Here's a pattern I used to see back when I researched the infectious AIDS model: a scientist undertakes a study already "knowing" what the results ought to be. When he doesn't find them, he invents an excuse for it. "Most models predict that both precipitation and temperature will increase over Antarctica with a warming of the planet." But when his data show no increase of temperature or snow, of course this can't contradict the human-caused global warming model: "We're looking for a small signal that represents the impact of human activity and it is hard to find it at the moment," he said. In other words, he already knows that such a trend exists (but how would he know if it's human-caused?), even if he can't find it in his data.