Read here. The North American Monsoon (NAM) is a fairly regular climate pattern that affects northern Mexico and southwestern U.S. The pattern is typified by a dry June that turns into a rainy July-September period.
Based on the relative simplicity of this climate pattern, one would expect that the unprecedentedly expensive, robustly powerful, cleverly sophisticated and devilishly complex climate models could at least provide accurate predictions of the NAM. New peer reviewed research finds that the global climate models are indeed worthless at such an assignment.
Cerezo-Mota et al. analyzed the simulated output of the IPCC's favored climate research concern (the Hadley Centre) and they found:
"The implication of their results, in the words of Cerezo-Mota et al., is that "precipitation in Arizona-New Mexico would not be correctly represented by a regional model driven by these GCMs." Thus, they write that RCMs driven by the "most commonly used" GCMs "would not give realistic simulations of the current climate of the region and therefore would not offer a realistic projection of climate change of the NAM." [Ruth Cerezo-Mota, Myles Allen, Richard Jones 2011: Journal of Climate]
Previous climate-model and peer-reviewed postings.