Australia: The Land Where Time Began

A biography of the Australian continent 

Atlantic Meridional Overturning Circulation (AMOC) Currently Weakest in Last Millennium

Heat on the Earth is redistributed by the Atlantic Meridional Overturning Circulation (AMOC), one of the major circulation systems of the Earth, has a major impact on climate. In this study a variety of published proxy records are compared in order to reconstruct the evolution of the AMOC since about AD 400. A picture of the AMOC has emerged that is fairly consistent: There was an initial weakening that began in the 19^th century following a long, relatively stable period, and this was followed by a 2^nd, and more rapid, decline that occurred in the mid-20^th century, which led to the weakest state of the AMOC, occurring in recent decades.

The ANOC is a major mechanism for the redistribution of heat on the Earth, and it is an important factor in the variability and change in the climate. The AMOC is a sensitive nonlinear system that depends on subtle density differences of the thermohaline in the ocean, and major transitions of the AMOC have been implicated in, for example, millennial climate events during the last glacial period (Rahmstorf, 2002). Evidence has been discovered that the AMOC is slowing down in response to anthropogenic global warming (Caesar et al., 2018), as is predicted by climate models, and that the AMOC is presently in its weakest state for more than 1,000 years (Thornalley et al., 2018). Longer-term reconstruction must be based on          proxy data because it was only in 2004 (Smeed et al., 2018) that measurements of the AMOC began. There are in general 3 different types of AMOC proxies:

1.    Reconstructions of surface or subsurface temperature patterns in the Atlantic Ocean that reflect the changes in the transport of ocean heat that are associated with the AMOC (Thornalley et al., 2018; Rahmstorf et al., 2015);  

2.    Reconstruction of the properties of water masses in the subsurface, e.g., the advance of subpolar water versus subtropical slope water that reflect changes in the AMOC (Sherwood et al., 2011); and

3.    Evidence of physical changes in deep sea currents, such as those that are reflected by changes in the sediment grains (Thornalley et al., 2018).

A combination of all 3 types of proxy is needed to provide robust evidence concerning the evolution of the AMOC, because all 3 types of proxy are limited in their representation of the AMOC (all 3 can be influenced by factors that are additional to changes in the AMOC).

This article uses several different proxy indicators of the evolution of the AMOC over the past 100 to nearly 2,000 years that are largely independent, to provide strong evidence of the decline of the AMOC in the 20^th century is unpresented over the past decades, and that AMOC is in its weakest state in more than 1,000 years.

The proxies were collected from various locations in the Atlantic Ocean or the surrounding land areas and represent either different subsystems that are associated with the AMOC (such as the density (Thornalley et al., 2018), the presence of subtropical versus subpolar slope waters along the East coast of North America (Sherwood et al., 2011; Thibodeau et al,, 2018), or the effect of changes in the Atlantic meridional transport of heat that are associated with the AMOC (Caesar et al., 2018; Thornalley et al., 2018; Rahmstorf et al., 2015; Cheng et al., 2017), as well as productivity changes in the surface that have been related to the AMOC (Osman et al., 2019; Spooner et al., 2020). The records that were collected from marine sediments (sortable-silt data (Thornalley et al., 2018), proxy records of subsurface ocean temperatures (Thornalley et al., 2018), ¹⁸O, obtained from benthic foraminifers (Thibodeau et al., 2018) δ¹⁵N of deep sea gorgonian corals (Sherwood et al., 2011), relative abundance of particular foraminifera (Turborotalita quinqueloba (Spooner et al., 2020), are the records that go back to (AD400), the furthest back in time. However, the temperature-based AMOC index (Rahmstorf et al., 2015) is based on a land-and-ocean temperature reconstruction from the Northern Hemisphere that uses a range of terrestrial proxies, which include, e.g., tree rings and ice core data (Mann et al., 2008). An estimate of AMOC related changes in productivity in the region of the subpolar gyre (Osman et al., 2019), is provided, furthermore, by data that was obtained from Greenland ice cores (the concentration of methanesulfonic acid). Most of these records extend into the modern era, for which there are additional AMOC proxies that are based on instrumental temperature records (Caesar et al., 2018; Cheng et al., 2017).

They provide a consistent picture of the evolution of the AMOC since about AD400: prior to the 19^th century, according to which the AMOC was relatively stable, in spite of different locations, time scales and processes that are represented by these proxies. Beginning during the 19^th century, a decline in the AMOC is evident in all the proxy records. A phase of particularly rapid decline that is found in several proxies that are largely independent began around 1960. In the 1990s there was a short-lived recovery is evident before a return to decline that began in the middle of the first decade of the 2000s. Additionally, all indices show multidecadal variability, albeit with different amplitudes and frequencies which make it questionable whether this is driven mainly by the AMOC. According to Caesar et al., it is probable that some of the references relate to the large range in temporal resolution in the proxies (from annual to 50 year binning), though others are probably due to complicating factors, such as influences on a proxy system that are not related to the AMOC (e.g., changes in trophic structure of coral’s food source in δ¹⁵N, local fluctuations in circulation that affects single site palaeoceanographic reconstructions of other controls on subpolar heat content (Keil et al., 2020). That different components of the AMOC respond on different timescales may be an additional factor. Changes in the deep ocean appear to function on different timescales, while the strength of the AMOC, which is typically measured at 26^ON, has been shown to be correlated with the multidecadal variability of sea surface temperatures (SST) (Zhang et al., 2019) (which suggests that a large part of this variability in the temperature-based proxies is due to changes in the AMOC). It is not surprising, therefore, that for the larger part of the last millennium, the multidecadal variabilities in the proxies differ.

According to Caesar et al. tracing the centennial and longer evolution of the AMOC is the strength of this multiproxy comparison. In order to test whether the reduction in strength of the AMOC that is present in all proxy records is significant, a change-point model is fitted to each time series and the means of the data before and after the comparison of the change points (methods). In the first approximation, assuming only a single change point, the model found a significant reduction in the mean in all except 1 proxy record. In the different proxy series the timing of the change point varies (also due to the different time series lengths), though it can be sorted into 2 clusters: 1 change in the 2^nd half of the 19^th century and second change that occurred in 1960s. Each time series was divided into intervals of 50 years for the Cheng et al. 2017 data, given the length of the time series is only 64 years and 100 year intervals for the Spooner et al. (2020) data given the coarse resolution of this time series) going backward from the present, and Caesar et al. estimated the means and uncertainty for each of the intervals, in order to test the significance of the differences between the different time periods.  The mean of any 50- (30-, 100-) year interval was assumed to be significantly lower when its uncertainty range des not overlap with the uncertainty range of mean of any other interval. It is shown by the results of this study that in 9 of the 11 proxy series, the most recent 50- (30-, 100- ) year mean value was significantly lower than any other before. Additionally, it is suggested by the high resolution proxies that there was a decline in the AMOC within the most recent interval.

Together, it is consistently shown by this data that the modern slowdown of the AMOC is unprecedented in more than 1,000 years. Improved understanding of this slowdown is required urgently. The resolution of which components and pathways of the AMOC have altered, and why, is the next step, and it requires a community effort that combines observational, modelling and palaeoclimatological approaches.

Caesar, L., et al. (2021). "Current Atlantic Meridional Overturning Circulation weakest in last millennium." Nature Geoscience 14(3): 118-120.