Back to Search Start Over

Key Role of Internal Ocean Dynamics in Atlantic Multidecadal Variability During the Last Half Century

Authors :
Kim, Who M.
Yeager, Stephen G.
Danabasoglu, Gokhan
Source :
Geophysical Research Letters; December 2018, Vol. 45 Issue: 24 p13,449-13,457
Publication Year :
2018

Abstract

This study presents multiple lines of evidence from observations and model simulations that support a key role for ocean dynamics, rather than external forcings, in Atlantic multidecadal variability (AMV) during the last half century. Observed AMV fingerprints considered here include the low‐frequency spatiotemporal evolution of sea surface temperature, surface heat fluxes, and deep ocean hydrography. While largely absent in the forced response of a large ensemble historical simulations (LENSs), these fingerprints are clearly discernible in a long control simulation where the variability is purely internal. Further evidence derives from initialized decadal prediction simulations, which exhibit much higher skill at predicting the observed AMV of the past 50 years than LENS. The high correlation between the observed AMV and the externally forced version derived from LENS, which has been invoked as evidence for externally driven AMV, is shown to be largely an artifact of concurrent warming since the 1990s. The basin‐wide sea surface temperature variability in the North Atlantic on multidecadal timescales, often called Atlantic multidecadal variability (AMV), has profound impacts on surface climate in the Northern Hemisphere. This work sheds light on the ongoing debate regarding the driving mechanisms of the AMV: whether it is primarily driven by factors associated with natural fluctuations of the coupled ocean‐atmosphere system (such as slow changes in the strength of the ocean circulation) or forced by factors that are external to that system (such as large volcanic eruptions). In this study, we present multiple lines of evidence, from ocean observations as well as climate model simulations, that suggest that AMV is largely attributable to the former and is thus primarily reflecting a natural mode of variability in the climate system. The observed AMV shows distinctive spatiotemporal patterns with concomitant variability in related fieldsWhile absent in the forced signals in historical simulations, this observed evidence is well reproduced in a preindustrial simulationA high correlation between the observed and externally forced, simulated AMV arises by coincidence from concurrent warming after the 1990s

Details

Language :
English
ISSN :
00948276
Volume :
45
Issue :
24
Database :
Supplemental Index
Journal :
Geophysical Research Letters
Publication Type :
Periodical
Accession number :
ejs48199414
Full Text :
https://doi.org/10.1029/2018GL080474