Variability of the Atlantic overturning circulation
June 2016 - New research at the Met Office suggests that the weakening of the Atlantic overturning circulation that has been observed since 2004 is part of variability occurring over decades.
The Atlantic Meridional Overturning Circulation (AMOC) is made up of a northwards flow of warm water in the upper layers of the Atlantic, which cools and sinks at high latitudes and a southwards flow of the cold dense water at depth. It plays an important role in our climate by transporting heat northwards in the Atlantic and keeping Europe relatively warm. Observations from theRAPID array have found a significant weakening of this circulation since the start of the observations in 2004, raising the question of whether this is part of an ongoing trend or part of climate variability.
The Met Office has produced a new ocean 'reanalysis' that combines a state-of-the-art model of ocean dynamics with observations of the ocean from satellites and ocean floats that sample the ocean below the surface. This produces a picture of how the ocean has evolved over the last couple of decades. The new study by Met Office scientists has found that this reanalysis reproduces the decrease in the Atlantic overturning circulation observed by the RAPID array since 2004, as well as its year-to-year variability, despite the RAPID observations not being ingested into the model.
The reanalysis extends back before the start of the direct RAPID observations in 2004, and so allows us to put them into a longer-term context. The weakening of the overturning circulation was found to follow a strengthening of the circulation since the mid-1990s. Thus it is possible that the recent weakening is a recovery from a period of anomalously strong circulation, rather than part of a long-term declining trend.
The results do not rule out the possibility that the observed weakening is a combination of decadal variability and a longer-term decreasing trend, as is projected by climate models in response to increasing greenhouse gases. Such a trend would only be detectable after more years of observations.
What will happen in the future?
Although the reanalysis does not predict the future, the research also shows a link between changes in the Atlantic overturning circulation and changes in density in the Labrador Sea several years earlier. The Labrador Sea is no longer becoming less dense, and has shown indications of increasing density in the last year. This suggests that the decline in the Atlantic overturning circulation might halt or even temporarily reverse. Continued monitoring of the Labrador Sea density may provide useful information about the future behaviour of the Atlantic overturning circulation, however there remain many uncertainties around how the circulation might vary over the next decade.
Decadal prediction of the Atlantic, including the overturning circulation and its impacts on the climate, is an ongoing area of research with the Met Office decadal prediction system. Several previous studies have suggested that the Atlantic overturning circulation is potentially predictable a few years in advance, and a recent Met Office study found that historical heat content changes in the North Atlantic would have been predictable a few years in advance, possibly because of the representation of the AMOC.
Although there are uncertainties over what might happen in the short term, the longer-term view is that a weakening of the circulation in the 21st century is still very likely as a result of increasing greenhouse gases.
Do we still need to observe the Atlantic overturning circulation?
Although our technique has been very good at reproducing the Atlantic overturning circulation so far, more than a decade of observations would be required for a thorough validation. In particular, the reanalysis may be less good at capturing future changes that occur through different processes, for instance from changes in the dense waters that overflow the underwater ridge between Greenland, Iceland and Scotland into the Atlantic basin. The deep ocean (below 2000m) is still very poorly observed, so changes in the deep ocean that affect the circulation might not be properly represented.
The reanalysis is a valuable tool for filling the gaps between observations and so obtaining a complete estimate of the state of the ocean over the past 25 years. The RAPID observations have proved to be a very useful benchmark for the reanalysis, as they are not directly used in its production. New observational arrays have recently been deployed further north as part of the OSNAP program and are expected to provide valuable data for validation and for understanding ocean processes at higher latitudes.
More details may be found in the article by L. Jackson et al., "Recent slowing of Atlantic overturning circulation as a recovery from earlier strengthening", recently published in Nature Geoscience