A recent Associated Press article suggested that humans have been changing the global climate since thousands of years before the industrial revolution. 8,000 years ago atmospheric levels of carbon dioxide began to rise as humans began clearing forests, planting crops and raising livestock, according to the article. Methane levels started increasing 3,000 years later.

The scientist quoted in the article was Bill Ruddiman, emeritus professor at the University of Virginia. Ruddiman suggested that the human-caused disruption would have been sufficient to create much warmer climate during ancient times than "natural" conditions would have suggested.

Right now there is a raging debate occurring regarding the temperatures of the last several thousand years. The longstanding view, championed by H.H. Lamb and other climate historians, was that the Holocene (the pleasant interglacial period in which civilization has flourished since the last Ice Age 15,000 years ago) has been marked by several notable climate fluctuations:

1. a very warm period about 3,000 B.C., the warmest of the Holocene (the "Holocene Maximum";

2. a gradual cooling thereafter, reaching a low point around the time of Christ;

3. a warming for about 1,000 years, peaking in roughly 1,000 A.D. ("Medieval Warm Period")

4. a cooling for several hundred years, with the coldest period from about 1560-1830 ("Little Ice Age")

5. brief warm and cool periods from 1830-1870 and 1870-1910, respectively

6. a warmer 20^th century, for the most part.

On shorter scales (decades or less) there have been numerous rises and falls in temperatures, some of them quite significant, but the long-term changes are those listed above.

Lamb and other historians using various recorded and anecdotal information concluded that the Medieval Warm Period temperatures were comparable to those of modern times, the Holocene Maximum was much warmer, and the little Ice Age was much cooler.

Enter Michael Mann of the University of Virginia, lead author of a paper that presented a much different viewpoint. Using tree ring and other data for estimating temperature histories back to 1,000 AD, Mann created a new and very different history. The Medieval Warm Period no longer existed, scarcely distinguishable from the Little Ice Age. Suddenly our current temperatures had become "the highest in 1,000 years." This was proof, so it was claimed, that human-induced global warming was occurring!

One of the most valuable and heavily-referenced "paleoclimate data sets" (data used to infer long-term climate change) is the Vostok ice core data. In January 1998, a collaborative ice-drilling project between Russia, the United States, and France at the Russian Vostok station in East Antarctica yielded the deepest ice core ever recovered, reaching a depth of 3,623 m (nearly 12,000 feet). The ice was deposited in layer upon layer, like dirt where the Grand Canyon intersects, each representing a year. The lowest layers were deposited about 400,000 years ago.

Ice cores are valuable because they contain tiny gas bubbles whose composition can be measured. CO2 is measured directly using a gas chromatograph, while temperature is estimated from concentrations of two gases, deuterium and Oxygen-18.

Early Vostok data analysis looked at samples centuries apart, and concluded (correctly) that there is a very strong relationship between temperatures and CO2 concentrations. The conclusion for many was obvious: when CO2 goes up, temperatures go up, and vice-versa. This became the basis for a number of scary-looking graphs in books by the scientist Stephen Schneider, former Vice President Al Gore, and others, predicting a much warmer future (since most scientists agree that CO2 will continue to go up for some time).

Well, it's not as simple as that. When the Vostok data were analyzed for much shorter time periods (decades at a time rather than centuries), something different emerged. H. Fischer and coauthors reported in Science (283: 1712-1714, 1999) that "the time lag of the rise in CO2 concentrations with respect to temperature change is on the order of 400 to 1000 years." In other words, CO2 changes are caused by temperature changes! Many other recent studies have shown similar results. Studies by Indermuhle et al (2000), Monnin et al (2001), and Mudelsee et al (2001) indicated a lag of 800-1500 years between temperature and CO2. References are available on request.

Ruddiman's study is interesting, and bears further analysis. But two counter-arguments stand out: it is unlikely that the rather low human populations of ancient times would have had the means to produce such high CO2 levels, aside from massive forest fires; and the high "spikes" in CO2 were more likely responses to the abrupt warm periods which are known to have existed. Warm periods would have triggered increases in plant life, which eventually would have died or been burned and released to the atmosphere -- as CO2. Warmer ocean temperatures would have released CO2 to the atmosphere (more CO2 is absorbed when water is cooler).

And why would temperature have risen and fallen if CO2 were not to blame? It's anyone's guess -- but my guess is that changes in earth's orbit, solar activity and ocean circulation were chiefly responsible for the warm and cool periods.