It’s been slow posting as I’ve headed back to school to combine my meteorological experience with some exposure to policy. I’m buried under tons of reading, most of which is good and will spawn plenty of posts in the future when I have more time. In the meantime, I wanted to point out a piece of writing that is absolute trash.
From the “Science and Public Policy Institute” (a D.C. “think tank”), I found a piece entitled 2011 U.S. Temperature Update: Alarmism Not by Chip Knappenberger. The purpose of this paper is to convince policymakers that temperatures in the U.S. aren’t noteworthy and evokes the frame of global warming alarmism. The methodology of the paper is the bothersome aspect.
Chip starts by showing a graph of U.S. surface temperature data from 1895-1997. Any guess why? Because the most intense El Niño in recorded history occurred in 1998 and denialists like to use data since 1998 to try to show that global warming is not occurring. Count Chip among this crowd, as his next graph includes surface data up to 2007 in an attempt to set up people who thought a new phase of temperatures across the U.S. had started as alarmists. He does this by showing his next graph of surface data up to 2010. Since 2008-2010 was cooler than the 1998-2007 period, Chip states that “the warm period was starting to look out of place”.
All of this is an introduction to his goal, which is to predict that 2011 will continue to be cooler than the aforementioned warm period. How specifically does Chip try to do this? By looking at a distribution of the difference between the annual temperature anomaly and the temperature anomaly observed during the first 6 months of the years 1895-2010. What does his distribution show? Supposedly, the mean of the data is 0°F. Then, even though it’s not, he claims the data is close enough to having a normal distribution that he can use the data as he pleases.
More egregiously, Chip does something that makes absolutely no sense. He uses that anomaly distribution to predict what 2011′s average U.S. temperature will be by combining the distribution alongside the same time series he used before. In other words, he is saying that based on the past 100+ years’ of data, 2011 is likely to be no more anomalously warm than 2008-2010. Put yet another way, he is claiming that climatology is the best predictor of this year’s average U.S. temperature.
Beyond the nonsensical use of statistics to prove an ideological point, allow me to provide additional information regarding our year-to-date temperature. Through June (the same data Chip used), the U.S. had recorded “Above Normal” temperatures: 79th highest out of 117 years. Giving Chip the benefit of the doubt that he wrote the paper in the middle of the year and conditions might have changed one way or the other, I will politely point out that through August, the U.S. had recorded the 92nd highest temperatures out of 117 years. Despite the Pacific Northwest and northwestern states recording below-normal temperatures, almost the remaining 2/3 of the continental U.S. were above normal, with 8 states at much above normal and 1 state recording the hottest year-to-date on record (Texas). Despite a cool start to the year, the number of heat records broken are outpacing cold records broken year-to-date by an incredible 3.4 to 1 ratio. Those kinds of things don’t happen in non-anomalous years. But it’s really the shoddy use of math that irks me.
As a side note, a paper by one of Chip’s colleagues read for one of my classes. Written by the President of the “think-tank”, it was filled with so many spelling and grammatical errors, the writer needs to do some serious thinking before publishing another paper. Or as some undergraduates are learning this semester, Word comes with a spelling and grammar check. Use it or you look like you don’t know or don’t care what you’re doing. These guys are trying to influence public policy?!