Global warming is usually presumed to be a modern problem: a byproduct of the Industrial Revolution and its intensive use of carbon dioxide-producing fossil fuels. But humans have been changing the earth's climate for thousands of years, says William Ruddiman, an emeritus professor of environmental sciences from the University of Virginia. And the changes have been substantial.
Early warming
Even before the start of the industrial era, Ruddiman says, our ancestors may have warmed the globe sufficiently to stave off a new ice age.
Ice ages are caused by variations in the Earth's orbit that alter the amount of sunlight reaching Canada, Siberia and Alaska during the brief arctic summer. During high-sunlight cycles, there's enough warmth to melt the previous winter's snows. During cold cycles, there isn't, and the snow gradually accumulates into glaciers.
These orbital variations occur in well-understood, long-term cycles. But the climate is also affected by cyclical changes in the atmospheric levels of two important greenhouse gases, carbon dioxide and methane, which trap solar heat that would otherwise radiate back into space.
Methane is produced when vegetation decays beneath swamps and marshes. Scientists can use ice cores from Greenland and Antarctica to measure the amount of methane in the atmosphere during the past 400,000 years. These studies indicate that during eras when the Northern Hemisphere receives weaker summer sunlight, swamps and marshes shrink. In eras when solar energy is higher, rainfall increases and marshes expand.
Based on the pattern of prior cycles, methane levels should have reached a peak 11,000 years ago and been dropping ever since. But the Greenlandic ice cores show that 5,000 years ago, something went awry, and methane levels began to rise.
"You have to throw 395,000 years of history out the window to come up with a natural explanation for this," Ruddiman said in San Francisco last December, at a meeting of the American Geophysical Union.
His not-so-natural alternative?
Five thousand years ago was just about the time people started creating artificial marshes to grow rice in Southeast Asia.
By 2,000 years ago, rice farmers had already used up the flat land of the valleys and were beginning to build the hillside terraces we see today, releasing more and more methane as more paddies were built.
At the same time, something was increasing the atmospheric content of carbon dioxide. The Greenlandic and Antarctic ice cores show that carbon dioxide levels fluctuate on natural cycles of 22,000, 41,000 and 100,000 years. According to these patterns, carbon dioxide, like methane, should have reached a peak 11,000 years ago and dropped ever since.
But it, too, dropped for only the first half of that cycle, then started to rebound -- so much so that at the start of the industrial age, the level was already 15 percent "too" high.
Previous scientists have posited various natural theories for the reversal of carbon dioxide levels. One theory is that changes in ocean chemistry are causing the seas to disgorge large quantities of previously absorbed carbon dioxide. Another is that it is due to a natural decline in forests, which remove large quantities of carbon dioxide from the air to form branches, leaves, bark and roots. When forests die, all of that stored carbon dioxide is returned to the air.
But Ruddiman suggests that humans might be the cause. Studies of pollen particles trapped in lake-bed sediments allow scientists to trace the spread of wheat, peas, lentils, flax and barley across regions that were naturally forest. As far back as 10,000 years ago, he says, people were beginning to cut down forests to make room for farming.
These facts may have been overlooked by climate modelers, but they are well-known to historical geographers. In 1989, Ian G. Simmons of the University of Durham, England, wrote that by 2,000 years ago, large segments of Southeast Asia, China, Southwestern Asia and the Mediterranean region were "greatly" deforested.
And in a 2003 book, "Deforesting the Earth," Oxford geography professor Michael Williams reported that humans were already cutting down European forests 6,000 years ago, and that American Indians were clearing forests in the Mississippi River Valley as far back as 7,000 years ago to plant squash, sunflowers, maize and beans.
"Most of Eurasia was deforested by the time of Christ," Ruddiman says.
Counting trees
In an effort to quantify the amount of preindustrial deforestation, Ruddiman turned to the Domesday Book, a census of Britain conducted by William the Conqueror in 1086 A.D. In addition to counting people, William's census takers tallied the extent of forests, fields and pastures. According to figures in the Domesday Book, the 1.5 million people then living in England had already cut down 85 percent of their nation's trees.
Extrapolating these per-capita land-clearing figures to the 57 million people living in China a thousand years earlier, plus the millions more in India, Southeast Asia and the Roman Empire, Ruddiman calculates that 2,000 years ago, deforestation was already quite extensive.
Already, he says, tree cutting had released 700 to 900 billion tons of carbon dioxide to the air -- enough to offset the natural decline and start driving levels of the gas back up again, thousands of years before anyone was using significant quantities of coal or oil.
All of that carbon dioxide, Ruddiman adds, would warm the Earth by about 1.4 degrees -- roughly the same amount that industrial-era emissions are believed to have warmed it to date (but not by as much as today's emissions are expected to warm it in the future).
How plague played in
The effect, he says, is stronger at high latitudes -- strong enough that climate models show that if people hadn't cut down so many trees, ice sheets might again be forming in parts of Canada such as Labrador and Baffin Island.
Ruddiman backs up his tree-cutting theory by pointing to several dips in the atmospheric level of carbon dioxide that occurred over the past 2,000 years. None was large -- only a few parts per million -- but they appear to be too large to be explained by natural factors such as volcanic emissions.
One of these dips occurred during the height of the Roman Empire. Another occurred in the 1400s, and a third occurred between 1500 and 1750 A.D. All of these, he says, link to periods when plagues killed off sizable fractions of the world's population.
The first occurred at a time when bubonic plague killed 20 million people in China and the Roman Empire. The second correlates to the Black Death, which killed one-third of the people of Europe in its first year alone. The third was during an era when 90 percent of the 50 million to 120 million people living in Central and South America died of smallpox, measles and other European diseases, the single largest mass mortality in history.
When that many people die, farms are abandoned and trees grow back quickly enough to take significant amounts of carbon dioxide back out of the air.
Historical accounts of the Black Death, Ruddiman says, are full of stories about millions of abandoned farms. "These accounts don't give numbers of farms or acreage," he says, "but it's immense."
Ralph Keeling, a professor of geochemistry at UCSD's Scripps Institution of Oceanography in La Jolla, says that confirmation of Ruddiman's theory will await the drilling of ice cores going back more than 400,000 years.
That's because astronomical factors make our current interglacial period more akin to one that occurred slightly too early to show up on today's oldest cores. If that earlier warm spell shows the same methane and carbon dioxide anomalies we see in the past 8,000 years, then the cause is natural, and Ruddiman's theory goes down the drain.
"He's making an appealing argument that is at least plausible," Keeling says, "but it could turn out to be wrong as we get better evidence."
Ruddiman admits that his thesis is controversial, referring to it as "an outrageous challenge to the conventional wisdom." But he also believes he's onto a previously overlooked link between history and climate modeling.
"The most in-your-face statement I can make is that humans stopped a glaciation," he says. "And I think there's a strong case that can be made for that."
Keeling doubts that the effect was strong enough to have staved off glaciers, but he agrees that Ruddiman is asking good questions. "At some level," he says, "it seems inevitable that early agriculture would have had an impact on the atmosphere. The question is simply how big that impact was."