See also a previous paper by Betti et al. Distance from Africa, not climate, explains within-population phenotypic diversity in humans.
The authors describe how the moderate effect of climate found in this study is compatible with their previous paper:
The fact that we detected a role for climate in driving between-population differentiation is not at odds with our recent finding on the same dataset that climate does not affect within-population phenotypic diversity (Betti et al., 2009). Between-population differentiation and within-population diversity can be affected in different ways, depending on the nature of the selective forces. For example, unless directional selection is exceptionally strong, it could shift a trait’s mean value without affecting within-population variability. This would translate in an effect of climate on between-population differentiation without leaving any signature on global patterns of within-population phenotypic diversity.In other words, selection may shift the nose height of arctic populations to lower mean value, but the dispersion around that mean will not be significantly less compared to populations from more temperate zones.
It is only extreme selection that might constrain phenotypes so strongly, resulting in a loss of variability, as phenotypes diverging greatly from the environmentally optimal phenotype will be discouraged.
The new study also identifies the main traits associated with climate:
Looking at single traits, the highest correlations with climate are observed for measurements of cranial and facial breadth, and measurements describing the nasal and orbital apertures. It has been suggested that the nasal index responds to selective pressures related to thermoregulatory breathing strategies (Carey and Steegmann, 1981; Franciscus and Long, 1991; Roseman, 2004). A thermoregulatory hypothesis has also been proposed to explain the pattern of brachycephalization in different populations, and the breadth of the cranium is shown to contribute highly to the degree of brachycephalization of the cranium (Beals et al., 1984).The impressive dataset of "4,666 male individuals belonging to 105 populations and 1,579 females drawn from 39 populations" repeesents the best, I believe, characterization of global cranial variation. I would certainly like to get my hands on it to repeat this experiment.
American Journal of Physical Anthropology doi:10.1002/ajpa.21115
The relative role of drift and selection in shaping the human skull
Lia Betti et al.
Human populations across the world vary greatly in cranial morphology. It is highly debated to what extent this variability has accumulated through neutral processes (genetic drift) or through natural selection driven by climate. By taking advantage of recent work showing that geographic distance along landmasses is an excellent proxy for neutral genetic differentiation, we quantify the relative role of drift versus selection in an exceptionally large dataset of human skulls. We show that neutral processes have been much more important than climate in shaping the human cranium. We further demonstrate that a large proportion of the signal for natural selection comes from populations from extremely cold regions. More generally, we show that, if drift is not explicitly accounted for, the effect of natural selection can be greatly overestimated.