Abrupt climate change and their effect on deep water masses

The change of ocean chemistry between glacial and interglacial periods has been studied since the fifties, by investigating the relationship of ocean carbonates and the ocean waters. Large changes in the marine carbonates associated with climate change were observed by early marine geologists (Arrhenius, 1952). Distinguishing the contribution of the change in biogenic production of carbonates in the surface oceans, by the change in ocean’s chemistry has been the focus since the first development of marine tracers. Data from ice cores clearly shows orbitally modulated glacial-interglacial cycles e.g., (Dansgaard et al., 1983; Cuffey and Clow, 1997; Grootes and Stuiver, 1997; Johnsen and Jensen, 2001). The combined use of the carbon and cadmium traces to investigate the difference in the ocean composition between glacial and interglacial periods, reveal an interesting story which lead inevitably to the conclusion that the deep waters formation dynamics are responsible for the observed differences.

An important observation was made by (Shackleton et al., 1983); the carbon isotopic signature of deep waters in the Atlantic and Pacific oceans was lighter during the glacial periods, and that the amount of depletion in the Atlantic was at least twice as much as in the Pacific. Interpreting this result as a global decrease in nutrients content in the deep ocean was unlikely as the timescale involved is too short. Alternatively, a transfer of light organic carbon from the terrestrial reservoir into the ocean would be assumed to cause the difference in the isotopic signal.

Deep-water circulation changes can also have occurred during the glacial periods, in principle it is possible to use benthic foraminifera δ13C record to infer differences in the deep water circulation between glacial period and the Holocene. The present day difference in deep water nutrient content between the two ocean basins, is consistent with the difference in isotopic composition of the two deep-water masses. The foraminifera record shows a δ13C difference of about 1 per mil. If we assume that carbon isotope is a proxy for nutrient, this correspond to a difference of about 1 mmol/kg in dissolve inorganic phosphorus (DIP) between the ocean basins, which is consistent with the present day Pacific-Atlantic nutrient distribution. In principle, is therefore plausible that changes in the isotopic composition during the glacial period, is a direct reflection of nutrients concentrations. During glacial periods the isotopic signature between the basins was almost the same, this implies that the nutrient content in the world’s ocean was rather homogeneous. As the present deep water ocean contributor is divided equally by the nutrient-poor NADW and the nutrient-rich AABW, we can conclude that the similar isotopic signal during glacial periods is a consequence of a much less influence of the NADW in glacial ocean deep waters.

Confirmation from this nutrient model for glacial and interglacial periods comes from analyses on cadmium carbon ratio on benthic foraminiferal tests. Results from Atlantic sediments show an increase in Cd/Ca ratio, therefore an increase in DIP content, during glacial periods, confirming the decreased NADW production hypothesis. More in-depth Cd/Ca studies on core transect in the North Atlantic show a highly stratified water column during glacials, consistent with a less intense and shallower deep-water production in this region.

Comparative analyses on cadmium and δ13C on cores located at Florida Strait at shallow depth, and western northern Atlantic at higher depth, confirm the role of the reorganization of the NADW with a less intense and shallower ventilation during Younger Dryas, and Heinrich 1 (Came et al., 2008). The study also highlight the still uncertain role of the Antarctic Intermediate Waters (AAIW), for example contrasting evidences from nutrient proxy show from one side an increased influence from the North Atlantic Intermediate Waters, during Younger Dryas (Marchitto et al., 1998) while evidence in the north Atlantic intermediate-depth show an increased influence of the AAIW (Rickaby and Elderfield, 2005).

This suggests a very different role of the two water masses in the glacial AMOC regime, thus is critical a better understanding of this dynamic to interpret millennial shifts in climate system. Nevertheless a general agreement exists among data and models upon the general North Atlantic response, which is characterized by a reduced flux or a shallowing of the northern deep water component, allowing a northward and upward extension of the southern component during glacial periods and Heinrich events. (Sarnthein et al., 1995).

