Human-induced greenhouse warming pumps heat into oceans, two Science studies report

Contact: Ginger Pinholster
gpinhols@aaas.org
202-326-6421
American Association for the Advancement of Science

Greenhouse gas emissions have caused the world’s oceans to heat up significantly over the last 50 years, according to two studies in the 13 April issue of the international journal, Science.

Researchers had previously reported in Science that the Atlantic, Pacific, and Indian oceans have collectively warmed an average of 0.06 degrees C since 1955.

Now, two new modeling studies have tied the ocean heating directly with global warming caused by human activity.

“I believe our results represent the strongest evidence to date that the Earth’s climate system is responding to human-induced forcing,” said Sydney Levitus of the National Oceanographic Data Center/NOAA, lead author of one of the new studies and the previous report.

“The parallel climate model, a cooperative product of the U.S. Department of Energy and the National Science Foundation, can reproduce the warming in the ocean seen over the last 50 years. This will make it much harder for naysayers to dismiss predictions from climate models,” said Tim Barnett of Scripps Institution of Oceanography, lead author of the second Science study.

Covering 72 percent of the Earth’s surface, the oceans are often called the “memory” of the earth’s climate system. They can absorb large amounts of heat and sequester it at depth for up to thousands of years before circulating it back to the atmosphere.

“Warming in the oceans is bad news and good news,” said Barnett. “It really does add strength to the claims that global warming is here. But, it also suggests that the immediate impact may not be as great, because the oceans may slow things down a little.”

Measurements of past air temperatures are more numerous and reach farther back in time than those of ocean temperatures. Earlier climate models didn’t include an ocean component, and therefore frequently predicted that modern air temperatures would increase more than they actually have, according to Levitus. This discrepancy has been useful fodder for skeptics, who have argued that global warming wouldn’t be as severe as commonly thought.

Last year, Levitus and his colleagues determined an average for how much the oceans had warmed by compiling millions of deep ocean temperature measurements from 1948-1995. (Science, 24 March 2000, p. 2225.) But, it wasn’t clear whether this heat came from greenhouse warming or just a natural swing in the climate cycle.

To investigate, Barnett and Levitus each used a different climate model to simulate how ocean temperature should respond to current levels of greenhouse gases and other modern day atmospheric conditions. Both models predicted an amount of warming quite similar to that which scientists have measured.

To be sure that their model results weren’t just a fluke, Barnett and his colleagues ran five simulations and averaged their results together. They also ran the model without the extra greenhouse gases and sulfate aerosols produced by human activity. Without the anthropogenic “fingerprint,” the simulated ocean did not warm significantly.

“What we found is that the signal is so bold and big, that you don’t have to do any fancy statistics to beat it out of the data; it’s just there, bang. The odds are exceedingly good that the model did not trick us with this signal; it couldn’t have done it by itself,” Barnett said.

Using a different model, Levitus and his colleagues also factored in the effects of the sun’s changing intensity and aerosol particles produced by volcanic eruptions over the last century. This simulation produced a very close match to the actual measurements.

“The fact that the model ocean warms by approximately as much as our observed estimate indicates that the models are very robust in simulating the observed changes in the Earth’s climate system during the past 100 years,” Levitus said.

More work must be done to make computer models deliver more consistent and specific predictions. Barnett says that an important benefit of the two Science studies, is that, “in the future, models will have to get the ocean right in order to be believable. These results raise the bar for sorting out the best models.”

The study by Levitus et al. was funded by NOAA, with smaller grants from the U.S. Department of Energy and NASA.
The study by Barnett et al. was funded by NOAA, the U.S. Department of Energy, the National Science Foundation, and the Scripps Institution.

A related News story by Richard Kerr will be available Wednesday, 11 April.

To receive copies of the Science papers or the related News story, call 202-326-6440, or send e-mail to scipak@aaas.org.