Tuesday, October 06, 2009

The More You Read Krugman, The Less You Know

I was reading this post today about my favorite partisan hack (h/t The Anchoress) when I came upon this quote from our hero made in 2006:

...it seems to me that [NASA climatologist Dr. James Hansen], whose predictions about global warming have proved remarkably accurate, didn’t believe that he could successfully be portrayed as an unreliable exaggerator.[emphasis added]


Heaven forbid, we criticize the guru of Anthropogenic Global Warming. After all, Krugman seems to regard him as a veritable Kreskin of prognostication.

So, I went back and looked at what Krugman feels represents "remarkably accurate" predictions from Dr. Hansen. Back in his famous 1988 testimony before Congress Hansen presented the following set of predictions:



So, now we are 22 years into this prediction. How did he do? Well, here is the UAH data on Global monthly average lower troposphere temperature since 1979:



So, let us do a little reality check for Dr. Hansen's predictions. He gave himself three potential scenarios (A, B and C), so we will look at all of them at 4 year intervals (1992, 1996, 2000, 2004, 2008). Luckily, 1988 in the UAH data is practically at zero, so we can just go with the numbers. (These are the running 37 month average and predicted values pulled from the graphs themselves, so they do lack a little specificity, but not enough to obscure the basic point.)

1992:

Observed: -0.05
Scenario A: 0.5 (predicted value off by 1100%)
Scenario B: 0.4 (predicted value off by 900%)
Scenario C: 0.35 (predicted value off by 800%)

Hmm...not great. But maybe Hansen will rally.

1996:

Observed: 0.05
Scenario A: 0.6 (predicted value off by 1100%)
Scenario B: 0.4 (predicted value off by 700%)
Scenario C: 0.2 (predicted value off by 300%)

2000:

Observed: 0.175
Scenario A: 0.9 (predicted value off by 414%)
Scenario B: 0.55 (predicted value off by 214%)
Scenario C: 0.55 (predicted value off by 214%)

2004:

Observed: 0.26
Scenario A: 0.95 (predicted value off by 265%)
Scenario B: 0.75 (predicted value off by 188%)
Scenario C: 0.5 (predicted value off by 92%)

2008:

Observed: 0.2
Scenario A: 0.95 (predicted value off by 375%)
Scenario B: 0.85 (predicted value off by 325%)
Scenario C: 0.65 (predicted value off by 225%)

OK, so obviously Scenarios A & B suck pretty badly, but Scenario C isn't too bad, right? (Relatively speaking, of course.)

Well, no. Sorry to be a bastard about it, but I didn't tell you what these "scenarios" are supposed to represent.



So, of these options, Scenario A would seem to be closest to the actual situation vis-a-vis trace gas emissions. Scenario C would only apply if drastic cuts had been enacted.

So Krugman was greatly impressed by predictions that averaged being wrong by a mere 651%.

That explains much.

No comments: