The year 2014 had scarcely expired before being declared the warmest year on record. First off was the Japan Meteorological Agency. The year 2014 surpassed 1998 to set a new record by all of five one hundredths of one degree Centigrade, according to the agency’s preliminary numbers. Then Britain’s Met Office announced that 2014 was the warmest year in the 355 years of the Central England Temperature series.
Each year, global-warming adherents anticipate an El Niño (a strong warming phase in the Pacific) as the physical manifestation of global warming’s Second Coming to herald the end of the barren years of flat-lining global temperatures. The Center for American Progress’s Joe Romm called the 2014 record doubly impressive. As Romm noted, “We’re still waiting for the start of El Niño” but got a temperature record nonetheless.
Despite seeing a record-breaking year overall, Britain had no major heat waves, and no new monthly records were set in 2014. Instead, each month was consistently warm, only one having below-average temperatures, and the year seeing the lowest number of frosts since 1961. Similarly with rainfall: Although 2014 was one of the 20 wettest years since 1766, no individual region had its wettest year on record. After a stormy January and February in Britain, the rest of the year was “relatively quiet,” as the Met Office describes it.
This isn’t part of the narrative being spun by Naomi Oreskes, Harvard professor of the history of science. Previewing the hottest-year-on-record announcements, Oreskes wrote in the New York Times last week that we were underreacting to the reality of dangerous climate change “now unfolding before our eyes.” The burden of proof should be lowered, Oreskes argued, but her excursion into statistical methodology to buttress this contention was widely panned (here, here, and here).
In science, models are used to produce predictions that can be tested against nature and thereby advance scientific knowledge. On this basis, the clear inference of the disagreement between climate-model simulations and observations over the past two decades is that scientists’ current understanding of the climate system is faulty. Climate science is not normal science: It has become the leading branch of global therapeutics. Climate-model outputs are used as a tool to win the political battle for policies believed necessary to save a dying planet.
Arguing for a relaxation of confidence levels (misdescribed by Oreskes the polemicist as burden of proof) to justify such policies, Oreskes the historian misattributes the widely used 95 percent confidence level to R. A. Fisher, the British statistician. Biologist Richard Dawkins once described Fisher, possessed of a towering intellect, as the father of modern statistics and, for his work in genetics, Darwin’s greatest 20th-century successor. Rather than the 95 percent confidence level, to Fisher belongs the credit of formulating the null-hypothesis test: For there to be a relationship between two variables, it must first be demonstrated that the null hypothesis — that the two variables are independent of each other — is false. As Fisher put it, every experiment may be said to exist only in order to give the facts a chance of disproving the null hypothesis.
The null hypothesis is widely used across the physical and social sciences, but not by the IPCC. In the approximately 1,500 pages of the IPCC fifth assessment report (2013) on the physical basis of climate change, it merited three paragraphs. Whether global temperatures rise by up to 4.8 degrees Centigrade by the end of century, as the IPCC believes, or, in line with 1900–2012 trends, rise by 0.7 degrees Centigrade by the century’s end, there will continue to be years that set new temperature records. However, this would tell us next to nothing about whether such events portend catastrophe or are essentially harmless.
There is a non-temperature trend that should worry alarmists. Since March 2001, Gallup’s environment poll has surveyed voters annually on the seriousness of global warming. In the first years of the century, the percentage who said global warming was exaggerated was generally in the low 30s (2004 saw a spike to 38 percent). The percentage then rose in the run-up to the December 2009 Copenhagen climate conference and peaked three months later at 48 percent. Since then, the percentage has slipped back to the low 40s — the skeptics, as they might be called, gaining ten points on the prior decade.
With the intensifying drumbeat of alarm and exaggeration anticipating this year’s Paris climate conference in December, climate-change alarmists face a conundrum: The more they act true to form, the more voters become skeptical and, in the United States, the greater the political incentives for Republicans to block climate-change policies. As yet, the alarmists don’t seem to have gotten the message.
— Rupert Darwall is the author of The Age of Global Warming: A History (Quartet, 2013).