RealClimate logo

What the 2018 climate assessments say about the Gulf Stream System slowdown

Filed under: — stefan @ 28 January 2019

Last year, twenty thousand peer reviewed studies on ‘climate change’ were published. No single person can keep track of all those – you’d have to read 55 papers every single day. (And, by the way, that huge mass of publications is why climate deniers will always find something to cherry-pick that suits their agenda.) That is why climate assessments are so important, where a lot of scientists pool their expertise and discuss and assess and summarize the state of the art.

So let us have a quick look what last year’s climate assessments say about the much-discussed topic of whether the Atlantic Meridional Overturning Circulation (AMOC, a.k.a. Gulf Stream System) has already slowed down, as predicted by climate models in response to global warming.

First, there is the IPCC 1.5 °C report (SR15) prepared for the Paris Climate Agreement and published in September 2018. It doesn’t say all that much about the AMOC, given that it is not a full IPCC assessment, but it does say this:

It is more likely than not that the Atlantic Meridional Overturning Circulation (AMOC) has been weakening  in recent decades, given the detection of the cooling of surface waters in the north Atlantic and evidence that  the Gulf Stream has slowed by 30% since the late 1950s (Srokosz and Bryden, 2015; Caesar et al., 2018).  There is only limited evidence linking the current anomalously weak state of AMOC to anthropogenic warming (Caesar et al., 2018). It is very likely that the AMOC will weaken over the 21 st century. […]

Weakening of the Atlantic Meridional Overturning Circulation (AMOC) is projected to be highly disruptive to natural and human systems as the delivery of heat to higher latitudes via this current system is reduced.

[Note: those “30% since the late 1950s” are probably in error; they are not supported by either of the two references provided.]

Then, in November, the 4th US National Climate Assessment was published that had been two years in the making. It says:

The primary concern related to ocean circu­lation is the potential slowing of the Atlantic Ocean Meridional Overturning Circulation (AMOC). An AMOC slowdown would affect poleward heat transport, regional climate, sea level rise along the East Coast of the United States, and the overall response of the Earth’s climate system to human-induced change. […]

As the atmosphere warms, surface waters entering the North Atlantic may release less heat and become diluted by increased freshwater melt from Greenland and Northern Hemisphere glaciers. Both of these factors would slow the rate of sinking and weaken the entire AMOC.

Though observational data have been insuffi­cient to determine if a long-term slowdown in the AMOC began during the 20th century, one recent study quantifies a 15% weakening since the mid-20th century and another, a weakening over the last 150 years. Over the next few decades, however, it is very likely that the AMOC will weaken.

Finally, Future Earth (a global Earth science research programme) and the Earth League (a grouping of leading institutions and individuals in the Earth sciences) have issued a climate science update for the Katowice climate summit in early December, called 10 New Insights in Climate Science 2018. It says:

 A weakening of the Atlantic overturning circulation, often referred to as the Gulf Stream system, has been expected from model simulations. Recent studies confirm that it has slowed down by 15% since the middle of the 20th century and is at its weakest in over a thousand years. This is already having observed effects, such as extreme weather in Europe, and further weakening is expected to strongly affect European weather as well as exacerbating sea-level rise at the east coast of North America.

In December also a new study – too late to be included in the assessments – was published by Thibodeau et al in Geophysical Research Letters, which further supports an unprecedented AMOC weakening during the past decades. The authors write:

In this study, we used geochemical evidence to highlight a slowdown in the North Atlantic Ocean circulation over the last century. This change appears to be unique over the last 1,500 years and could be related to global warming and freshwater input from ice sheet melt.

My view

Our regular readers know that one of my topics of interest is the stability of the Gulf Stream System – I’ve worked on this on and off for over 25 years, ever since finishing my PhD in physical oceanography. So let me add my own comments on the findings cited above.