When in the science community the concept of abrupt climate change was widely accepted around the mid 80’s, the same methods and techniques were used to analyze the water dynamics during these abrupt events, thanks also to improvements in the analytical methods, which allows higher temporal resolution. Very quickly the role of the Meridional Overturning Circulation in the Atlantic begins to be appreciated. δ13C mapping o the North Atlantic shows evidence of a three- flow state of the MOC. (i) A modern intergalcial (stadial) state, with vigorous overturning circulation and deep water formation at high latitudes. (ii) A glacial (interstadial) flow state, with less vigorous and shallower MOC, together with southern deep water reaching northern latitudes. And a Heinrich flow state, with little or no overturning circulation. (Sarnthein et al., 1995; Elliot et al., 2002; Rahmstorf, 2002).

Consequently, the concept of heat piracy describes the heat distribution as the MOC state is the primary responsible of the neat inter-hemispheric heat flux. What has thoroughly been observed in ice cores from Greenland and Antarctica, is an anti phasic relationship of main temperature proxy reflecting a heating of the Antarctic during Heinrich events, a stationary condition during B/A warm events and a resume of the Antarctic heating during Younger Dryas, where temperature in the northern hemisphere are plunging down of 8 to 16 degrees Celsius. This has for a long time been described as the bipolar see-saw.

The explanation of this mechanism is still under refinements, but can be generally described as the “heat piracy” performed by the NADW during interglacial periods, that transport heat to higher latitudes leaving the southern hemisphere relatively deficient. When a sudden Heinrich event occur, an almost complete shutdown of MOC and consequently of the NADW occurs, leaving an accumulation of heat in the southern Hemisphere.

During these cold events, the major effect on global climate is characterized by a reorganization of the atmospheric circulation with a southern shift of the mean Inter Tropical Convergent Zone (ITCZ), which lead to intensification of summer monsoon in South America. Speleothem records from cave located in eastern Brazil, provide evidence of an increased monsoonal activities, during the shift of the ITCZ caused by defects in NADW as well as orbitally induced southward shift of ITCZ during peak in summer insolation at southern latitudes (Wang et al., 2004; Cruz et al., 2005). Continental consequences of the shut down of the THC are helpful to draw a general picture on global climate, and to test the consistency of the related interpretation. For example a question that might arise is: how the continental climate might be affected during a Heinrich cold event? Strengthened north eastern trade winds as a southern shift of the ITCZ during Heinrich events can determine a positive feedback loop where stronger winds determines a further shift in ITCZ or a cooling of a certain region of the equatorial and subtropical sea, feeding more energy in the wind pattern development (Xie, 2004). (Jaeschke et al., 2007) shows an increase in Ti/Ca in the northeast Brazil during Heinrich events caused by increased weathering and terrigenous transport. (Jennerjahn et al., 2004) shows interestingly a cooling in the southern hemisphere during a Heinrich events, this still is consistent with a positive temperature gradient (NH-SH) with higher temperatures in the southern hemisphere and cooler in the northern hemisphere. (Wang et al., 2006) shows an example of teleconnection between northern and southern hemisphere and an almost perfect anti-correlation in the isotope curves measured on two cave records. This is consistent with an intensification of the monsoons in the southern hemisphere and a cooling in the northern hemisphere.

It is abundantly clear from the paleoclimate record that abrupt changes have occurred in the global climate system at certain times in the past. Large climate transitions happened in human timescales, for examples the end of the Younger Dryas, and various D-O events. Nonlinear responses apparently have occurred as critical thresholds were passed. Despite the huge amount of data gathered in the last three decades, our knowledge of what these thresholds are is completely inadequate; we cannot be certain that anthropogenic changes in the climate system will not lead us, inexorably, across such a threshold, beyond which may lie a dramatically different future climate state (Broecker, 1987). Only by careful attention to such episodes in the past can we hope to comprehend fully the potential danger of future global changes due to human-induced effects on the climate system. The mechanisms responsible for the triggering of this atmosphere-ocean reorganization are yet not well understood. Both external and internal forcing play a part, however a fundamental role must be played by the ocean, especially the by the formation of deep water masses in controlling past rapid climate changes. Future possible climate change must consider this key aspect of the climate system, and the science effort in abrupt paleoclimatolgy must be concerned by the role of deep water system.