Climate argument solved?
Technical errors blamed for mismatch in temperature readings.
The end may be in sight for a 15-year argument over a discrepancy in the data on global warming.
Three papers published in Science today say that temperature trends in the lower atmosphere are consistent with a warming world, countering earlier claims to the contrary.
One study deals with satellite measurements, the second with data from weather balloons and the third with predictions of climate models. "Taken together, these three results are a major step forward," says Carl Mears of Remote Sensing Systems in Santa Rosa, California, an author of one of the papers.
The problems began in 1990, when an analysis of satellite observations showed the troposphere - the lowest few kilometres of the atmosphere - was warming too slowly compared to the surface for climate models to be correct1. Global-warming sceptics seized upon the result.
"It has been the main crutch of the sceptics when it comes to pooh-poohing global warming, with some success," says Kevin Trenberth, from the National Center for Atmospheric Research in Boulder, Colorado. Some politicians have also cited the data as evidence of the uncertainties in global warming.
The 1990 study, headed by John Christy from the University of Huntsville, Alabama, was criticized for the way it used satellite data. This prompted multiple rounds of revision, but never really solved the problem. A recent re-analysis in Nature2, which claimed to solve the debate, also fell short of being definitive.
So why pin any hopes on the new papers? "We are converging, we are definitely getting closer," says Mears.
He and Frank Wentz, also at Remote Sensing Systems, tackled a problem with how the satellite data are corrected for the time of day. Although the satellites nominally pass over the same point at the same time each day, in practice drag causes their orbits to sink and the time to drift.
Christy's group estimated the temperature at the exact time they wanted, rather than when the satellite actually was overhead, by looking at temperature measurements the satellite took to the east and west of its position. From this they concluded the troposphere was warming by about 0.09 °C per decade.
Mears and Wentz instead used data generated by a complex model of the atmosphere to adjust the satellite measurements. On doing so, the troposphere suddenly appears to be warming by almost 0.2 °C per decade, in agreement with climate models3.
The new work is bound to draw its own criticism, however. "It's going to be very interesting to see how this reverberates through the climate-sceptic blogosphere. I expect by Tuesday there will be plenty of articles calling me a fraud," says Mears.
The second paper points out problems in the temperature record from weather balloons. Steven Sherwood and co-authors argue that changes in instruments have made the records untrustworthy4. The problems arise because different manufacturers' instruments heat up by different amounts during the day. At the moment, they say, the errors on this data are so big that one can't pin down how much the troposphere is warming.
The third piece of research, led by Ben Santer from the Lawrence Livermore National Laboratory in California, concludes that disagreement between 19 climate models and measurements are more likely to be due to errors in measurements, rather than models5.
"I don't think this will be the last word," says Phil Jones, a researcher from the University of East Anglia, UK, who has looked at the new results for the next assessment of the Intergovernmental Panel on Climate Change.
"Certain people have too much invested to admit they accept all these reports," Jones says. "But there's quite a bit of agreement in the whole community that these papers are getting closer to the truth."
- Spencer R. W. & Christy J. R. Science, 247. 1558 - 1662 (1990).
- Fu Q., Johanson C. M., Warren S. G. & Seidel D. J. Nature, 429. 55 - 58 (2004).
- Mears C.A. & Wentz F.J. Science published online, doi: 10.1126/science.1114772 (2005).
- Sherwood S., Lazante J. & Meyer C. Science published online, doi:10.1126/science.1115640 (2005).
- Santer B. D., et al. Science published online, doi: 10.1126/science.1114867 (2005).