Fucking "dysgenics"? What in the actual scientific racist eugenicist 1920s bullshit is this?!
I mean there's the fact that he's attempting to use IQ as his response variable without acknowledging that it is pretty flawed and heavily environmentally influenced.
Secondly... I mean come on, he's trying to relate intelligence to population genetics via admixture. It's kinda paradoxical to try and make a non-racist argument for intelligence differing significantly and across populations by genetics.
Thirdly specifically the phrasing "human biodiversity" is often used as a pretty strong dogwhistle by current scientific racists alongside ranting about replacement. We are really not at the risk of major genetic bottlenecks across the world right now. (Also biodiversity is a term used specifically to mean the richness and abundance of disparate species, it's fairly nonsensical when applied to a single species)
Bonus point for the quantitative biologists around: if you're resorting to pcas, you probably either don't understand the mechanisms behind what you're trying to show, or it is an effect only visible by considering the small effects of many other variables. Usually it's first worth some plotting followed by a glm (in this case a spatially explicit glm).