paper (though I would say it leans toward production by the zooxanthellae) so I may have been mistaken about that earlier when I said they were produced by the coral. If they are produced by the zooxanthellae they are likely transported through the coral tissue. Note that there are other colored proteins aside from MAAs that definitely ARE produced by the coral host, including a green fluorescent protein and pocilloporin, which is pink and found in Pocilloporidae.
These are my own thoughts: Given that they are absorptive to UV, it would make sense that these pigments would generally appear to be blue or green, since they would absorb photons in the 310-360nm range and fluoresce at a slightly longer wavelength, 400-500 is a blue-green color.