Known variations in the sun's total energy output cannot explain recent global warming, say researchers who have reviewed the existing evidence. The judgment, which appears in the September 14 Nature, casts doubt on the claims of some global warming skeptics who have argued that long-term changes in solar output, or luminosity, might be driving the current climate pattern.

The evidence for human-induced global warming is neatly captured in a plot of the planet's reconstructed temperature over the last 1,000 years. The temperature takes a dramatic upswing starting 100 years ago, creating the so-called hockey stick graph. A reasonable question is whether natural changes such as solar activity could have caused or contributed to the upturned blade of that stick, perhaps because the sun's luminosity varies widely over centuries or more. "The question is, were there times in the past when it was equally warm, and the answer is no," says Tom Wigley of the National Center for Atmospheric Research. He and three colleagues compared the average of a number of temperature reconstructions based on tree rings, ice cores and other data with models of Northern Hemisphere temperature that include different levels of solar variation, from little to a speculatively high amount. In all cases, "what you get out looks very much like the observations" from real samples, he says. "The warming [of the past 100 years] is greater than any in the last 1,000 years."

The consistency meshes with solar physicists' latest understanding of how the sun works, the group notes. The sun's luminosity swings up and down by less than 0.1 percent in accord with an 11-year cycle of sunspots and faculae, bright areas of heightened output [see image above]. This cycle accounts for most of the sun's variability. Recent simulations reinforce the idea that convection inside the sun rapidly smoothes out internal hot spots before their concentrated heat can escape like an upwelling of magma, the researchers note. This inertia allows surface changes to have a discernible effect and explains why no additional sources of variation have been identified so far, they say.

Skeptics might hope to take refuge in some lingering uncertainties. Variability in solar ultraviolet or magnetized plasma output are poorly understood and could affect the climate in ways that luminosity doesn't. Some researchers have also claimed to see signs of longer-term or hyperactive solar luminosity in the last 10,000 years, although "if they are playing some role, these possible longer-term solar effects are small compared to the unprecedented human-induced changes," says geologist Feng Sheng Hu of the University of Illinois, who has reported evidence for one type of longer cycle.