The world is getting hotter, and humans are at least partly to blame--but that's about all that climate researchers can confidently say about global warming. Their bottom-up approach--trying to understand the role of every part in the dizzyingly complex climate machine--has left the crucial question of how bad things could get unanswered. In the 4 January issue of Science, a group of researchers describe a new top-down approach: They plugged different combinations of values for fundamental properties of the climate system into a computer model and looked to see how well the model's output matched long-term observations. The results suggest that global warming is probably a serious threat and that it could get even hotter than most scientists believe.
Climate dynamicist Chris Forest of the Massachusetts Institute of Technology and his colleagues built a model of the climate between 1860 to 1995, when greenhouse gas levels soared, that included three adjustable variables. They picked various combinations of values for these variables, then ran the model and compared the outcomes with observed temperature records. The most disturbing finding comes from the team's analysis of climate sensitivity, in other words, how much global temperatures will increase given a certain amount of atmospheric carbon dioxide. If atmospheric carbon dioxide doubles the least temperature will rise is 1.4 kelvin--comparable to the long-cited, subjective 1.5 K lower limit recently repeated by the Intergovernmental Panel on Climate Change (IPCC) (ScienceNOW, 22 January 2001). At that level, "future changes in climate are of considerable concern," notes climatologist Tom Wigley of the National Center for Atmospheric Research in Boulder, Colorado. But the team came up with an upper limit even higher than the IPCC's: a scorching 7.7 K, compared with the IPCC's 4.5 K.
Another result is the ability of aerosols--microscopic particles found in pollutant hazes--to change solar heating of the atmosphere. Forest and his colleagues found that aerosols have most likely cooled the planet, by reflecting 0.30 to 0.95 watt per square meter of solar energy back into space; IPCC suggested that number could be as high as 4. If correct, Forest's modest cooling would mean that most of early greenhouse warming is not being masked by aerosols. But aerosol modeler Joyce Penner of the University of Michigan, Ann Arbor, cautions that Forest probably shouldn't be lumping all types of aerosols together.
The third variable--the rate at which the ocean takes up heat and counteracts greenhouse warming--couldn't be usefully constrained and needs more research, they write.