Interannual Variability in the Global Meridional Overturning Circulation Dominated by Pacific Variability: Possible Implications for ENSO Variability

Neil F Tandon, University of Toronto, Department of Physics, Toronto, ON, Canada, Paul J Kushner, University of Toronto, Toronto, ON, Canada, Mark A Cane, Lamont -Doherty Earth Observatory, Palisades, NY, United States and Oleg Saenko, Canadian Centre for Climate Modelling and Analysis, Victoria, BC, Canada
Abstract:
The view prevails that the global meridional overturning circulation (GMOC) is primarily reflective of the Atlantic meridional overturning circulation (AMOC). While this is true in the mean, this is likely not true when considering interannual variability. This paper shows that interannual variability of the GMOC in nine coupled climate models is heavily dominated by variability in Pacific Ocean overturning. This is apparent not just near the surface, but down to depths exceeding 4km. We perform more detailed analysis on large initial condition ensembles of the CESM1 and CanESM2 models. This reveals that the dominant mode of variability in deep Pacific overturning (DPO) shows strong coupling to sea surface temperature in the NINO3.4 region of the eastern tropical Pacific. In CESM1, annual mean NINO3.4 anomalies show a strong positive correlation with an index of DPO at lead times of two years: positive NINO3.4 anomalies lead to counterclockwise anomalies in DPO, implying enhanced southward volume transport in the upper equatorial Pacific Ocean. There is also strong negative correlation between NINO3.4 and DPO at lag zero, suggesting a 4 year feedback cycle between NINO3.4 and DPO anomalies. Such an interannual feedback cycle is also apparent in the CanESM2 model, but with a shorter timescale. Note that the NINO3.4 index in these models does not show an unrealistic quasi-biennial spectral peak like that in the older CCSM3 model. Assessing the overall realism of surface-deep ocean coupling in these models is challenging due to the lack of DPO observations. It is unclear if available near-surface observations place a strong constraint on DPO variability. Altogether, these results point to a possible role for DPO in shaping long-term ENSO variability.