I recently discussed the ridiculous claims from Anthony Watts in a post at his WUWT blog, based on comments from Ryan Maue. Nonetheless, a reader seems to think he had a point. Specifically, he says this:
robertok06 | March 17, 2016 at 11:33 am | Reply
You may object, and I concur with you, that Mr Watts sometimes exaggerates and has an agenda, but there is one thing he discusses in his blog that you seem to forget… i.e. the almost perfect overlapping of the temperature trend at the tropics (where the El Nino is located) and global temperature, since Oct 2015.
Anything to say/comment on that pretty bizarre coincidence?
He further elaborates in a later comment:
Sorry, I didn’t “run” any correlation test, I simply judged the visual correlation of the two curves, global temp and tropical one, which seem to be out of sinch for a while last year and all of a sudden get in synch with one another… peak with peak, valley with valley, more or less.
The belief in “almost perfect overlapping” is based on this graph:
First: the “almost perfect overlapping” isn’t. There are times when peaks and valleys seem to be in sync, but there are also plenty of times when they’re not. The synchronicity isn’t any more than we would expect from randomness, because there’s another factor at work.
As another reader pointed out, the tropics as defined (latitudes 20S to 20N) cover 34% of the globe. That means that the data making up the red line (tropics) is 34% of the data making up the black line (globe). When one data set takes 34% of its value directly from the other, you expect them to be correlated.
Consider, for instance, these two curves — a simulated “global” curve in black, a simulated “tropical” in red, covering a 500 day period (about the same length as the original):
Notice how strongly they give the visual impression of correlation? Notice how peaks line up with peaks, valleys with valleys, even better than the Maue graph shown by Watts?
The red line is random noise (constructed with the same autocorrelation structure as daily temperature). The black line is the average of three such random noise sets, one of which is the red line. We end up with two random curves, with 33% of the “global” being the “tropical.” That overlap, combined with the quite nontrivial autocorrelation of these series, sure does make ’em look like they’re correlated.
By touting the apparent correlation of day-to-day fluctuations as meaningful, Watts & Co. are misleading readers, probably because they just don’t get it. It’s fine for Anthony Watts to be clueless about what’s going on. It’s not fine for him to pontificate from a position of ignorance.
But there’s something else to consider, something even more important.
Nobody disputes that el Niño influences global temperature. It has been established time and again in the peer-reviewed scientific literature, by lots of people, including me.
But Anthony Watts doesn’t just say that “El Niño is the driver of record high temperatures,” he also says “not carbon dioxide.” The implication is clear, Watts’ belief that CO2 (more generally, global warming) isn’t why we’ve seen temperature records get smashed, so it isn’t anything we should worry about.
We’ve had el Niño events before, lots of times. They bring extra high temperature, higher than the background level. But to get to the levels we’ve seen recently, it’s not enough to rise above the background level as much as el Niño makes happen. You have to raise the background level itself, by a lot — enough to beat the pants off all those other el Niño events. But el Niño doesn’t raise the background level.
Global warming does.
Imagine a basketball game of one-on-one between me and Michael Jordan. If MJ had a great breakfast, was in training, in top form, had practiced a lot recently, and was in the best possible spirits, it would improve his game. Not only would he beat me, he’d cream me by more than he had before.
If I then whined about how MJ beat me because he had a great breakfast, was in training, in top form, had practiced a lot recently, and was in the best possible spirits, but not because he was just plain better at basketball than I am, I’d be misleading you. If I actually believed it, I’d be misleading myself.
If you like what you see, feel free to donate at Peaseblossom’s Closet.