First of all, while we don’t have regular direct measurements of the AMOC going back throughout the 20th Century, indirect evidence for an AMOC slowdown is not new. Dima and Lohmann already concluded in 2010 that “the conveyor has been slowing down over the last seven decades” (where ‘conveyor’ refers to the AMOC).

Strangely, this finding was not discussed at all in the fifth IPCC report published in 2013. Therefore, the IPCC now finding that an ongoing slowdown is “more likely than not” is progress, yet still a very cautious statement. Likewise the statement about the “limited evidence” for the slowdown being human-caused is also very cautious. Why do I find this overly cautious?

The main points there are that an AMOC slowdown leads to a particular fingerprint pattern in sea surface temperature change – which is basically what Dima and Lohmann already identified, and this pattern is predicted by high-resolution climate models in response to rising greenhouse gases, and it is also found in the observations. There is no known alternative explanation for what might cause this fingerprint. That fingerprint is not subtle: it is so strong that the subpolar Atlantic is the world’s only region which has resisted global warming over the past hundred years and even has cooled down, reaching record low temperatures in 2015 when the globe as a whole was record-hot.

Although the AMOC slowdown fingerprint is most clearly seen in long-term sea surface temperature trends, it is also apparent in the 2018 temperature anomaly, despite a single year including a lot of short-term variability noise. No place on Earth had a larger cold anomaly than the subpolar Atlantic. Image: Berkeley Earth project.

In fact, the strength of this pattern and the conclusion that it corresponds to a 15% AMOC slowdown just matches the median slowdown found in the historic climate runs of the CMIP5 climate models – in other words, it is exactly what the models predict as a response to human-caused climate change. The physical mechanism is understood – how warming and ice melt weakens the AMOC (and that these factors are human-caused), and how an AMOC weakening causes the observed surface temperature fingerprint. In addition, there are several independent data sets that show this slowdown to be unprecedented for at least a millennium.

In IPCC jargon, personally I would therefore give the statement that the AMOC has slowed down since the early-mid 20th , and that this is at least partly human-caused, a “very likely” rating.


If you doubt that the AMOC has weakened, read this

Stronger evidence for a weaker Atlantic overturning circulation

The underestimated danger of a breakdown of the Gulf Stream System

AMOC slowdown: Connecting the dots

What’s going on in the North Atlantic?

20 Responses to “What the 2018 climate assessments say about the Gulf Stream System slowdown”

  1. 1
    Geoff Beacon says:

    Corrected comment:

    Strangely, this finding was not discussed at all in the fifth IPCC report published in 2013.

    I take this as a criticism of the IPCC

    And your If you doubt that the AMOC has weakened, read this shows further frustration.

    Far too often we find the scientific consensus, as filtered through the official bodies and MSM, underestimates the speed of change. The SR15 report may be underestimating in another way. Did it use climate models that did not take enough feedback mechanisms into account?

    Lowe and Bernie’s The impact of Earth system feedbacks on carbon budgets and climate. Their modelling incorporated more in the way of feedback and found much smaller remaining carbon budgets for 1.5°C or 2°C than SR15.

    Richard Betts replied on twitter saying

    Personally I find the Lowe & Bernie estimate to be credible (full disclosure: they’re close colleagues of mine).

    The Lowe & Bernie paper doesn’t seem to have been cited much. How seriously should we take its worrying results?

    P.S.The twitter thread.

  2. 2
    Mike Risk says:

    Thanks for that…one searches in vain for good news these days.

    Here are some more, to add to that yearly 20,000 (another depressing number).

    There are other reports of weakening of the AMOC, derived from geochemical analysis of the climate records encoded in the skeletons of deepwater coelenterates (corals and gorgonians). In 1997, Smith et al. (Nature 386: 818) presented data from deepwater corals, suggesting that the Gulf Stream had shut down within four years during the onset of the Younger Dryas. They suggested this was triggered by meltwater-sound familiar? In 2006, Sherwood et al analysed deepwater gorgonians and reported an anomalous weakening of the Labrador Current off Nova Scotia (only as a GAC Abstract at this time).

