We last met Shaun Lovejoy when he claimed that mankind caused global temperatures to increase. At the 99.9% level, of course.
He’s now saying that the increase which wasn’t observed wasn’t there because of natural variability. But, he assures us, we’re still at fault
His entire effort is beside the point. If the “pause” wasn’t predicted, then the models are bad and the theories that drive them probably false. It matters not whether such pauses are “natural” or not.
Tell me honestly. Is this sentence in Lovejoy’s newest peer-reviewed (“Return periods of global climate
fluctuations and the pause”, Geophysical Research Letters) foray science or politics? “Climate change deniers have been able to dismiss all the model results and attribute the warming to natural causes.”
The reason scientists like Yours Truly have dismissed the veracity of climate models is for the eminently scientific reason that models which cannot make skillful forecasts are bad. And this is so even if you don’t want them to be. Even if you love them. Even if the models are consonant with a cherished and desirable ideology.
Up to a constant, Lovejoy’s curious model says the global temperature is caused by climate sensitivity (at double CO2) times the log of the ratio of the time varying CO2 concentration, all plus the “natural” global temperature.
There is no such thing. I mean, there is no such thing as a natural temperature in the absence of mankind. This is because mankind, like every other plant and animal species ever, has been influencing the climate since its inception. Only a denier would deny this.
Follow me closely. Lovejoy believes he can separate out the effects of humans on temperature and thus estimate what the temperature would be were man not around. Forget that such a quantity is of no interest (to any human being), or that such a task is hugely complex. Such estimates are possible. But so are estimates of temperature assuming the plot from the underrated pre-Winning Charlie Sheen movie The Arrival is true.
Let Lovejoy say what he will of Tnat(t) (as he calls it). Since this is meant to be science, how do we verify that Lovejoy isn’t talking out of his chapeau? How do we verify his conjectures? For that is all they are, conjectures. I mean, I could create my own estimate of Tnat(t), and so you could you—and so could anybody. Statistics is a generous, if not a Christian, field. The rule of statistical modeling is, Ask and ye shall receive. How do we tell which estimate is correct?
Answer: we cannot.
But—there’s always a but in science—we might believe Lovejoy was on to something if, and only if, his odd model were able to predict new data, data he had never before seen. Has he done this?
Answer: he has not.
His Figure shown above (global temp) might be taken as a forecast, though. His model is a juicy increase. Upwards and onwards! Anybody want to bet that this is the course the future temperature will actually take? If it doesn’t, Lovejoy is wrong. And no denying it.
After fitting his “GCM-free methodology” model, Lovejoy calculates the chances of seeing certain features in Tnat(t), all of which are conditional on his model and the correctness of Tnat(t). Meaning, if his model is fantasia, so are the probabilities about Tnat(t).
Lovejoy concludes his opus with the words, “We may still be battling the climate skeptic arguments that the models are
untrustworthy and that the variability is mostly natural in origin.”
Listen: if the GCMs (not just Lovejoy’s curious entry) made bad forecasts, they are bad models. It matters not that they “missed” some “natural variability.” The point is they made bad forecasts. That means that misidentified whatever it was that caused the temperature to take the values it did. That may be “natural variability” or things done by mankind. But it must be something. It doesn’t even matter if Lovejoy’s model is right: the GCMs were wrong.
He says the observed “pause” “has a convincing statistical explanation.” It has Lovejoy’s explanation. But I, or you, could build your own model and show that the “pause” does not have a convincing statistical explanation.
Besides, who gives a fig-and-a-half for statistical explanations? We want causal explanations. We want to know why things happen. We already know that they happened.