But he emphasized that their findings do not provide a forecast; they only give an impression of the amount of freshwater required to alter or halt the North Atlantic ocean current.
"It is of course very unlikely that the freshening [freshwater inflow] will simply continue at the same rate it has for the past few decades," he wrote in an e-mail to National Geographic News.
Rahmstorf said the freshening could be part of a natural fluctuation in Earth's climate system that will stop and reverse.
He added, however, that if the phenomenon is due to global warming, which he said is likely, then the freshening will probably accelerate as glaciers melt and more rain falls at high latitudes in response to rising temperatures.
According to Curry, scientists are uncertain as to the exact course global warming will take and how it will affect the amount of freshwater flowing in the North Atlantic. A particular wild card, she noted, is Greenland.
"As it does melt, it will release freshwater into the Nordic seas"water bodies found between Iceland, Greenland, and Norway"and that probably represents the biggest source of freshwater that could have an impact on the conveyor belt," she said.
There are a number of mechanisms that could inject large amounts of freshwater into the Nordic seas at the precise region that is critical to the conveyor belt. They include
pooling and release of glacial meltwater,
collapse of an ice shelf followed by a surge in glacier movement, or
lubrication of a glacier's base through increased melting.
According to an unpublished survey by Potsdam University researchers Kirsten Zickfeld and Anders Levermann, expert scientific opinion varies widely on the likelihood that excess freshwater runoff from the Arctic will alter the North Atlantic conveyor belt in this century.
Some scientists consulted for the survey said there is no chance that the current will break down. Others estimated that the chance of a complete shutdown exceeds 50 percent if global warming climbs by 7.2° to 9° Fahrenheit (4° to 5° Celsius) by 2100.
Rahmstorf believes the chance of a circulation shutdown is as high as 30 percent. He said any possibility of such a scenario, even if slight, is cause for concern.
"Nobody would accept expanding nuclear power if there was a 5 percent risk of a major accident," he said. "Why would we accept expanding oil and coal power if there is a 5 percent risk of a major climate accident?"
SOURCES AND RELATED WEB SITES