Deep Water Formation Sites Research Articles

A Simulation of the Last Glacial Maximum climate using the NCAR-CCSM

The National Center for Atmospheric Research-Community Climate System Model (NCAR-CCSM) is used in a coupled atmosphere–ocean–sea-ice simulation of the Last Glacial Maximum (LGM, around 21,000 years ago) climate. In the tropics, the simulation shows a moderate cooling of 3 °C over land and 2 °C in the ocean in zonal average. This cooling is about 1 °C cooler than the CLIMAP sea surface temperatures (SSTs) but consistent with recent estimates of both land and sea surface temperature changes. Subtropical waters are cooled by 2–2.5 °C, also in agreement with recent estimates. The simulated oceanic thermohaline circulation at the LGM is not only shallower but also weaker than the modern with a migration of deep-water formation site in the North Atlantic as suggested by the paleoceanographic evidences. The simulated northward flow of Antarctic Bottom Water (AABW) is enhanced. These deep circulation changes are attributable to the increased surface density flux in the Southern Ocean caused by sea-ice expansion at the LGM. Both the Gulf Stream and the Kuroshio are intensified due to the overall increase of wind stress over the subtropical oceans. The intensified zonal wind stress and southward shift of its maximum in the Southern Ocean effectively enhances the transport of the Antarctic Circumpolar Current (ACC) by more than 50%. Simulated SSTs are lowered by up to 8 °C in the midlatitudes. Simulated conditions in the North Atlantic are warmer and with less sea-ice than indicated by CLIMAP again, in agreement with more recent estimates. The increased meridional SST gradient at the LGM results in an enhanced Hadley Circulation and increased midlatitude storm track precipitation. The increased baroclinic storm activity also intensifies the meridional atmospheric heat transport. A sensitivity experiment shows that about half of the simulated tropical cooling at the LGM originates from reduced atmospheric concentrations of greenhouse gases.

Benthic foraminiferal record of ecosystem variability in the eastern Mediterranean Sea during times of sapropel S 5 and S 6 deposition

High-resolution benthic foraminiferal and geochemical investigations were carried out across sapropels S 5 and S 6 from two sediment cores in the Levantine Sea to evaluate the impact of climatic and environmental changes on benthic ecosystems during times of sapropel formation. The faunal successions indicate that eutrophication and/or oxygen reduction started several thousand years prior to the onset of sapropel formation, suggesting an early response of the bathyal ecosystems to climatic changes. Severest oxygen depletions appear in the early phases of sapropel formation. The initial reduction of deep-water ventilation is caused by a warming and fresh water-induced stratification of Eastern Mediterranean surface waters. During the late phase of S 5 formation improved oxygenation is restricted to middle bathyal ecosystems, indicating that at least some formation of subsurface water took place. During S 6 formation oxygen depletions and eutrophication were less severe and more variable than during S 5 formation. Estimated oxygen contents were low dysoxic at middle bathyal to anoxic at lower bathyal depths during the early phase of S 6 formation but never dropped to anoxic values in its late phase. The high benthic ecosystem variability during S 6 formation suggests that water column stratification at deep-water formation sites was in a very unstable mode and susceptible to minor temperature fluctuations at a millennial time-scale.

About process times in some deepwater formation events in the Mediterranean Sea

Experimental data gathered on some Mediterranean deepwater formation sites have been analysed in order to check the Chapman (Ch), Visbeck and Maxworthy scale laws, their final mixing water depth and process times estimates. All the analysed sites have a high probability to be “internally constrained” so that the baroclinic Rossby radius is the dominant horizontal length scale and the final chimney depth is independent of the rate of rotation as indicated by Visbeck and Maxworthy criteria; the Eady instability times are generally very fast (∼3/ f) but the least final mixing process times are too long compared with the meteorological forcing ones. It is attempted here to identify some new criteria to give some insight on the processes time scale, which is still an open problem. Conclusions about the Rhodes gyre appear questionable, because the experimental data appear inconsistent with any theoretical criterion.

Appraisal of a molecular approach to infer variations in surface ocean freshwater inputs into the North Atlantic during the last glacial

The thermohaline circulation is considered responsible for regulating climate variability at a global scale [Geochim. Cosmochim. Acta 53 (1989) 2465]. It has been argued that it can be disrupted by changes in surface ocean salinity in the sites of deepwater formation [J. Geophys. Res. 96 (1991) 16811]. However, to date reconstruction of sea surface salinity (SSS) has lacked the accuracy to identify the location, extent and amplitude of past meltwater events [Science 282 (1998) 61]. Recently, a measurement based on alkenones (%37:4) has been shown to be related to types of water masses and perhaps surface salinity in the Nordic Seas [Paleoceanography 13 (1998) 694; Terra Nova 10 (1998) 86]. Here, we present new data from surface sediments that further confirm that it is possible to obtain a linear correlation between %37:4 and salinity. Records of %37:4 from three nearby sediment cores in the NE Atlantic show equivalent profiles, with the highest values occurring in the Heinrich layers (HLs). We propose that downcore %37:4 data can be interpreted in terms of decreases in surface ocean salinity (freshening of surface waters) in response to incoming freshwater from sea-ice and/or icebergs to the core relation. The %37:4 data provides further indication of the locally lower surface ocean salinity that prevailed in the North Atlantic during the last glacial period, and particularly during the Heinrich episodes due to the influx of iceberg meltwater.

