I had started on an essay to describe this meteostatistical failure last night, but Lucia beat me to it, so the credit goes to her. She’s a sharp lady. Besides the fact that short term weather events are not climate you can’t apply the coin flip statistical logic to it like Weather Underground’s Jeff Masters did, citing a calculation that suggests a one in 1,594,323 chance heat wave. Lucia reduces this probability estimate calculation to rubble and writes:
In fact– the entire reason weather forecasting is possible at all is that we know weather patterns persist. Master’s calculation is based on the assumption weather patterns do not persist!.
As we’ve said many times before , the Eastern US heat wave was a result of a quasi-stationary blocking high pressure pattern that persisted a few days. It has already moved on and has been absorbed in the weather noise.
But the wailing about the significance of the recent heat wave has reached a fever pitch, blinding rational people who should know better in their quest to show that climate and weather are the same thing. They aren’t, and they never will be. The same folly of opinion occurred two years ago in the summer of 2010 during the Russian Heat Wave, with many of the MSM and pundits saying that its was a sure case of global warming affecting the weather. Then, NOAA published a peer reviewed paper holding global warming/climate change blameless, basically saying it was nothing more than a persistent weather pattern. A follow up paper by other scientists confirmed it was due to natural variability. But the people who believe that AGW will be dangerous and world changing can’t let go of the idea when it comes to blaming short term weather patterns on global warming.
Dr. Jeff Masters wrote of the recent heat wave:
Each of the 13 months from June 2011 through June 2012 ranked among the warmest third of their historical distribution for the first time in the 1895 – present record. According to NCDC, the odds of this occurring randomly during any particular month are 1 in 1,594,323. Thus, we should only see one more 13-month period so warm between now and 124,652 AD–assuming the climate is staying the same as it did during the past 118 years. These are ridiculously long odds, and it is highly unlikely that the extremity of the heat during the past 13 months could have occurred without a warming climate.
What this really is is a meaningless statistics.
Let’s tweak Dr. Master’s rather imperfect calculation by retaining his assumption that “climate is staying the same as it did during the past 118 years” but accounting for “persistence”. More specifically we will assume that there is non-zero serial auto-correlation in the monthly data. Since he used “white” noise, I’m going to pick the next simplest model: Red noise (i.e. AR1 noise.)
She runs her calculations and concludes:
Taking the mean of the series, I found that “assuming the climate is staying the same as it did during the past 118 years” the probability of 10% ± 0.4% the final 13 months would fall in the top 1/3rd of historic temperatures observations.
A 1 in 10 (10%) probability is a statistical galaxy away from a one in 1,594,323 chance.
And even other warmists agree, Lucia adds in this update:
Update: I googled to read who’d blogged. Michael Tobis commented:
Actually that’s bad form from both Masters and NCDC.
1.6 million (more precisely, 1,594,323) to one is just the thirteenth power of 1/3, which overstates the case to the extent that successive monthly anomalies are correlated. (Also the 1/3 is somewhat arbitrary and could be a cherry pick, but leave that aside). I don’t doubt that something very odd is going on but the number represents a common elementary statistical error and is in this case excessively alarmist.
Is this the first time MT and I have agreed on something?
Unfortunately, the damage is done, and Dr. Jeff Masters million to one lie is all over the net, aided by an unquestioning press.
Read Lucia’s full explanation here.