August 11, 2005

Anyone still holding onto the idea that there is no global warming ought to hang it up. All data sets—satellite, surface, and balloon—have been pointing to rising global temperatures. In fact, they all have had upward pointing arrows for nearly a decade, but now all of the data sets are in closer agreement due to some adjustments being published in three new articles in Science today.

People who have doubted predictions of catastrophic global warming (and that includes me) have long cited the satellite data series derived by climatologists John Christy and Roy Spencer at the University of Alabama Huntsville (UAH). That data set showed a positive trend of 0.088 degrees centigrade per decade until recently. On a straight line extrapolation that trend implied warming of less than 1.0 degree centigrade by 2100.

A new article in Science by researchers Carl Mears and Frank Wentz from Remote Sensing Systems (RSS) identified a problem with how the satellites drifted over time, so that a slight but spurious cooling trend was introduced into the data. When this drift is taken into account, the temperature trend increases by an additional 0.035 degrees per decade, raising the UAH per-decade increase to 0.123 degrees centigrade. Christy points out that this adjustment is still within his and Spencer's +/- 0.5 margin of error. What's the upshot? Although reluctant to make straight-line extrapolations, Christy notes in an e-mail, "The previous linear extrapolation indicated a temperature of +0.9 C +/- 0.5 C in 2100, the new data indicate a temperature of +1.2 +/- 0.5 C."

However, the Remote Sensing Systems team has made some additional adjustments, such that their global trend is 0.193 degrees per decade. Christy and Spencer disagree with those additional RSS adjustments, but acknowledge that it's an open scientific question which team is correct. If RSS is right, a straight-line extrapolation of future temperature trends implies that global average temperatures in 2100 will be about 2.0 degrees centigrade (3.6 degrees Fahrenheit) warmer than they are today—more than double the original Christy and Spencer trend. The RSS trend is more in accord with the higher projections of future temperature increases generated by climate computer models.

Is there a way to tell which data set is more accurate? Long term weather balloon data provide an independent measure of temperature trends; however, they also have some problems. Another of the Science articles looks at daytime biases in the radiosonde balloon data sets. A team led by Yale University climate researcher Steven Sherwood, suggests that researchers overcorrected for temperature increases caused by daytime solar heating of the instruments, and thus projected a spurious cooling trend. The researchers acknowledge that there are also nighttime biases, but do not correct for those in this article, coming to the not very robust conclusion that "the uncertainty in the late 20th century radiosonde trends is large enough to accommodate the reported surface warming."

The UAH temperature data set differs from a set of six different recent analyses of weather balloon radiosonde data by range from a low of 0.002 degrees centigrade to a high of 0.023 degrees centigrade. All are well within the +/-0.5 degree margin of error for the adjusted UAH data and lower than the adjusted RSS temperature trend. In other words, the balloon data suggest the global temperature trends are closer to the UAH number than they are to the RSS number. In its article, the RSS team agrees, "Trends from temporally homogenized radiosonde data sets show less warming than our results and are in better agreement with the Christy et al. results."

But what about the future? As the U.S. National Oceanic and Atmospheric Administration notes, "taking into account uncertainty in climate model performance, the IPCC [UN Intergovernmental Panel on Climate Change] projects a global temperature increase of anywhere from 1.4–5.8°C" by 2100.

So what's the bottom line? The UAH team finds warming of 0.123 degrees per decade. The balloon data tend to support the UAH team's findings. The RSS team finds warming of 0.193 degrees per decade. And the surface measurements show a warming trend of 0.15 degrees per decade.

Christy notes, "If you want to say model trends are bolstered, you must remember model trends are all over the map. Which trend is bolstered? Perhaps you want to say those model trends less than 0.2 C per decade are bolstered." Right now the available data sets appear to strengthen the case for arguing that the lower-end model projections for future temperature increases are more likely ones. Christy concludes, "The new warming trend is still well below ideas of dramatic or catastrophic warming."

Ronald Bailey is Reason's science correspondent. His book Liberation Biology: The Moral and Scientific Case for the Biotech Revolution is now available from Prometheus Books.

Subscribe to Reason today! It's easy. Do it right from your computer. Subscribe to the print edition or the electronic edition.