Water mass transformation by surface buoyancy flux in the Southern Ocean and its possible impact on AMOC in CMIP6 experiments of JMA/MRI

Shogo Urakawa1, Hiroyuki Tsujino2, Hideyuki Nakano1, Takahiro Toyoda3, Kei Sakamoto4 and Goro Yamanaka3, (1)Meteorological Research Institute, Tsukuba, Japan, (2)Meteorological Research Institute, Japan Meteorological Agency, Tsukuba, Japan, (3)Meteorological Research Institute, Tsukuba, Ibaraki, Japan, (4)Meteorological Research Institute, Ibaraki, Japan
Abstract:
Atlantic Meridional Overturning Circulation (AMOC) is one of ocean general circulations that involves North Atlantic Deep Water (NADW) formation. It accompanies a large northward heat transport and is considered to be important for climate. NADW flowing into the Southern Ocean is transformed into Circumpolar Deep Water (CDW) by mixture with ambient waters. A part of CDW is considered to be pumped up to the surface layer in the Southern Ocean by Ekman upwelling and transformed to the lighter waters which go back to the NADW formation area. So, the water mass transformation in the Southern Ocean plays an important role in the AMOC. The magnitude of AMOC is underestimated by more than 3 Sv in CMIP6/OMIP experiments with our global ocean model developed in Meteorological Research Institute. On the other hand, our CMIP6 historical experiments with our coupled general circulation model well reproduces the magnitude of AMOC, although its ocean model component is the same as one used in the OMIP experiments. This study investigates the difference in AMOC from the point of view of the water mass transformation in the Southern Ocean. Our coupled model represents stronger zonal wind stress over the Southern Ocean and its peak slightly shifts north. This results in a larger northward transport of sea-ice and the subsequent increase of fresh water input to the ocean. It leads to more transformation of upper CDW into the lighter waters and possibly accelerating the AMOC in the model. This result is consistent with the earlier studies on the water mass transformation in the Southern Ocean such as Abernathey et al. (2016) and Pellichero et al. (2018). On the other hand, the increase in sea-ice melt accompanies the enhanced sea-ice formation around Antarctica in our coupled model. However, it does not result in better representation of dense shelf water around Antarctica and Antarctic bottom water. Our ocean-only model shows the stronger bottom cell of MOC, which might be partially supported by dense water formation by SSS restoring only applied in the CMIP6/OMIP experiment. This inconsistency might result from coarse horizontal resolution, lack of important processes around Antarctica in our models such as water mass transformation within ice cavities, and inadequate representation of descending currents along the Antarctic continental slopes.