Haline stratification and the onset of glaciation in the Pliocene

The northern part of the Pacific Ocean is characterized by conspicuously low salinity waters in comparison with other parts of the world (see map below)In fact the low salinity levels are a feature of only the upper part of the ocean in this region, as there exists a strong and persistent haline stratification (halocline), with a change to typical salinity at a water depth of around 200 m.  This stratification has significant climate and environmental implications for the north Pacific region, and for adjacent regions as well.  

Up until about 2.7 million years ago the northern Pacific behaved like most other high-latitude oceanic regions.  There is evidence that a strong halocline did not exist before that time, and hence there was ready vertical mixing of deep and shallow waters.  This vertical mixing brought abundant nutrients such as nitrate, phosphate and silicate to surface, and also ensured that sea-surface temperatures were maintained at moderate levels year-round.  The steady supply of nutrients fed a healthy population of planktonic species, including diatoms that produced a steady rain of siliceous sediment to accumulate on the sea floor.

North Pacific Ocean Drilling Project cores show a sudden drop in the rate of accumulation of silica (opal) on the northern Pacific sea floor at 2.7 m.y. ago. This is generally interpreted to be a product of the initiation of the halocline stratification (Haug et al., 1999).  Coincident with the drop in silica levels, is a pronounced increase in the magnetic susceptibility of sea-floor sediments.  This feature represents an increase in the volume of terrigenous rock and mineral fragments delivered to the ocean floor by ice rafts.

The critical question is: What caused the north Pacific halocline to form 2.7 m.y. ago?  

Such a halocline could be related to a number of different factors, including the input of a large volume of fresh river water, a change in ocean circulation patterns or a change in overall ocean water temperature.  There is little evidence of a change in river flow rates or patterns at that time, and although the completion of the Isthmus of Panama did affect Pacific and Atlantic currents, the major closure happened almost 2 m.y. previously, at 4.6 m.y. 

Stratification of the ocean is enhanced when the upper water has a low salinity, and is reduced when the upper water is relatively cold.  According to Sigman et al. (2004) the thermal control over stratification becomes less significant when the overall temperature drops. They suggest that the long-term cooling during the late Tertiary, and the superimposed effects of Milankovitch cycling, finally led to a situation where, at 2.7 m.y., the thermal gradient was reduced to a point where an already existing salinity gradient became capable of maintaining stratification.  Such a salinity gradient has a positive feedback because the halocline prevents mixing with subsurface water, and ongoing precipitation and river input contribute to the low salinity.

Sigman et al. (2004) also point out another important positive feedback related to an extensive area of halocline in polar waters.  The haline stratification restricts the venting of deep-ocean carbon dioxide (which typically sinks in equatorial regions and is released in polar regions), and thus leads to an amplification of the climate cooling.

In a recent paper in Nature climate scientists from Europe and the US describe another positive-feedback consequence of the development of the northern Pacific halocline this one with an interesting twist ( Haug et al., 2005).  One of the consequences of any sort of ocean stratification is that the surface water temperatures are no longer moderated by mixing with deeper water.  As a result, the north Pacific surface water was able to become relatively warm in the summer and very cold in the winter.  Haug et al. (2005) provide evidence that summer-time sea surface temperatures increased by as much as 7 C after 2.7 m.y., and they suggest that the warm summer water of the northern Pacific may have played an important role in the initiation of the glaciation in northern North America.  The air over the warm northern Pacific would have been relatively moist, and in the later summer and early fall this moist air would have migrated east over northern Canada, resulting in heavy snow falls, eventually accumulating enough to form the Laurentian Ice Sheet.

An important corollary of this work is the possibility that the current anthropogenic warming trend could lead to destabilization of the northern Pacific halocline, with significant consequences.  Apart from impacting the weather throughout the northern hemisphere, loss of the halocline would enhance the rate at which deep-ocean carbon dioxide is vented to the atmosphere, leading to even greater warming.


Haug, F, Sigman, D, Tiedemann, R, Pedersen, T, and Sarnthein, 1999, Onset of permanent stratification in the subarctic Pacific Ocean, Nature, V. 401, p. 779-782

Haug, F, Ganapolski, A, Sigman, D, Rosell-Mele, A, Swann, G, Tiedemann, R, Jaccard, S, Bollmann, J, Maslin, M, Leng, M, and Eglington, G, 2005, North Pacific seasonality and the glaciation of North America 2.7 million years ago, Nature, V. 433, p. 821-825 (Feb. 2005)

Sigman, D, Jaccard, S, Haug, F, 2004, Polar ocean stratification in a cold climate, Nature, V. 428, p. 59-63.

Steven Earle, 2005. Malaspina University-College, Geology Department, Return to Earth Science News