The Southern Influence

CLIMATOLOGY Changes in oceanic thermohaline circulation, triggered by variations in the flux of glacial melt water to the North Atlantic Ocean that affect the formation rate of North Atlantic Deep Water, can produce rapid dramatic changes in climate. Deep water does not originate exclusively in the North Atlantic, however; it also is made around the edges of Antarctica, where it is referred to as Antarctic Bottom Water. Although most of the attention given to the question of how the modification of sea surface salinity by the addition of glacial melt waters affects thermohaline circulation is still directed at Northern Hemisphere deepwater formation sites, the role of the Southern Ocean is now being investigated more vigorously. Seidov et al. use an ocean circulation model to examine the character of deepwater formation as a function of systematic changes in freshwater fluxes in both the Northern and Southern Hemispheres. They find that thermohaline circulation is driven by both deepwater sources and that the Southern Ocean can overpower the North Atlantic and become a major factor in long-term climate change, illustrating the importance of high-latitude freshwater fluxes in governing climate.— HJS Global Planet. Change 30 , 257 (2001).

GCM studies of intermediate and deep waters in the Mediterranean

Results from GCM simulations of the Mediterranean thermohaline circulation are presented under repeating year wind and surface buoyancy forcings and the reproduction of important physical processes is discussed. It is shown that baroclinic eddies are critical to the effective dispersal of Levantine intermediate water (LIW) throughout the eastern Mediterranean basin. These develop rapidly even in a 1/4 degree model which does not resolve the deformation radius, provided horizontal friction is small enough. It is shown that LIW enters the Adriatic basin and pre-conditions deep water formation which would not otherwise occur due to low surface salinity in winter. The dispersal of Adriatic deep waters is modelled, and it is shown that the introduction of the Gent and McWilliams advective scheme greatly improves the distribution of deep waters in the eastern basin by permitting the formation of dense overflow waters from the Otranto straits. This is achieved with very small parametrised advection that still permits the formation of baroclinic eddies unlike most applications which use the scheme to replace eddies. Results from a 100-year climate simulation are then presented in which the thermohaline circulation has reached equilibrium conditions. Dense water formation in both eastern and western basin still occur after 100 years. While the eastern basin water masses are reasonably realistic, the western basin is a little too cold and fresh, suggesting that insufficient LIW is reaching the deep water formation site in the northwest basin. Further work is needed in this area.

Decoupling of δ13CΣCO 2 and phosphate in recent Weddell Sea deep and bottom water: Implications for glacial Southern Ocean paleoceanography

On a section between 72°S and 42°S and a transect between 60°E and 10°E through the Weddell Sea and the southernmost eastern Atlantic Ocean, the water column was sampled on 72 stations, and the stable carbon isotopic composition of total dissolved inorganic carbon (δ13CΣCO2) as well as the stable oxygen isotopic composition of seawater (δ18O) was determined. These data were compared with potential temperature, salinity, dissolved oxygen and phosphate data from the same stations. The observed δ13CΣCO2/PO43− relationship in the deep Weddell Sea strongly differs from the global Redfield‐driven deep water relationship. We attribute this to enhanced thermodynamic fractionation at sites of bottom water formation that decouples the nutrient signal from the δ13CΣCO2 signal not only in surface and intermediate water masses but also in deep and bottom water. Different, water‐mass specific thermodynamic imprints due to different modes of bottom water formation are assumed to cause the observed deviation from the global δ13CΣCO2/PO43− relationship in the deep Weddell Sea. The influence of increased photosynthetic fractionation, i.e., a more negative than low‐latitude isotopic organic carbon composition, is shown to be minor. As a result, Recent Weddell Sea deep and bottom water δ13CΣCO2 is by 0.4–0.5‰ higher than expected if solely biologic fractionation would occur. A discussion of simple hypotheses of Weddell Sea deep and bottom water formation during glacial times reveals that regardless of what scenario is considered, the thermodynamic imprint on Southern Ocean deep water would increase. This makes it difficult to explain low glacial δ13C values observed in benthic foraminifera from the subpolar Southern Ocean as being calcified in Antarctic source bottom water and thus is in support of hypotheses looking for additional sites of deep water formation.

The influence of air and sea exchange on the carbon isotope distribution in the sea

We explore here the influence of the temperature dependence of isotope fractionation between atmospheric CO2and ocean ΣCO2 on the distribution of carbon isotopes in the ocean. This is accomplished by an analysis of departures from the expected Redfield tie between PO4 and δ13C. We find that for the surface ocean, the temperature influence largely compensates for the biologic influence. In the deep ocean, the temperature influence imprinted at the sites of deepwater formation reduces somewhat the biologically induced difference between the carbon isotope ratios for deep waters produced in the northern Atlantic and in the Antarctic. These same features are reproduced in the Hamburg ocean model. In order to assess the impact of changes in the ratio of ocean mixing rate to wind speed, we have made a model run in which CO2 exchange rates between air and sea were everywhere doubled. As expected, the influence of the thermodynamic effect on the oceanic carbon isotope distribution is magnified.

The Cenozoic ocean circulation based on ocean General Circulation Model results

A series of ocean General Circulation Model (GCM) experiments for Paleocene, Eocene, Miocene and present-day continental geography with atmospheric forcing prescribed from atmospheric GCM experiments is utilized to investigate the changes in surface and deep-water ocean circulation through the Cenozoic. The experiments illustrate a number of similarities with previous observation-based reconstructions, although the timing of the development of particular gyre systems and the nature of the high southern latitude circulation are generally different than previous interpretations. In addition, the model results suggest substantial changes in the sites of deep water formation, including a middle Eocene subtropical source. Because of a number of experimental limitations, the experiments should be considered largely as a sensitivity analysis of the role of changing geography in modifying the ocean circulation.