Everyone Freaks Out About Two New Climate Change Studies

New reports on global temperature trends cause a predictable media uproar


How fast is the planet warming? Two new not-yet-peer reviewed studies were published online earlier this week that suggest somewhat different answers. The first agrees with earlier findings about the pace of global warming, while the second argues that half of the recent warming in the U.S. is artificial.**

The first is from the Berkeley Earth Surface Temperature (BEST) project run by physicist Richard Muller. That paper, "A New Estimate of the Average Earth Surface Land Temperature Spanning 1753 to 2011," reported that "the rise in average world land temperature globe is approximately 1.5 degrees C in the past 250 years, and about 0.9 degrees in the past 50 years." The BEST group concluded that the increase in carbon dioxide emissions from humanity was the likely cause of the increase in global average temperature. In its study, BEST claimed to have taken into account "issues raised by skeptics, such as possible biases from urban heating, data selection, poor station quality, and data adjustment. We have demonstrated that these do not unduly bias the results."

At the same time, Anthony Watts and his colleagues published their new study, "An area and distance weighted analysis of the impacts of station exposure on the U.S. Historical Climatology Network temperatures and temperature trends," which found "for the years 1979 to 2008, yields a trend of +0.155C per decade from the high quality sites, a +0.248 C per decade trend for poorly sited locations, and a trend of +0.309 C per decade after NOAA adjusts the data." The press release for the new Watts study suggests that weather station quality would be a significant issue that undermines the higher global average temperature trends implied by the data reported in Global Historical Climatology Network and in the BEST network.

As is usual in the debates over issues in climatology, the online discussion was measured and polite. Left-wing climate blogger Joe Romm over at ClimateProgress calmly headlined the new BEST study, "Bombshell: Koch-Funded Study Finds 'Global Warming Is Real', 'On The High End' And 'Essentially All' Due To Carbon Pollution." With regard to the new Watts study, the folks over at the right-wing site serenely reported, "New Study Crushes Global Warming Data Claims."

So there you have it—a crushed bombshell!

The Watts study applies a nifty new classification system for the quality of weather station siting devised by French researcher Michel Leroy and adopted by the World Meteorological Association. For example, in order to be considered a class 1 or 2 station no artificial surfaces like parking lots or brick walls must be within 100 meters or 30 meters respectively. Such surfaces spuriously boost the amount of warming that a weather station thermometer would detect. For years, Watts has tirelessly analyzed the quality of weather stations in the United States to identify how their siting might bias their data. In his new study, Watts and his fellow researchers report that U.S. class 1 and 2 weather stations find that temperatures are increasing at a rate of +0.155 per decade in the continental U.S. Poorly sited stations (classes 3, 4, and 5) show a +0.248 per decade trend.

In an email, Watts additionally noted, "We identified the most representative thermometers for climate trend capture are the well sited, non-airport, Class 1 and 2 rural MMTS stations, which have a superior instrument shelter, an electronic memory to capture the high and low, and are removed both from the siting issues as well as urban heat island and airport growth and instrumentation problem effects." Other considerations include alterations to how the land in rural areas [PDF] is used by people. The trend for those stations is +0.032 C per decade. Watts added, "The value is not much greater than zero."

Over the years, the folks at the National Oceanic and Atmospheric Administration have tried to adjust the temperature data from the U.S. Historical Climatology Network to take into account many factors—urbanization, station and instrument changes, airports—that could affect the accuracy of the temperature record. Watts and his fellow researchers claim that while the well-sited stations show an overall trend of +0.155 C per decade and the trend at poorly sited stations is +0.248 C per decade, NOAA adjusts the data so the reported trend is +0.309 per decade. Watts reckons that the problem is that NOAA researchers use temperature data from poorly sited stations to adjust upward the data from the well-sited stations.

"This disparity suggests that a combination of siting issues and adjustments are creating a spurious doubling of the U.S. surface temperature record for the 30 year period of this study," concludes the Watts paper. Via email, Watts generously observes, "I don't believe the errors associated with NOAA adjustments are deliberate, but simply a case of confirmation bias. They expect to find global warming because a popular theory says they should."

Specifically with regard to the BEST study, the Watts study notes, "Given that USHCN [U.S. Historical Climatology Network] stations overlap and are part of the GHCN [Global Historical Climatology Network], the siting issue should be examined for all the GHCN and BEST sites." Translation: The BEST data very likely include a spurious warming trend of which they have not taken adequate account. Via email, Watts also points out that BEST study lumps class 1, 2, and 3 sites together as OK. "Only stations classified as 1 and 2 are acceptable," Watts argues. "I expect that errors such as these have contributed to the failure of all BEST papers in peer review."

Meanwhile, Muller and his colleagues report that their analysis of temperature data shows that the globe has warmed +0.87 degrees C over the past 50 years. Via email, Elizabeth Muller, co-director of Best with her father observes that they "have not looked at the new analysis by Watts in great detail," but add "as near as we can tell we are in agreement with his results. He is critical of our methods, but reached the same conclusions we did—that station quality does not unduly bias the temperature record." When they get around to looking in more detail, they will find that Watts and his colleagues do think station quality problems have in fact significantly biased their temperature records. In any case, Muller notes that an earlier study by BEST analyzed temperature data from only stations that are far from urban areas. "We found that in these very rural areas, there was the same amount of warming as for the global land," she writes.

To get some idea of the magnitude of the trends over which the argument is being conducted, let's do some rough calculations. The BEST trend is +0.87 C over the past 50 years. That implies a temperature increase of +0.174 degrees per decade. The satellite record from the University of Alabama Huntsville researchers John Christy and Roy Spencer finds a per decade increase of +0.14 C. The new Watts study looking only temperature trends for the continental U.S. of +0.155 C per decade.

Admittedly, these very rough calculations are a bit like comparing peaches to nectarines. For example, the satellites measure the temperature of the lower troposphere not the surface. But looking at the Watts study one finds that according to the satellite data, the U.S. tropospheric temperature trend is +0.24 degree C per decade. Watts notes that surface warming is typically amplified in the troposphere by a factor of 1.1 to 1.4, which implies surface temperatures "in the range of 0.17 to 0.22, which is close to the 0.155 degrees C per decade trend seen in the compliant class 1 & 2 stations." (A puzzlement: if Watts' best-sited stations exhibit a trend of +0.032 per decade and the amplification factors are applied, that would imply a U.S. tropospheric temperature per decade trend of +0.035 to +0.045, yet the satellite trend of +0.24 is more than five times that.)

The good news is that both Watts and BEST have been completely transparent about their calculations and their data. This should make it possible for each to check each other's work and get back to the rest of us with their results. They both are aiming to have their work published in peer-reviewed journals as well. In the meantime, I fully expect that the public conversation over climatology will continue to remain as just civil as it always has been.