Abstract

Numerical experiments with a two-dimensional nonhydrostatic model were carried out to investigate the generation processes of open-ocean deep convection, deep water formation and ventilation around the Maud Rise in the Weddell Sea, Antarctica. The thermobaric effect, i.e. an increase in the thermal expansion rate of sea water with pressure (water depth), is essential for the onset of deep convection in this region and for the overturning of the water column to occur abruptly. Thermal-like plumes induced by the thermobaric instability destroy the thermocline (halocline) and transport the cold and less-saline mixed layer water into the warm and more-saline underlying layer. Then the underlying water ascends to push the thermocline (halocline) up until it disappears at the sea surface. It takes only a few days for the thermocline (halocline) to disappear without sea-ice cover. On the other hand, it takes more than 30 days with a sea-ice cover because of the reduction of the cooling rate. The stability of the water column around the Maud Rise in nopolynya winter 1986 was examined. The area over the rise is likely to overturn before the end of the cooling season while the marginal area is not. This suggests that the area over the rise may be one of the source regions for the convective features observed throughout the Weddell Sea. The rate of deep water formation due to this overturning and the associated upward heat flux are estimated at 0.91 × 105 m3 s−1 and 36 W m−2 over one year, respectively, which are 2 ∼ 3 times of those caused by the entrainment of the Weddell Deep Water into the mixed layer when no overturning occurs. No-overturning stations on the margin of the rise are classified into two types. At one type of station, salinity in the relatively thick mixed layer is too low for the water column to be destabilized before the end of the cooling season; at the other type of station, the relatively thin mixed layer with a relatively high salinity and the warm Weddell Deep Water inhibit an overturning through a previously described negative feedback process. Sensitivity of the model result to several physical parameters expressing turbulent mixing was also examined.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call