Police detectives aren’t the only people who look for fingerprints. Climatologists do, too: they’ve been looking for the collective fingerprint of humanity on Earth’s climate. Most of them suspect that the 6 billion tons of carbon we pump into the atmosphere each year, in the form of carbon dioxide, could warm the planet through the greenhouse effect. In the coming century the warming could be dramatic; but is it detectable already? This past year two teams of climate modelers said yes: man-made global warming is happening--almost certainly, anyway, and it’s getting more certain every year.
Certainty would be easier if it were just a matter of looking at the thermometer. We know that Earth has warmed by roughly a degree Fahrenheit in the past century, says Benjamin Santer, an atmospheric scientist at Lawrence Livermore National Laboratory in California, but you could have many different combinations of factors--volcanoes, the sun, carbon dioxide--that give you the identical temperature change. To exclude the natural suspects, researchers have been looking not just at the average global temperature but at the geographic pattern. The idea is that if we are warming the planet by polluting it, we’d produce a different temperature pattern than the sun would.
The two teams that said in 1995 that they’d found our geographic fingerprints (Santer’s at Livermore and a group at the British Meteorological Office) both used computer models. No other method is possible: you can’t put the planet in a laboratory and run experiments on it. And though they used different methods, they were successful for the same basic reason: they took into account that we are able not only to warm the planet but also to cool large regions of it.
That’s because each year we release not just 6 billion tons of carbon but 23 million tons of sulfur, mostly from fossil fuels and mostly in the form of sulfur dioxide. This gas turns into sulfate aerosols that reflect sunlight back into space even as carbon dioxide is trapping heat near Earth. The cooling effect of the sulfates is more regional--they tend to stay close to their sources, mainly in the Northern Hemisphere, while CO2 spreads around the globe--but in just the past few years it has become clear that they have a big impact on the geographic temperature pattern. The Livermore and British teams were the first to include the effect in supercomputer climate models.
The Livermore researchers first simulated the atmosphere with preindustrial levels of CO2 and measured the natural variability it might experience over the course of a few centuries. Then they added today’s levels of CO2 and sulfur. Overall, the combined gases did warm the planet, although not as much as CO2 would alone. But the more striking result came when the researchers compared the geographic temperature pattern predicted by their model for today’s polluted world with year-by-year records of the real world’s climate over the past 50 years. They found that with each passing year the real-world pattern grew more like the model--which makes sense, because the real-world levels of sulfates and greenhouse gases were climbing toward today’s levels. Santer and his colleagues calculate that the chances of this trend’s being a coincidence caused purely by natural climate variability--and unrelated to air pollution--are slim at best.
The British team reached essentially the same conclusion by a different approach; you have to simplify something to model climate even on a supercomputer, and the two teams chose different things. The British used a more realistic ocean than the Livermore group did, one that could transport heat to and from its depths, but a less realistic atmosphere: rather than re-creating the complicated chemistry of sulfate aerosols, they simply estimated how much sunlight Earth would reflect for a given level of sulfur emissions. Then they put in the actual measured increases of atmospheric carbon dioxide year by year since 1860 and tracked the response of the model. After 1950, the real-world temperature pattern conformed increasingly to the one predicted by the model--suggesting, just as the Livermore study did, that CO2 and sulfates were taking increasing control of climate. The chance of natural variability’s producing the pattern was less than 10 percent.
The British team let their model run into the future. As sulfur emissions rise and CO2 rises faster, they found, global temperature should rise 2.3 degrees Fahrenheit by the year 2050. Without sulfates, it would rise 3.3 degrees. (North America would get a bigger break--it would warm only 3 degrees instead of 4.5.)
Other climatologists have been praising these studies, but the two teams themselves are quick to point out weaknesses. The Livermore correlations between model temperature pattern and reality are much looser in winter and spring than in summer and fall. The British correlations work on the global scale, but when the researchers analyze a region like Europe or North America in detail, the correlations fall apart. And though both models now include sulfates, neither includes soot or other haze-producing hydrocarbons, which can either cool the planet or warm it. No one understands these effects well enough yet to put them in a computer model.
Yet the fact that both teams found the same strengthening pattern may nevertheless hint that our influence on climate is making itself felt. We’ve found emerging evidence that we’re beginning to see a fingerprint, but we’re not quite there yet is the cautious conclusion of John Mitchell, who led the British team. As far as understanding climate change goes, this is the end of the beginning, not the beginning of the end.