Skip to main content

Historical record of AMOC variability inconsistent in ocean reanalysis products

While the circulation of the Atlantic Ocean has a complex three-dimensional spatial structure, the zonally integrated flow in the basin, referred to as the Atlantic Meridional Overturning Circulation (AMOC), is largely responsible for the net northward oceanic heat transport on climate-relevant timescales.  This makes the AMOC a key metric for quantifying the mean state of the ocean and its historical change. Karspeck et al. (2015) examines the mean and variability of the AMOC as represented in six ocean reanalysis products over the period 1960-2007, with particular focus on multi-decadal trends and interannual variability at 26.5°N and 45°N. Ocean reanalysis products, developed independently at modeling centers around the world, use historical surface observations and subsurface ocean hydrography to constrain the solution of ocean general circulation models. They represent an estimate of the spatially and temporally complete state of the global ocean given the sparse network of historical ocean observations.

Ocean reanalysis
Time series of AMOC anomaly at 1000 m depth at 45°N (top panels) and 26.5°N (bottom panels) for the set of ocean reanalysis products (left panels) and the set of No Assimilation forced ocean model simulations (right panels). The 1960-2007 time-mean has been removed from each time series. The means (in Sv) are indicated within brackets at 45°N and 26.5°N (respectively) in the legends. Time series from RAPID are included for comparison.

One of the key findings of this work is that, contrary to the conventional expectation that the imposition of subsurface data constraints will draw the AMOC in reanalysis products into agreement, the historical AMOC variability is less consistent among the reanalysis products than in corresponding simulations without subsurface data constraints.  This finding has implications for how the scientific community understands the implicit uncertainties in the current generation reanalysis products.

Written by
Alicia Karspeck, National Center for Atmospheric Research

A. R. Karspeck1, D. Stammer2, A. Koehl2, G. Danabasoglu1, M. Balmaseda3, D. M. Smith4, 
Y. Fujii5, S. Zhang6, B. Giese7, H. Tsujino5, and A. Rosati6

1 National Center for Atmospheric Research

2 University of Hamburg, DE

3 European Centre for Medium-range Weather Forecasts, UK 

4 Met Office Hadley Centre, UK 

5 Meteorological Research Institute, Japan Meteorological Agency, Japan 

6 Geophysical Fluid Dynamics Laboratory, NOAA

7
Texas A&M University

Topics