Dynamics of Interdecadal Climate Variability: A Historical Perspective*

Abstract

Abstract The emerging interest in decadal climate prediction highlights the importance of understanding the mechanisms of decadal to interdecadal climate variability. The purpose of this paper is to provide a review of our understanding of interdecadal climate variability in the Pacific and Atlantic Oceans. In particular, the dynamics of interdecadal variability in both oceans will be discussed in a unified framework and in light of historical development. General mechanisms responsible for interdecadal variability, including the role of ocean dynamics, are reviewed first. A hierarchy of increasingly complex paradigms is used to explain variability. This hierarchy ranges from a simple red noise model to a complex stochastically driven coupled ocean–atmosphere mode. The review suggests that stochastic forcing is the major driving mechanism for almost all interdecadal variability, while ocean–atmosphere feedback plays a relatively minor role. Interdecadal variability can be generated independently in the tropics or extratropics, and in the Pacific or Atlantic. In the Pacific, decadal–interdecadal variability is associated with changes in the wind-driven upper-ocean circulation. In the North Atlantic, some of the multidecadal variability is associated with changes in the Atlantic meridional overturning circulation (AMOC). In both the Pacific and Atlantic, the time scale of interdecadal variability seems to be determined mainly by Rossby wave propagation in the extratropics; in the Atlantic, the time scale could also be determined by the advection of the returning branch of AMOC in the Atlantic. One significant advancement of the last two decades is the recognition of the stochastic forcing as the dominant generation mechanism for almost all interdecadal variability. Finally, outstanding issues regarding the cause of interdecadal climate variability are discussed. The mechanism that determines the time scale of each interdecadal mode remains one of the key issues not understood. It is suggested that much further understanding can be gained in the future by performing specifically designed sensitivity experiments in coupled ocean–atmosphere general circulation models, by further analysis of observations and cross-model comparisons, and by combining mechanistic studies with decadal prediction studies.