The two places that are believed to be the major locations of oceanic deep water formation that are thought by many to drive the global thermohaline circulation are the North Atlantic and Southern Oceans, each of which is surmised to have supplied equivalent amounts of new deep water to the system for most of the past 800 years.  Periodically, however, it is believed that the strength of each of these deep water sources waxes and wanes; and Broecker et al. (1999) have suggested that these oscillations in deep water formation may be responsible for millennial-scale climate oscillations, such as the recent Little Ice Age and prior Medieval Warm Period.

In a study of grain sizes of deep-sea sediments obtained from the northeast Atlantic Ocean, Bianchi and McCave (1999) discovered a similar millennial-scale oscillation of the thermohaline circulation that extended back in time over the past 11,000 years and seemed, at the younger end of the record, to be related to these two historically-well-known climatic regimes.  Analyzing cadmium/calcium ratios of benthic foraminifera shells in sediment cores retrieved from the Bahama Banks region of the Northwest Providence Channel that connects the North Atlantic Basin (Sargasso Sea) to the Florida Straits, Marchitto et al. (1998) found evidence of similar oscillations in thermohaline circulation over the same 11,000-year period.

In another study, Ruhlemann et al. (1999) derived a 29,000-year history of sea surface temperature in the western tropical North Atlantic Ocean from measurements of alkenone unsaturation in sedimentary organic matter obtained southeast of Grenada.  At the times when two dramatic cooling events were occurring in the northern North Atlantic Ocean (16,900 to 15,400 years ago and 12,900 to 11,600 years ago), they found significant warming at their site.  Both of these observations, they noted, were consistent with the contemporaneous slow-downs of North Atlantic Deep Water formation that are known to have occurred at those times: with less heat being transported northward, temperatures rise in the south.

Finally, in a study of physical and chemical characteristics of a sediment core obtained south of Iceland, Raymo et al. (1998) determined that millennial-scale oscillations of this type were occurring well over a million years ago, leading them to conclude that associated alternating warm and cold periods of this time-scale "may be a pervasive and long-term characteristic of Earth's climate, rather than just a feature of the strong glacial-interglacial cycles of the past 800,000 years."

In view of this strong and persistent coupling of climate and thermohaline circulation strength - which appears to be unrelated to either past changes in atmospheric CO2 concentration or long-term global mean air temperature, in that paired oscillations of climate and oceanic conveyor belt strength continue unimpeded through glacials and interglacials alike - there is good reason to believe that the modest global warming that has followed on the heels of the apparent demise of the Little Ice Age may be totally unrelated to the concomitant increase in the air's CO2 content that is generally attributed to anthropogenic CO2 emissions.  Nevertheless, a number of scientists continue to believe that the ongoing rise in the air's CO2 content could well perturb this finely-tuned system and upset earth's climatic equilibrium.

The general thinking of this school of thought is that further warming would intensify the planet's hydrologic cycle, resulting in increased precipitation with a consequent increase in continental freshwater runoff into the North Atlantic Ocean that would lessen the density of surface seawater there and reduce its rate of sinking.  This phenomenon is then postulated to reduce the driving force for the great circulatory system of the world's oceans that brings warmth to Europe via the Gulf Stream, the result of which would be a cooler Europe and a reduced rate of global warming (see our Editorial of 15 July 1999 and our Journal Review A Natural Thermostat to Keep a Lid on Global Warming).  And to the credit of those who lean toward this scenario, a smaller-scale version of a portion of this phenomenon has actually been observed over the past several decades in the history of deep water outflow from the Nordic seas to the Atlantic Ocean (Bacon, 1998).

Additional support for this scenario was provided by Barber et al. (1999), who pulled evidence from several different sources together to make a strong case for the proposition that the dramatic regional cooling of 1.5-3°C that is known to have occurred at marine and terrestrial sites around the northeastern North Atlantic Ocean some 8200 years ago was caused by a catastrophic release of freshwater into the Labrador Sea from the final outburst drainage of glacial lakes Agassiz and Ojibway.  They envisioned the sudden increase in freshwater flux to have reduced the formation rates of Labrador Sea Intermediate Water and North Atlantic Deep Water sufficiently to have strongly reduced the strength of the oceanic conveyor belt, retarding the northward transport of heat to this region of the world.

