Public Release: 

Core samples reveal evidence for a wet Ice-Age South America

Duke University

DURHAM, N.C. -- A 25,000-year precipitation record deduced from sediment core samples taken in lofty Lake Titicaca in Bolivia and Peru suggests that, contrary to widely accepted previous analyses, tropical South America may have been wet rather than dry during the last Ice Age and later cold periods, scientists from five universities reported Friday.

"We think we have a consistent story that's a fabulous story,"said lead author Paul Baker, a geology professor at Duke University's Nicholas School of the Environment and Earth Sciences in an interview. "Lake Titicaca is a beautiful rain gauge."

Situated on the Altiplano, a 12,000-foot plateau below the peaks of the Andes Mountains in Bolivia and Peru, Titicaca is South America's largest and deepest fresh water lake. Its location, across the high mountain range from one of Earth's largest water vapor sources -- the Amazon jungle -- makes it "an important place in terms of the present global climate picture," added Baker, who drafted the Science report.

The research, in the Bolivian section of the lake, was supported by the National Science Foundation. The findings were published in the Jan. 26, 2001 edition of the journal Science. The study's results suggest that the South American tropics were wet during cold eras and advancing ice in the Northern Hemisphere.

"We have a unique record of climate change in tropical South America that shows when global climate conditions cooled and the glaciers advanced, wetter climates prevailed in the Andes," said Geoffrey Seltzer, an associate professor of Earth sciences at Syracuse University, who is another of the report's lead authors.

"Today, most textbooks about paleoclimatology [the study of ancient climates] say the global tropics during the Last Glacial Maximum were arid," Baker said.

Using the Neecho, a refurbished 38-foot research boat whose ownership was transferred from the U.S. Geological Survey in Woods Hole, Mass. to Duke, the research team drew cores as long as 46 feet from the lake bottom at three different locations, where water depths were 121, 498 and 754 feet respectively.

While the Peruvian government has monitored lake levels since 1915, the researchers claimed success at extending those records back another 250 centuries by analyzing the core sediments for their magnetic values, fossilized diatoms, which are tiny silica-encased aquatic algae, calcium carbonate concentrations and oxygen isotope ratios.

While all these analyses contributed to the authors' conclusions, studies of fossil diatoms by co-author Sherilyn Fritz, an associate geology professor at the University of Nebraska at Lincoln, provided "what I believe is the single most important record," Baker said.

"The most unambiguous data we had came from the fossil diatoms in the sediment cores," agreed Fritz. "And it was (Nebraska) graduate student, Pedro Tapia, who did all that research as part of his work toward his dissertation. His research was the critical linchpin in the study."

Diatoms occur in different assemblages of species depending upon whether the water is deep or shallow and saltier or fresher. Lake Titicaca currently takes in more water than it loses through evaporation and discharge from its sole outlet, the Rio Desaguadero. But during consistently dry times lake levels drop too low for the river to discharge any of its water, and salt levels thus begin to rise.

Calcium carbonate levels also vary with lake depths and salinity, while magnetic values of the sediment change when upland erosion ceases -- another indicator of climatic fluctuations. And oxygen isotopes provide a way for the researchers to predict the sources and temperatures of water vapor.

Using all this evidence, the researchers found that Titicaca's environs were wet not only during the last glacial age, which began 25,000 years ago and lasted until 15,000 years ago, but also during later cold periods in the North Atlantic Ocean region. Conversely, warm spells in the North Atlantic were marked by drought on the Altiplano.

Using additional data from other paleoclimatic studies, the researchers found signs of two mechanisms driving wet conditions in the South American tropics. The first mechanism, operating on cycles of 20,000 years and more, triggered the last Ice Age when the extreme tilt of Earth's axis delivered less solar radiation to the far Northern Hemisphere.

The second, operating on cycles of about 1,000 years, is triggered by low seawater temperatures in both the high-latitude North Atlantic and the northern part of the tropical Atlantic. "When the water is cold in the northern tropics and warm in the southern tropics, those conditions increase the northeast trade winds that bring moisture from the Atlantic into the Amazon basin," Baker said.

"Today, most of the time when you have a really wet year on the Altiplano you also have a wet year in the Amazon basin; but not every time," added Baker, who acknowledged that the researchers were "stepping out on a limb" by equating wetness on Titicaca with conditions in the jungles to the east. "But most evidence points that way," he said.

Baker cautioned against forecasting future moisture levels in the tropics based upon past conditions, because of the implications of human-caused climate change.

"Human influence is so dominant now that whatever is going to go on in the tropics has much less to do with sea surface temperatures and Earth's orbital parameters and much more to do with deforestation, increasing atmospheric carbon dioxide and global warming," he said. Other authors of the study included Robert Dunbar of Stanford University, former Duke doctoral students Matthew Grove and Scott Cross, and James Broda, an engineer at the USGS in Woods Hole.

Broda, Baker said, "played a major role in helping me obtain and refurbish the Neecho, ship it to Bolivia, get it running and in taking cores."



Judy Holmes, Syracuse
(315) 443-5172

Tom Simons, Nebraska
(402) 472-8514

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.