A new understanding of ocean-heat transfer

Climatologists have had trouble squaring computer simulations of earlier climates with geological data. Skeptics have used this failing to cast doubt on computer projections of future global warming. Now it appears that, in two key aspects, the geological data have been misleading.

Computer simulations that tried to reproduce what happened during natural changes in the atmosphere's concentration of heat-trapping carbon dioxide (CO2) couldn't get the tropics right. They showed the tropics should have warmed when CO2 concentrations increased. However, the geological record seemed to show that the tropical sea temperatures didn't correlate with CO2 buildups, but stayed relatively cool.

Now Paul Pearson at the University of Bristol in England and 7 co-workers believe that previous analysts read the record wrong. They explain in today's issue of the journal Nature that what the geological record says today may not be what was originally recorded.

The ocean temperature records are written in the shells of microscopic animals called foraminifera. These animals use oxygen from sea water to build their shells. The oxygen comes in two forms - atomic weight 18 and weight 16. The ratio of the amount of oxygen 18 to oxygen 16 is a measure of the temperature of the sea water - a record presumably preserved in foraminifera shells in ancient sediments. Dr. Pearson and his colleagues suspected that chemical processes in many of the sediments that have been sampled could have changed that crucial ratio in the ancient shells.

To check this out, they studied shells that came from sediments that should have protected them from such changes. The shells come from two warm epochs - about 50 million and 70 million years ago, respectively. Sure enough, the oxygen "thermometers" in these shells show the tropics warmed as atmospheric CO2 increased in line with the computer simulations.

Commenting on this finding, Lee R. Kump at Pennsylvania State University in University Park says it "seems to solve a puzzle that has troubled people in the field for years."

The way heat energy flows from the warm tropics to the poles has been another such puzzle. Computer models show the atmosphere carrying most of the burden. Until now, observations have suggested the ocean plays a bigger role. Kevin Trenberth and Julie Caron at the National Center for Atmospheric Research in Boulder, Colo., find that satellite data from Feb. 1985 to April 1989 tell another story.

The atmosphere actually carries 78 percent of the heat transport in the Northern Hemisphere and 92 percent in the Southern Hemisphere at 35 degrees latitude, where the heat transport peaks. "This new analysis makes the observations more consistent with the most stable global climate models and gives us confidence that the models are on target," Dr. Trenberth says.

Until now, scientists had fudged climate models to make simulations conform to misleading real-world data. It's important to get this right. It would take a million 1,000-megawatt electric power stations - the largest power plants now used - to produce a quantity of energy equivalent to the heat the atmosphere carries each year from tropics to polar regions. Mishandling this massive energy flow would be like trying to simulate a steam engine without knowing how heat flows from the boiler and through the pistons.