In 1801, the eminent British astronomer reported that when sunspots dotted the sun's surface, grain prices fell. When sunspots waned, prices rose.
He suggested that shifts in grain prices were a stand-in for shifts in climate. Large numbers of sunspots led to a warmer sun, he reasoned. With more warmth reaching Earth, crop yields would increase, depressing grain prices.
With that, a 200-year hunt began for links between shifts in the sun's output and changes in climate.
No one doubts that the sun drives Earth's climate. Nor do researchers doubt that over long time spans, changes in the level of sunlight reaching Earth's surface leave their imprints on climate.
The vast bulk of research to date, however, points to greenhouse gases - mainly carbon dioxide from burning coal, oil, and natural gas - as the main force behind the current warming trend, most climate scientists say.
Still, over the past decade some researchers say they've found puzzling correlations between changes in the sun's output and weather and climate patterns on Earth. These links appear to rise above the level of misinterpreted data or faulty equipment.
'There are some empirical bits of evidence that show interesting relationships we don't fully understand,' says Drew Shindell, a researcher at NASA's Goddard Institute for Space Studies in New York.
For example, he cites a 2001 study in which scientists looked at cloud cover over the United States from 1900 to 1987 and found that average cloud cover increased and decreased in step with the sun's 11-year sunspot cycle. The most plausible cause, they said: changes in the ultraviolet (UV) light the sun delivers to the stratosphere.
Clouds can cool, or clouds can heat
Others claim to have linked shifts in levels of cosmic rays reaching deep into the atmosphere to changes in average cloud cover. Depending on how thick and how high they are, clouds either cool the planet by reflecting sunlight back into space or act as a blanket and trap heat. The valve controlling the flow of cosmic rays from deep space is the sun's magnetic field - which shifts with sunspot activity.
But this broad line of inquiry faces an enormous credibility problem, Dr. Shindell notes. From Herschel's day through the early 20th century, scientists have offered correlations that 'fall apart the longer you look at them,' he says.
Moreover, when scientists report a new correlation, some enthusiastic advocates go beyond what the data show and imbue it with too much significance. Such is the case with cosmic rays, many scientists say, whose poorly demonstrated ties to cloud formation have nevertheless been touted in the public arena - if not the scientific arena - as an explanation for most of the warming in the 20th century.
To say that current warming trends are 'all cosmic rays and no carbon dioxide is totally ludicrous, in the same way that people say that it's all [human-induced] carbon dioxide and nothing natural. That is equally ludicrous,' says Jasper Kirkby, a physicist who is actively exploring potential links between cosmic rays and clouds at CERN, Europe's center for high-energy physics research in Geneva.
'Climate is a cocktail,' he explains. 'The effect of cosmic rays on clouds - if there is a significant effect - will be part of the mix. The question is: Is it a significant part of the mix, or insignificant?'
Mainstream scientific skepticism about a strong direct link between changes in the sun's output and today's global warming stems from a tiny shift in sunlight.
Generally, peak periods of high sunspot activity deliver more sunlight to the top of the atmosphere than periods of minimum activity. Scientists measure this 'total solar irradiance,' which includes infrared and ultraviolet light as well as visible light.
In 1970, Russian researchers using high-altitude balloons to measure sunlight reported a 2 percent rise in the sun's output as the sun moved from periods of little sunspot activity to peak activity. Today, using better measurements from satellites over the past 28 years, the change in total solar irradiance is estimated to be much smaller, between 0.05 percent and 0.07 percent. The most important component for climate-change purposes - visible light - represents about half of this change, says Tom Woods, a researcher at the University of Colorado's Laboratory for Space and Atmospheric Physics, based in Boulder.
'Pesky' correlations with sunspots
Last fall, solar physicists and climate scientists in the US and Europe reviewed the latest studies of changes in total solar irradiance driven by the 11-year sunspot cycle. They concluded that those changes are unlikely to have had a 'significant influence' on global warming since the 1600s. In particular, satellite measurements since the late 1970s showed changes too weak to have 'contributed appreciably to accelerated warming over the past 30 years.'
The effect 'is really small, unless you can come up with ways to amplify it,' says Tom Wigley, a senior scientist at the National Center for Atmospheric Research in Boulder, who took part in the study.
Other studies suggest that changes in sunlight - as well as the cooling effect of volcanic activity, which sends sunlight-reflecting particles high in the sky - probably played a major role in climate during preindustrial times and even into the early 20th century. But even these find that CO2 emissions have dominated the scene over the past half century.
Some pesky correlations - such as the one between sunspot cycles and cloud cover - linger. This has led some scientists to ask if some process in the atmosphere may be boosting those tiny changes.
One candidate is UV light. During swings in sunspot cycles, the largest fractional changes in the sun's output occur in the ultraviolet range, Shindell notes. But much of that is absorbed by ozone in the stratosphere - which may be the connection, he suggests. The rise and fall of UV light can alter the amount of heat-trapping ozone in the stratosphere, changing its circulation patterns. These changes can work their way into the layer below, the troposphere, where weather and people meet. Instead of warming the troposphere, changes in solar UV output appear to redistribute warmth, chill, rainfall, and other conditions already present.
This mechanism may account for plunging winter temperatures in the Little Ice Age (1450 to 1850) - at least over land in the Northern Hemisphere, he says.
Another possibility: cosmic rays
But if changes in ultraviolet light tied to sunspot cycles merely stir the climate pot, might something else affect long-term global average temperatures?
Enter galactic cosmic rays. In 1997, Danish researcher Henrik Svensmark and a colleague at the Danish Meteorological Institute injected new life into this debate with the first in a set of papers that suggested a strong correlation between an increase in galactic cosmic rays reaching Earth's surface during low points in the sunspot cycle and increased cloud cover.
The idea of a big effect on climate from cosmic rays is controversial. For instance, the team that studied sunspots and cloud cover over North America found that average cloudiness rose and fell with the sunspot cycle, but didn't track with cosmic ray trends.
Still, a study published last year in Britain showed a small but statistically significant effect from cosmic rays, notes Rasmus Benestad, who specialized in solar-climate interactions at the Norwegian Meteorological Institute in Oslo. He is highly skeptical that cosmic rays play a big role in climate, he says. But, he adds, the phenomenon is worth exploring.
Dr. Kirkby and colleagues at several institutions aim to do just that. They've designed an aerosol chamber to test how cosmic rays might affect cloud formation and how significant the effect might be. 'You really can't settle the issue by more heated debate,' he says. 'You need experimental data.'