Intermittency of the AMOC multi-decadal variability: Quantification and Mechanisms

Wei Cheng1, Wilbert Weijer2, John Chiang3, Gokhan Danabasoglu4, Stephen G Yeager5, Who M Kim6, Peter R Gent5, Jiaxu Zhang7 and Dongxiao Zhang8, (1)University of Washington/CICOES and NOAA/PMEL, Seattle, United States, (2)Los Alamos National Laboratory, Los Alamos, NM, United States, (3)UC Berkeley, Geography, Berkeley, CA, United States, (4)National Center for Atmospheric Research, Climate and Global Dynamics, Boulder, United States, (5)National Center for Atmospheric Research, Boulder, CO, United States, (6)National Center for Atmospheric Research, Boulder, United States, (7)University of Washington, CICOES, Seattle, United States, (8)CICOES/University of Washington and NOAA/PMEL, Seattle, United States
Abstract:
Traditional spectral analyses suggest that modeled AMOC indices in the North Atlantic have elevated spectral energy on decadal to multi-decadal timescales, which may or may not be statistically significant; meanwhile, wavelet analysis reveals that the increased spectrum energy is intermittent in time, even in control simulations where modeling parameters are held constant. The reason behind such intermittency is unclear. In this study, we first quantify the intermittency by examining AMOC45N index wavelet on the multi-decadal time scales in the CESM control simulation (using the last 1200 years of a 2200-year run). Results show that such intermittency corresponds to centennial modulation of the multi-decadal AMOC variability. The potential roles of “remote” forcing from both the Arctic and the South Atlantic on the centennial modulation are then examined. In particular, we seek to understand if salt transport from these regions may perturb the subpolar ocean stratification and precondition it for particularly strong or weak AMOC variability. Preliminary results suggest that the active AMOC multi-decadal variability periods are concurrent with increased variability in the southward volume and salt transport east of the Greenland and across the Davis Strait, but show lagged and weaker correlation with variability in salt transport from the South Atlantic. When the AMOC multi-decadal variability is strong, southward volume and salt transport along the Eastern Greenland Current and across the Davis Strait is weak while the thermocline depth in the eastern subpolar gyre is depressed and ocean stratification there weakens.