The American Journal of Human Genetics has a paper in its pre-print section titled "A geographically explicit genetic model of worldwide human settlement history." I quickly skimmed it (and uploaded it into the GNXP forum). I have serious issues some of the inferences made in regards to the "obvious" fit of such coalescence data with a particular demographic history. I am convinced that meta-population dynamics tend to be ignored (in part because they are just another complication) even though they can also explain the data. Nevertheless, this jumped out at me:
We further neglected key events such as spatial and temporal environmental variation. Our results thus suggest that various environmental factors tend to be spatially relatively homogenous for human migration patterns, when considered over a large geographic distance.
As I stated above, meta-population dynamics, local extinctions and recolonizations, are issues that the authors seem to ignore when it comes to ignore their there results, especially given that environmental parameters are likely to be very relevant to marginal groups. They even allude to what seems like an abortive extra-African colonization in the Levant by anatomically modern H. sapiens sapiens which ended in local population extinction. But, the idea that humans are relatively buffered from environmental variation is roughly correct, and I'm talking about "humans" in a very broad sense. Our genus, Homo, broke out of the African continent almost immediately after its genesis. Dmanisi shows that beyond a shadow of a doubt.
