Variations Of The Thermohaline Circulation Under Different Atmospheric CO₂ Scenarios In A Climate Model

A climate model (ECHAM5/MPI-OM1) newly developed for the fourth assessment report of the Intergovernmental Panel on Climate Change (IPCC) at Max-Planck Institute for Meteorology is used to study the variations of the Atlantic Thermohaline Circulation (THC) under different increased CO₂ scenarios. Especially, the mechanism of the inter-decadal variability of the THC, the changes of the global oceans due to increased CO₂ in the atmosphere and the responses of the THC in the North Atlantic to increased Atmospheric CO₂, are discussed in details. These topics not only have very important scientific, social and economic meanings; but also have profound influences on international policies making.The control run experiment of the climate model, in which the CO₂ concentration is set at a fixed value, is used to study the mechanism of the inter-decadal variability of the THC from atmosphere-ocean interaction viewpoint and the dynamics of the variations of Greenland-Scotland overflow. A series of meaningful conclusions are drawn, the main of these are as follows:1. The dominant period of interannual variability of THC is 4 years, while the dominant period of interdecadal variability of THC is 24 years, which appears as the strongest signal and the first principal component. The interdecadal variability of the THC works as follows: when the THC has the weakest strength, negative temperature anomalies are induced in the up upper ocean north of the Gulf Stream region due to the reduced northward ocean heat transport. About 5 years later, the negative temperature anomalies reach its maximum. Associated with the build up of negative upper ocean temperature anomalies positive anomalies in the upper ocean density and an acceleration of the subpolar gyre and North Atlantic Current (NAC) take place. After another 3 years the subpolar gyre and NAC receive their strongest values. Stronger subpolar gyre and NAC mean increased transport of salinity into the Greenland–Iceland–Norway (GIN) Seas and will lead to a maximum in the upper ocean density in GIN. After 1 further year enhanced convection is triggered, leading to an increase in the rate of deep water formation and acceleration of the THC. The THC reaches its maximum approximately 4 years after the maximum of subpolar gyre and NAC strength. The total time for the phase reversal is 12 years, consistent with a period of about 24 years.2. The dynamic mechanism of the variations of Greenland-Scotland overflow of control run is: The Faro-Bank (FB) Channel saline inflow increases, and the upper layer density in GIN seas increases. Enhanced convection is triggered, leading to rise of isopycnal. So the density contrast across the ridge and effective height increase, and the hydraulic transports increase. At that time, the FB Channel inflow increases, so the anticlockwise circulation in the GIN seas will be strengthened. Thus the positive barotropic effect for overflow increases. Therefore the hydraulic effect is consistent with barotropic effect for overflow.The results of coupled model (ECHAM5/MPI-OM1) under increased atmospheric CO₂ scenarios are analyzed to study the variation of the THC with global warming. Especially we focus on the variation of the THC strength, the changes of North Atlantic deep water (NADW) formation, the regional responses of the THC in the North Atlantic to increased Atmospheric CO₂ and dynamics of the changes of Greenland-Scotland overflow. The main conclusions are as follows:1. From 2000 to 2100, under increased CO₂ scenarios (B1, A1B and A2), the strength of the THC decreases by 4Sv, 5.1Sv and 5.2Sv respectively, or equivalently reduced by 20%, 25% and 25.1% of the present THC strength. In total approximately 16.2Sv deep water are formed in the Labrador sea (8.2Sv) by convection, both overflows (5.9Sv),and entrainment (2.1Sv) in south of the Denmark Strait region (SDSR) in the 20th century. In the 21st century approximately 12.9Sv deep water are formed in the Labrador sea (5.2Sv) by convection, both overflows (6.1Sv), and entrainment (1.6Sv) in SDSR.2. Analyses show that oceanic deep convective activity strengthens significantly in the GIN Seas owing to saltier (denser) upper oceans, but weakens in the Labrador Sea and in the south of the Denmark Strait region (SDSR) because of surface warming and freshening due to global warming. The saltiness of the GIN Seas is mainly caused by the increase of the saline North Atlantic inflow through FB Channel.3. Since the density contrast across the sills decreases with global warming, the hydraulic transports of Denmark strait and FB channel both decrease under the A1B scenario. The overflow of FB channel also decreases from 3.2 Sv to 2.9Sv with global warming, so the variation of FB channel overflow is hydraulically controlled. However the overflow of Denmark strait increases from 2.7 Sv to 3.3 Sv under the A1B scenario, because it is not a kind of hydraulic transport. There will be more north Atlantic water through FB channel entering the GIN seas, and there will be more water which flows out of the GIN seas and enters the Barents Sea. The inflow through the Fram Strait will increase not only in the upper layer but also in the deeper and bottom layer. Not only in the upper layer but also in the deeper and bottom layer, the anticlockwise circulation in the GIN seas will be strengthened with global warming. Therefore the increase of Denmark Strait overflow is attributed to barotropic effect.