How Important Is the East Anglia Climate Data Set?
Let's say it's irredeemably corrupt. What would that mean for our understanding of climate change?
The focus of the e-mail hacking incident commonly known as "climategate" has shifted to whether scientists at East Anglia's Climate Research Unit threw away raw temperature data. A Sunday Times piece on the alleged information-dumping notes that CRU is "the world's leading centre for reconstructing past climate" and that the material in question "was used to build the databases … showing how the world has warmed by 0.8C over the past 157 years." How vital to climate change hypotheses is the CRU data set?
It's important, but hardly a sine qua non. Three organizations assemble global temperature data sets, which researchers then use to identify long-term trends: CRU, the National Oceanic and Atmospheric Administration, and the National Aeronautics and Space Administration. (The Japan Meteorological Agency also conducts similar work.) There are subtle differences among the sets, but they all point to the same general conclusion—that the earth has warmed by about 1.6 degrees Fahrenheit over the past century and a half.
The question of whether CRU dumped "raw" data is a little deceptive, because CRU, NASA, and NOAA don't put together first-order temperature measurements—that's up to various national meteorological services, which rely on satellites as well as thermometer readings on land and at sea. The National Weather Service in the United States, and equivalent organizations abroad, then sort through the numbers and clean them up. This cleanup operation is, in part, a form of proofreading, like if station agents in Siberia report a temperature of 102, they probably meant 10.2. It's also a "homogenizing" process that tries to account for the many variables that affect temperature over time—like when a population boom in a formerly rural area leads to an "urban heat island." A national weather service might adjust the data so that urbanization isn't mistaken for an increase in global temperatures. Much of these data are then stored at the Global Historical Climatology Network's database.
East Anglia's research unit uses a subset of this very large pool of information, while NASA and NOAA take slightly different pieces of it. NASA, for example, relies on data that are in the public domain. CRU takes the public numbers but also integrates more fine-grained data, which are sometimes governed by nondisclosure agreements. Each group then uses its chosen subset to create estimates of how global temperatures have changed over time and how they may change in the future.
The three groups account for data limitations in different ways. For example, each must deal with the fact that there are no permanent weather stations in the Arctic Ocean—making it difficult to get accurate readings. NASA's approach is to extrapolate temperatures from the nearest land-based stations, like those in Greenland. The much-maligned CRU doesn't "fill in" the Arctic Ocean in this way, which makes it seem as though the Arctic is warming at the same rate as the global mean. As a result, the CRU approach suggests there's been less warming over the last 10 years than does NASA's—something climate skeptics rallied around before they decided the set was tainted.
Got a question about today's news? Ask the Explainer.
Explainer thanks Richard Alley of Pennsylvania State University, Judith Curry of the Georgia Institute of Technology, Michael Schlesinger of the University of Illinois, and Gavin Schmidt of the Goddard Institute for Space Studies.
Become a fan of the Explainer on Facebook.
Juliet Lapidos is a former Slate associate editor.
Photograph of Greenland by Digital Vision/Getty Images Creative.