First the link to this week’s complete list as HTML and as PDF.
The long awaited comments on Bayon et al. (list of 2012-04-16) have finally been published. Neumann et al. don’t convince me though. She and Bayon agree on climate being the primary driver enabling the forest to be settled and she herself assumes the ease secondary forest can be cleared with to have been taken advantage of. All Bayon claims is the observed pulse of erosion being far in excess of what can be explained by climate alone and showing a clear anthropogenic signature. Maley et al.’s objections seem far more founded but are convincingly countered in Bayon’s reply.
Beall comments on a new (to me) assault on science’s integrity. In the 1920s scientists were not afraid of calling published rubbish what it was and instead of pussyfooting around like today made their criticisms sharp and to the point. If we’re not all to sink in the quagmire of unsubstantiated noise it’s high time for this openness to come back. And if some debates descend into name-calling like they did then (Nernst and Haber being an example of barely skirting the edge) so be it.
It seems Callaway widely overestimates the regularity of the molecular clock. A factor of two over a shorter time span seems quite acceptable to me. As Reich says the long-term average is well established. In the case of Homo and modern humans there are a number of well attested bottlenecks where many mutations may have become lost, giving the appearance of a lower rate of genesis.
Min et al. are somewhat unconvincing. Similar social status is far from the same as a similarly placid life. With palace intrigues and the fight to maintain rank the opposite is true. It is well known that the longer life of modern women vanishes when correcting for profession, a far stronger influence than sex.
I’m not sure if it’s Ashraf & Galor’s English or mine, but I don’t quite understand what exactly it is, their figures show. As far as I can tell in e.g. figure 5 the ordinate shows measured values corrected by all their regression results except genetic diversity and the abscissa not genetic diversity itself but its approximation from distance. So what do we see except a wide lot of scatter? We have somewhat low values for Africa, high ones for Europe and North America, and no effect of the independent value at all through the width of Asia. This leaves us with Papua New Guinea and a strong north–south dependency across the Americas. The former is exceptional in many ways and the latter, according to figure 1, the one area where the distance vs. genetic diversity connection fits least well if at all. So do I see any explanatory value in their hypothesis? I think not.
Gibbons adds one more argument in favour of last week’s hypothesis by Eriksson & Manica.
Rand et al. is one more example of the false use of standard errors in place of standard deviations empoyed to overstate a result in the social sciences. It may well be that sufficiently large groups converge in their averages, but what is claimed here is some sort of predictive power for random individuals, and that’s characterized by the standard deviation alone (provided the distribution is approximately Gaussian in the first place and does not require description through the median and quartiles). Still their result as such seems to hold.
A model with more parameters than data to calibrate them with becomes meaningless, can fit anything, and loses all predictive power. This simple truism is proved again by Wallace et al. and by Taylor et al. The latter shows how even widely off and opposite sign effects can easily be swamped by all the other parameters to make a model fit. The simple fact is that time and time again simple models with few parameters, those based only the sun alone are one example, have proved far more reliable in their predictions and extrapolations than any of the GCMs used by the IPCC.
In desperation I asked Fermi whether he was not impressed by the agreement between our calculated numbers and his measured numbers. He replied, “How many arbitrary parameters did you use for your calculations?” I thought for a moment about our cut-off procedures and said, “Four.” He said, “I remember my friend Johnny von Neumann used to say, with four parameters I can fit an elephant, and with five I can make him wiggle his trunk.”
Freeman Dyson, nature 427 (2004), 297
On the other hand the straightforward simple models of Sirocko and Spracklen do display genuine explanatory power. From Bayon, Spracklen and many others it becomes clear that anthropogenic climate change, through deforestation, ground sealing, river regulation (see New Orleans and Katrina), and soil degradation is a reality and it’s high time to tackle these actual problems instead of following the global warming chimera.
Hansen has cherry-picked his reference again. Global temperature fell from about 1940 to 1980 and the 1950–1980 average about equals 1925’s value. As Karl & Katz show 1950–1980 are the three decades with the least extremes in over a hundred years. So a growing variability in moving away from them is only to be expected. The general warming since about 1850 has never been in dispute, it is the regression from the Little Ice Age to the long term normal.