The Internal Generation of the Atlantic Ocean Interdecadal Variability

Abstract

Numerical simulations of a realistic ocean general circulation model forced by prescribed surface fluxes are used to study the origin and structure of intrinsic interdecadal variability of the ocean circulation. When eddy-induced turbulent diffusivities are low enough, spontaneous oscillations of the Atlantic meridional overturning circulation (AMOC) with periods O(20) yr and amplitude O(1) Sv (1 Sv ≡ 106 m3 s−1) emerge. The transition from the steady to the oscillatory regime is shown to be consistent with a supercritical Hopf bifurcation of the horizontal Peclet number. Adding atmospheric thermal damping is shown to have a very limited influence on the domain of existence of intrinsic variability. The spatial structure of the mode consists of a dipole of sea surface temperature (SST)/sea surface height (SSH) anomalies centered at about 50°N with stronger variance in the western part of the subpolar gyre, in agreement with the observed Atlantic multidecadal oscillation (AMO) signature in this region. Specific features include a westward propagation of temperature anomalies from the source region located on the western flank of the North Atlantic Current (NAC) and a one-quarter phase lag between surface and subsurface (800 m) temperature anomalies. Local linear stability calculations including viscous and diffusive effects confirm that the North Atlantic Current is baroclinically unstable on scales of O(1000) km with growth rates of O(1) yr−1. Both the spatial structure of the mode and the period agree in magnitude with in situ measurements in the North Atlantic, suggesting that this intrinsic ocean mode participates in the observed Atlantic bidecadal climate variability.

Direct temporal cascade of temperature variance in eddy-permitting simulations of multidecadal variability

“…More specifically, the variability disappears for diffusivity values that are typically in the range of observational estimates (500 − 1000 m 2 s −1 ) casting therefore some doubts on the relevance of this self-sustained internal mode for the observed variability. The same conclusions are found for the influence of the isopycnal diffusivity within the Gent and Mcwilliams (1990) parameterization in a realistic configuration (Arzel et al 2018). Moreover, (LaCasce and Pedlosky 2004) have shown that large-scale anomalies associated with Rossby waves are prone to meso-scale baroclinic instability; they further suggested that the presence of Rossby waves in the mid to high latitudes is unlikely because at these latitudes, Rossby wave periods are slow compared to the destabilization 5 Accepted for publication in Journal of Climate.…”

“…The same mode of variability was later demonstrated to exist in a realistic configuration of the North Atlantic (Sévellec and Fedorov 2013). Arzel et al (2018) further showed that the internal ocean mode produces maximum SST variance in the western part of the subpolar gyre, in agreement with the observed signature of the AMO in the North Atlantic.…”

“…DOI 10.1175/JCLI-D-19-0921.1. jkstra and Ghil 2005;Arzel et al 2007;Sévellec and Huck 2015;Arzel et al 2007;Huck et al 2015;Jamet et al 2016;Arzel et al 2018). The mode's existence was first shown in a rectangular flat-bottomed single-hemispheric basin, with prescribed surface heat fluxes and Planetary Geostrophic dynamics (Greatbatch and Zhang 1995;Colin de Verdière and Huck 1999).…”

“…Sensitivity studies carried out at low resolution based on either idealized or realistic oceanic configurations have demonstrated the critical influence of the horizontal and eddy-induced turbulent diffusivities on the properties of the internal ocean mode (e.g. Arzel et al 2018).…”

North Atlantic Ocean Internal Decadal Variability: Role of the Mean State and Ocean‐Atmosphere Coupling

“…The processes responsible for the AMOC decadal climate variability were reported to be due to (1) the atmospheric stochastic forcing (Frankignoul & Hasselmann, ; Delworth & Greatbatch, ), (2) self‐sustained oceanic intrinsic variability (Arzel et al, ; Sérazin et al, ), or (3) coupled mode of variability (Timmermann et al, ). Most low‐resolution ocean models show little variability when using bulk formulation as surface boundary condition (Danabasoglu et al, ).…”