In a comment on a previous post about lack of convincing evidence for climate models - the ones that predict catastrophe - I wrote:
At any time - right now, 5 years ago, 10 years, 15 years ago - the people who work with those models and claim we should pay attention to their predictions could make/have made a set of predictions: next year, the year after that, and so on. Then, as time passed, we would have found out if the models predict correctly. The modelers haven't done that.
From this talk by Richard Muller, a Berkeley physicist, I learned of two instances where the modelers did what I said they haven't done.
1. In 2009, James Hansen predicted, based on his model, that 2010 would be the hottest year on record. Since temperatures had been roughly constant - not rising - for the previous 12 years, this was an interesting prediction. When 2010 ended, Hansen's own data (analyzed in an unusual way, according to Muller) found this to be true, but two other datasets found it to be false.
2. In 2001, several scientists, based on nine climate models, predicted that Antarctic ice would increase over the next ten years. In fact, it decreased ("exactly the opposite of the prediction"). In response, John Holdren, President Obama's science advisor, said, according to Muller (at 26:22):
Well, those models were really wrong. But now we've changed those models. And now if we run them again they show that the ice will decrease. And therefore this is evidence in favor of global warming.
The audience tittered.
If you think Climategate was not important, and that the scientists whose email was revealed did nothing seriously wrong - as several official investigations, Bill McKibben ("if you managed to hack 3,000 emails from some scientist's account, you might well find a few that showed them behaving badly") and New Yorker staff writer Elizabeth Kolbert have concluded - you should see what Muller says about it (starting about 30:00).
I liked the talk. Muller makes several points with which I agree and presents helpful data. However, there were several things I didn't like. There is one big gap. Muller thinks current climate models are probably right, but doesn't explain why. I would have liked to know. And he says two things with which I deeply disagree. First, he says scientists shouldn't say what the data will show - that is, make predictions. I believe that making and testing predictions is by far the best way to learn how much we know. Second, Muller says that buying a Prius does nothing. The improvement is too small, the cost of a Prius too great. People in China can't afford a Prius, says Muller. This misses a really important point. When you buy a Prius you are supporting innovation. To solve the big problems that arise in any society, innovators need support. My theory of human evolution goes on and on about this. Long ago, connoisseurs supported innovation. Festivals such as Christmas supported innovation. Art lovers supported innovation. Fashion supported innovation. The great achievement of Tyler Cowen's new book The Great Stagnation, which I will discuss tomorrow, is its focus on rate of innovation. What controls rate of innovation is a supremely important question usually ignored by economists - as Muller ignores it.
In any case, here are two actual predictions and how they fared.