How Confidently Can We Reconstruct the Past 2,000 Years?
Average global surface temperatures rose by about 1 degree Fahrenheit (0.6 degrees Celsius) over the last century, according to data from thermometers stationed around the world. How does this increase, especially the rapid warming in recent decades, compare with temperature swings over the last one or two thousand years?
It's relatively easy to make such comparisons as far back as the middle of the 19th century when meteorologists first started using instruments to collect geographically widespread temperature data. But what climate scientists really want to know is how modern temperature fluctuations compare with natural variations that took place before the Industrial Revolution, when levels of heat-trapping greenhouse gases in the atmosphere were much lower. With limited or no instrumental temperature records, researchers rely on proxy evidence such as tree rings, boreholes, ice cores, corals, and ocean and lake sediments to reconstruct past temperatures. Even the length of glaciers documented in historic paintings is telling.
Like a lot of climate change research, however, using proxy evidence to reconstruct historical temperatures has not been without public controversy, much of it centered on climatologist Michael Mann. He and his colleagues published a thousand-year reconstruction in 1999 that concluded the Northern Hemisphere warming of the late 20th century was unprecedented and that the 1990s was the warmest decade, and 1998 the warmest year. When debate arose in Congress last year over Mann's research, it turned to the National Research Council to weigh in on the matter, and a committee was convened to assess the ability of scientists to reconstruct surface temperatures for the past 2,000 years.
In its report, the committee's first observation was that the warming recorded by instruments in the last century is also reflected in temperature reconstructions derived from borehole measurements, retreating glaciers, and other proxies. The report adds that there are sufficient proxy data from enough places to say with high confidence that the mean surface temperature globally was higher in the last few decades of the 20th century than during any other comparable period in the past 400 years.
The evidence is shakier for the period between A.D. 900 and 1600, however, a time frame for which there is less proxy data and the uncertainties associated with them are larger. Uncertainties are greater surrounding decades and single years especially, because not all proxies record temperatures for such short timescales. So while the committee found Mann's conclusion that the warming during the final decades of the 20th century is unprecedented in the past thousand years to be plausible, it placed much less confidence in his claims about the 1990s and 1998 in particular.
For periods before 1600, there are fewer proxies in fewer locations to provide temperatures. Nevertheless, the committee noted that proxy data does indicate that many locations were warmer in the last 25 years than during any 25-year period since 900.
In addition, the report finds that surface temperature reconstructions for the last millennium are generally consistent. They show relatively warm conditions around A.D. 1000 and a "Little Ice Age" from roughly 1500 to 1850. The exact timing of the medieval warming is unclear and it may have varied from region to region. None of the reconstructions indicates higher temperatures during the Middle Ages than during the most recent decades, the report notes.
Very little confidence can be put in statements about average global surface temperatures earlier than A.D. 900, the committee added. Proxy evidence that is scarce prior to 1600 -- especially in the Southern Hemisphere -- gets even scarcer before 900.
-- Bill Kearney
Surface Temperature Reconstructions for the Last 2,000 Years. Committee on Surface Temperature Reconstructions, Board on Atmospheric Sciences and Climate, Division on Earth and Life Studies (2006, approx. 196 pp.; ISBN 0-309-10225-1; available from the National Academies Press, tel. 1-800-624-6242; $40.00 plus $4.50 shipping for single copies).
The committee was chaired by Gerald R. North, distinguished professor of meteorology and oceanography and Harold J. Haynes Endowed Chair in Geosciences, Texas A&M University, College Station. The study was funded by the National Academies.