Piers M. de F. Forster and Susan Solomon of the National Oceanic and Atmospheric Administration analyzed temperature data collected from more than 10,000 surface stations over the past 40 years. Using information from stations operated by certified observers for which complete weeks of data were available the researchers calculated the difference between the highest temperature recorded during the day and the lowest recorded at night, the so-called diurnal temperature range (DTR). They determined that more than 35 percent stations in the U.S. had a significant difference between weekend DTR and weekday DTR on the order of several tenths of a degree. The direction of the effect was not always the same, however. Some cities (particularly those on the coast) exhibited higher DTRs on the weekends than during the weeks, whereas many in the midwest showed smaller DTRs on the weekends. Outside of North America, the magnitude of this weekend effect is smaller, Forster and Solomon report, with cities in Japan and China showing the largest swings.
The scientists tested the data set for both 28-day cycles (which could reflect a lunar influence) and random variations and found that neither could explain the findings. Because weekly cycles are rarely if ever found in nature, the observed fluctuations must therefore be anthropogenic in origin, the researchers write. In particular, they propose that cloud changes associated with aerosol particles in the atmosphere could be causing the weekend effect, though other pollution processes cannot be ruled out at this time. The authors conclude that "the data strongly support the view that human emissions play an important role in climate change and represent a key test for climate change theory."