This is fairly typical for genetic studies. We can observe an association, without understanding the causal relationship. Those of us who take the molecular biology/biophysical view of genes and proteins have been watching the geneticists slowly discover that "the allele of a gene that codes for" mental model is very limited; at least they're starting to pay more attention to the regulatory changes and indels and structural rearrangements, even if they don't have the tools to explain the underlying mechanism.

Like most hair/skin coloring, it's related to melanin production, regulation, and distribution.