The Nature of Decadal Variability of Atlantic Meridional Overturning Circulation

Stochastic Forcing vs. Air-Sea Coupling

In search of the origin of decadal variability in the climate system, the Atlantic meridional overturning circulation (AMOC) has long been regarded as one of the best candidates. Both the observational and the modeling studies support the idea that the decadal variability in the north Atlantic sector has been closely related with AMOC. However, the nature of the decadal variability of AMOC has been highly debated: does the decadal variability of AMOC result from a coupled mode similar to ENSO, or simply an oceanic response to low-frequency atmospheric forcing?

In north Atlantic, FOAM produces pronounced decadal variability (10-20 years) not only in the AMOC, but also in air temperature. Because of the chaotic nature of the atmosphere, the pronounced decadal spectrum peak from air temperature (in concert with AMOC) has been regarded as strong evidence to support the conclusion that the decadal variability of AMOC in FOAM comes from a coupled mode. However, one stochastic atmosphere run with the same ocean component from FOAM has been performed to test the necessity of the coupling. Surprisingly, the striking spectrum peak of AMOC still comes out. The two experiments lead us to conclude that the decadal variability of AMOC could be produced solely by stochastic forcing, but the coupling provides the decadal variability in the atmosphere (NAO). The stochastic resonance in the ocean model starts from the Labrador Sea, where one giant ocean wave could be produced by convective activity from stochastic atmosphere forcing. This giant ocean wave propagates along the west boundary of Atlantic Ocean all the way to the tropics. The pressure anomaly associated with this wave produces decadal variability of ocean current as well as temperature and salinity anomaly in the subpolar region. The ensuing density anomaly eventually produces the decadal variability of AMOC.

Lagged regression map of dynamic height upon the first principle component of AMOC at 30 m (Left), 1500 m (middle) and 3000 m (right). Notice the shading intervals are different at three depths.

He F., Z. Liu and L. Wu, 2006: The role of oceanic circulation for the decadal variability of the NAO. US CLIVAR Multidecadal to Centennial Global Climate Variability Workshop. Nov 2006, Honolulu, Hawaii.