Although there is thus much evidence to support the thermohaline-circulation-driven hypothesis of millennial-scale cyclical climate change, the concept is not without its problems.  One of them is that since the demise of the last great continental ice sheets in the Northern Hemisphere, large and abrupt climatic changes have no longer been observed (Stocker, 2000).  In addition, under more normal circumstances, the model study of Gent (2001) suggests that global warming would lead to the northwest Atlantic becoming both warmer and more saline, with little net effect on surface water density and, therefore, "no evidence of a significant weakening of the thermohaline circulation."  Likewise, the model study of Latif et al. (2000) also predicts a stabilized thermohaline circulation in response to global warming, due to large-scale air-sea interactions that produce anomalously high salinities in the tropical Atlantic that are subsequently advected into the region of North Atlantic Deep Water formation.

Further problems with the hypothesis come from the contention of Wunsch (2000) that "there cannot be a primarily convectively driven circulation of any significance" in the oceans.  As an alternative, Munk and Wunsch (1998) claim that the primary driving force for the thermohaline circulation must be tidal energy, with one-third of the required three terawatts of energy being derived from mixing-driven tidal dissipation in the deep ocean.  Interestingly, Egbert and Ray (2000) have been able to confirm this prediction by means of Topex/Poseidon satellite altimeter data, which they used to empirically quantify the spatial distribution of deep-sea tidal energy dissipation.  Their finding that one terawatt of tidal energy is indeed dissipated in the vicinity of rough bottom topography must be extremely gratifying to Munk and Wunsch; for together with the two terawatts of energy estimated to come from shallow-water tidal energy dissipation, the three terawatts of energy needed to account for the yearly energy loss from the earth-moon system due to the moon's receding from the earth at about four centimeters per year has now been found.  Wunsch, for example, remarks that "it appears that the tides are, surprisingly, an intricate part of the story of climate change, as is the history of the lunar orbit."

"There remain," however, in the words of Egbert and Ray, "many questions … about the implications of these processes for large-scale ocean circulation and climate."  We could not agree more (see our Journal Review Lunar Tides and Climate Change); for the subject is indeed complex.  It would appear, however, that there are mechanisms aplenty to account for most of what we know about tides, ocean currents and climate change - and even the orbit of the moon about the earth! - without the need to invoke a major role for atmospheric CO2 in any part of the story.

References
Bacon, S.  1998.  Decadal variability in the outflow from the Nordic seas to the deep Atlantic Ocean.  Nature 394: 871-874.

Barber, D.C., Dyke, A., Hillaire-Marcel, C., Jennings, A.E., Andrews, J.T., Kerwin, M.W., Bilodeau, G., McNeely, R., Southon, J., Morehead, M.D. and Gagnon, J.-M.  1999.  Forcing of the cold event of 8,200 years ago by catastrophic drainage of Laurentide lakes.  Nature 400: 344-348.

Bianchi, G.G. and McCave, I.N.  1999.  Holocene periodicity in North Atlantic climate and deep-ocean flow south of Iceland.  Nature 397: 515-517.

Broecker, W.S., Sutherland, S. and Peng, T.-H.  1999.  A possible 20th-century slowdown of Southern Ocean deep water formation.  Science 286: 1132-1135.

Egbert, G.D. and Ray, R.D.  2000.  Significant dissipation of tidal energy in the deep ocean inferred from satellite altimeter data.  Nature 405: 775-778.

Gent, P.R.  2001.  Will the North Atlantic Ocean thermohaline circulation weaken during the 21st century?  Geophysical Research Letters 28: 1023-1026.

Latif, M., Roeckner, E., Mikolajewicz, U. and Voss, R.  2000.  Tropical stabilization of the thermohaline circulation in a greenhouse warming simulation.  Journal of Climate 13: 1809-1813.

Marchitto Jr., T.M., Curry, W.B. and Oppo, D.W.  1998.  Millennial-scale changes in North Atlantic circulation since the last glaciation.  Nature 393: 557-561.

Munk, W.H. and Wunsch, C.  1998.  Abyssal recipes II: Energetics of tidal and wind mixing.  Deep-Sea Research 45: 1977-2010.

Raymo, M.E., Ganley, K., Carter, S., Oppo, D.W. and McManus, J.  1998.  Millennial-scale climate instability during the early Pleistocene epoch.  Nature 392: 699-702.

Ruhlemann, C., Mulitza, S., Muller, P.J., Wefer, G. and Zahn, R.  1999.  Warming of the tropical Atlantic Ocean and slowdown of thermohaline circulation during the last deglaciation.  Nature 402: 511-514.

Stocker, T.F.  2000.  Past and future reorganizations in the climate system.  Quaternary Science Reviews 19: 301-319.

Wunsch, C.  2000.  Moon, tides and climate.  Nature 405: 743-744.