That makes sense, when you consider the immediate effects of geocatastrophes (The Toba eruption comes to mind, although not the best example) would tend to be limited by mountain ranges and other geographic features, as would disease outbreaks.
We also have to escape from the limiting vice of attempting to approximate prehistoric climates based on today's climates.
While most understand this at higher latitudes, where the effects of glaciation/warming were more imminently obvious (there was a kilometer thick ice sheet where I am sitting now, just a 10-12,000 years ago--only an inch of snow now), the effects also shifted more temperate and tropical regions as well. They were not the same then, and have varied since, both in short (catastrophic) spurts, and in long termed trends.
It is entirely possible that what we regard as Asian populations today had either not migrated into the region yet or had been severely thinned out by any of a number of natural killers.
That's about where ten years of reading and study in this area has lead me too.
Professor Stephen Oppenheimer says that the oldest (undisputed) Mongoloid skeleton ever found is only 10,000 years old.
It is my opinion that some group like the Jomon/Ainu produced todays' Mongoloid and Caucasians. The South East Asians (according to Oppenheimers study) produced the North Asian with the flatter face, peculiar eye lid, lighter skin and the 'shovel' teeth (Sindont) that are characteristic of all American Indians. I think Kennewick Man has sunadont teeth like south east Asians.