The measured increase in the Earth’s surface temperature during the 20th century is based upon thermometer measurements, which become increasingly incomplete further back in time. For example, at the beginning of the 20th century, thermometer measurements covered only 20 percent of the Earth’s surface, compared to more than 87 percent in 1987. Some greenhouse-warming dissenters have claimed that the gradual increase in coverage during the 20th century introduced an artificial warming trend into the temperature record, which accounts for most or all of the 20th century’s measured warming.

In an article titled "Effect of Mission Data on Estimates of Near-Surface Temperature Change Since 1900," in the July 1 edition of the Journal of Climate, LLNL researchers Philip B. Duffy, Charles Doutriaux, Imola Fodor and Benjamin Santer studied effects of the incompleteness of surface thermometer records on the estimated 20th century warming by examining 16 climate model simulations of the surface temperature changes from 1899 to 1998.

The scientists compared temperature trends obtained from globally complete model output with temperature trends derived by sampling the model output at only those locations where temperature observations are actually available. The comparison enabled the researchers to assess the effect of missing observational data on the apparent temperature trend during the 20th century.

"We found no evidence to support the hypothesis that incomplete observational data has caused us to overestimate the true warming trend," said Duffy, lead author of the paper. "On the contrary, our results suggest that the actual warming during the 20th century may have been slightly larger than the warming estimated from the incomplete observational data of -about 0.7 degrees Celsius instead of 0.6 degrees Celsius."

Livermore scientists examined climate models that incorporated estimated historical changes in both greenhouse gases and anthropogenic sulfate aerosols. Scientists concluded that in 10 of the 16 climate change simulations, missing data led to significant underestimates of the true global warming trend. In the remaining six simulations, missing data had no significant impact on the 20th century’s warming trend.

If the climate simulations are credible estimates of human effects on historical climate and of natural climate variations, it is extremely unlikely that missing observational data caused the 20th century’s warming to be overestimated.

"I hope that we’ve laid to rest the theory that warming that occurred during the 20th century is an artifact of missing data," Duffy said. "Knowing the accurate amount of the 20th century’s warming is important because if it were much less than we’ve thought all along, we would have to fundamentally rethink our ideas about global warming."

Founded in 1952, Lawrence Livermore National Laboratory is a national security laboratory, with a mission to ensure national security and apply science and technology to the important issues of our time. Lawrence Livermore National Laboratory is managed by the University of California for the U.S. Department of Energy's National Nuclear Security Administration.

Note: If no author is given, the source is cited instead.

• Global Warming
• Environmental Issues
• Climate

• Instrumental temperature record
• Consensus of scientists regarding global warming
• Temperature record
• Temperature record of the past 1000 years
• Climate model
• Greenland ice sheet