December 5, 2005

Is there a long-term trend in the thermohaline circulation?

Filed under: Climate History

In the December 1st issue of Nature magazine, Harry Bryden and colleagues at Britain’s National Oceanography Centre report that the Atlantic meridional circulation (also known as the thermohaline circulation (THC))—the density driven current that carries warm surface water northward and returns colder deep water southward—has slowed by 30 percent between 1957 and 2004.

The significance of this finding is difficult to assess in light of other recent observations.

Climate model simulations estimate that a complete shutdown of the THC would result in a cooling of Europe of 4ºC or more. So, shouldn’t a 30% slowdown have some noticeable impacts, i.e. a pretty sharp cooling trend?

Just two days before the Bryden results were published, a report from the European Environment Agency detailed all of the ills that Europe has been facing recently because of how warm it has been, and prominently proclaimed that Europe’s four hottest years on record were 1998, 2002, 2003 and 2004!

In the nuclear bombing business, this is “fratricide,” where two closely placed devices—in this case science bombshells timed to influence the ongoing yet-another-UN-Kyoto confab in Montreal—blow each other up.

Even NASA’s Gavin Schmidt told the New York Times that the decline in the THC as reported by Bryden et al. should have produced a noticeable decline in surface temperatures, but that such a decrease had not yet been found.

How many breathless news stories, like the one in London’s politically quaint Guardian, played Bryden as a long-term (read: anthropogenic influenced) trend in the THC? A close read, however, shows that the THC changes have really only taken place sometime since 1992. Since 1957, the characteristics of the Atlantic Ocean that Bryden et al. used in making their calculations were sampled 4 more times—in 1981, 1992, 1998, and 2004. No remarkable changes were detected between 1957 and 1992, but since then, Bryden found indications that the THC had slowed a bit by 1998, and further in the 2004 data.

A large-scale, arguably “natural” event took place during that same time. Karcher et al. recently reported a large freshwater release from the Arctic Ocean into the North Atlantic Ocean in the mid-1990s as a result of atmospheric circulation patterns resulting from an extremely intense North Atlantic Oscillation (or NAO, a measure of the pressure distribution over the North Atlantic Ocean basin). Since the early to mid-1990s, the NAO has returned to more normal values indicating that the trend from low NAO values characteristic of the 1960s to the historically high values in the early 1990s has ended. The return to normal NAO conditions has also, according to Karcher et al., resulted in a more normal degree of freshwater input to the subpolar North Atlantic. Addition of a pulse of freshwater to the North Atlantic is one mechanism for slowing the THC, and so it is possible that the freshwater release in the mid-1990s identified by Karcher could have acted to slow, temporarily, the THC—perhaps an effect picked up in Bryden’s analysis.

And further, a recent paper by Knight et al. reconstructed the history of the THC in the Atlantic for the past 125 years or so based upon a combination of climate model simulations and sea surface temperature observations. They concluded that the THC had increased substantially since the 1970s—a finding in opposition to that of Bryden et al. Squaring these disparate findings is not a simple matter and indicates that the situation is much more complex than perhaps realized.

Science magazine’s Richard Kerr covers the Bryden et al. findings in the December 2nd issue of the magazine with an article titled “The Atlantic Conveyor May Have Slowed, But Don’t Panic Yet.” Kerr’s look at the issue is a bit more critical than most other mainstream press reports. He notes that the trend reported by Bryden et al. is hardly bigger than the uncertainty in the calculations, and even quoted Bryden as telling him “we don’t know enough about the ocean to know whether [our result] represents a trend” that will persist.

Here is how Kerr concludes his look into the issue:

The picture is still fuzzy, however. “It would be dangerous to jump to the conclusion that there’s a persistent weakening” of the conveyor circulation, says ocean and climate modeler Richard Wood of the Hadley Centre for Climate Prediction and Research in Exeter, U.K. Wood, Rhines, and Bryden all worry that the near-instantaneous snapshots taken by the ocean surveys might have been misleading. Like any part of the complex climate system, the conveyor is bound to slow down at times and speed up at others. The two latest surveys, Wood says, may have happened to catch the Atlantic as the conveyor slowed temporarily, giving the impression that a permanent change had taken place.

On the other hand, the [Bryden’s] analysis may not have even captured what happened in the past decade or so. Climate models simulating the conveyor in a warming world don’t call for such a large slowdown until sometime in the next century, Wood notes. In fact, climate researcher Jeff Knight of the Hadley Centre and colleagues recently reported that changing sea surface temperatures suggest that the conveyor has speeded up a bit since the 1970s (Science, 1 July, p. 41). And physical oceanographers Carl Wunsch and Patrick Heimbach of the Massachusetts Institute of Technology have just crunched far more oceanographic data from a variety of sources over the interval of dramatic change (1993 to 2004) in the NOC analysis. In a paper submitted for publication, they report a small slowdown, a quarter the size of the NOC group’s. The change in heat transported northward is negligible, they calculate.

So has the conveyor slowed? Might it continue to slow? “We don’t know,” says Wunsch. And it may take a decade or two more of watching and waiting to know for sure.

(For another critical (and more technical) look at the Bryden et al. results, see Comment #25 from physical oceanographer Dr. Martin Visbeck at

Together, all of this points to a far less clear picture about the state of the circulation of the Atlantic Ocean than is generally being reported. On the other hand, if Bryden et al. have discovered a real long-term change in the THC, then this will in turn change the paradigm as to how the THC relates to a huge host of climate parameters—parameters that, at present, don’t seem to be behaving like they should if the THC is indeed slowing dramatically. Not often does one anomaly break a paradigm. It happens—but rarely.


Bryden, H.L., et al., 2005. Slowing of the Atlantic meridional overturning circulation at 25ºN. Nature, 438, 655-657.

Karcher, M., et al., 2005. Arctic Ocean change heralds North Atlantic freshening. Geophysical Research Letters, 32, doi:1029/2005GL023861.

Kerr, R., 2005. The Atlantic Conveyor May Have Slowed, But Don’t Panic Yet. Science, 310, 1403-1404.

Knight, J.R., et al., 2005. A signature of persistent natural thermohaline circulation cycles in observed climate. Geophysical Research Letters, 32, doi:10.1029/2005GL024233.

No Comments

No comments yet.

RSS feed for comments on this post. TrackBack URI

Sorry, the comment form is closed at this time.

Powered by WordPress