  3. 3
    Geoff Beacon says:

    Short-lived climate pollutants(SLCPs),such as methane, have effects on ocean heat content for longer periods than their effects on global surface temperatures.

    Because of these shorter term effect on surface temperatures, some have argued that their danger is overstated. For example, since methane has a lifetime of about 10 years it does not contribute much to long-term surface temperature rise in the longer term.

    (See Allen et al. A solution to the misrepresentations of CO2-equivalent emissions of short-lived climate pollutants under ambitious mitigation.)

    Does this miss the effects of SLCPs on the AMOC because during the lifetime of SLCPs the oceans have warmed?

  4. 4
  5. 5
    t marvell says:

    What is the ETA for AMOC stoppage?

  6. 6
    prokaryotes says:

    In addition to the effects caused by the rising bottom water
    temperature due to global warming, there are other processes
    and factors in the ocean that may influence the stability and
    CH4 emission of sediment gas hydrates. The current climate
    change may eventually alter the ocean circulation pattern and
    the intensity of some currents. For example, changes in the
    Gulf Stream intermediate water temperature could cause a
    rapid destabilization of gas hydrates along a broad path in the
    North American margin seas

  7. 7
    CM says:

    Thanks for this. Have you mentioned to the IPCC the need for an erratum on the 30%? While I think their conservatism on the likelihood/evidence statements is more of a concern than this slip of the pen, it /is/ a pretty glaring slip.

  8. 8
    William Rees says:

    Is the IPCC inherently conservative? Does the IPCC underestimate?
    Try this:

  9. 9
    Al Bundy says:

    Yeah, if water doesn’t get dense enough to hit bottom, it will still sink. Random geographic facts will steer where it goes. Some clatrate beds might get a nice warm flowing bath. Depends, perhaps, on how far south the freshwater lens gets. Evaporation releases heat to the atmosphere. Every joule that doesn’t warm Scotland et al will instead heat some non-surface level of the ocean.

  10. 10
    JCH says:

    In recent years, along with the discussions about a weakening AMOC, there has also been speculation about the possible imminent onset of the negative phase of the AMO. So far, it hasn’t happened.

    Is the continuance of a positive AMO incompatible with the weakening AMOC?

  11. 11
    BojanD says:

    My apologies in advance if my question might seem like laziness on my part but how would one go about establishing figure 55 as the daily rate of papers published that are related to climate change?

  12. 12
    Mitch says:

    A relatively strong warming of shelf water temperature will not cause catastrophic clathrate decomposition. The change in temperature at the sea floor is only a few degrees C, and this temperature change must diffuse through about 200 m of sediment above the clathrate. That requires centuries, which is why we can estimate earth surface temperatures for the last 500 years from boreholes.

  13. 13
    Matthew R Marler says:

    Thank you stefan for this nice focused essay.

    How much, if you can tell, of the AMOC slowdown has resulted from the early 20th century warming, and how much from the late 20th century warming?

  14. 14
    stefan says:

    Go to Web of Science (subscription required), enter “climate change” (with the quotation marks) as search term, find 2017 under publication years and see that 20,191 studies are listed. Divide by 365 days. For 2018 the number is even 61 per day now, given a total of 22,329 papers. (I used the 2017 total because I started to write this piece before Christmas when 2018 wasn’t over yet.)
    As an aside, at Web of Science you can also see that the Caesar et al. paper mentioned above (with the 15% AMOC decline finding) is marked as “hot paper”, meaning “in the top 0.1% of papers in the academic field of Geosciences” according to how often it is cited.

  15. 15
    Matt Skaggs says:

    Unless there is some factor specifically related to CO2, this phenomenon should have also occurred during the Holocene Thermal Maximum, right? Given how warm it was then elsewhere, there should be proxy evidence of extreme cold in the regions currently warmed by the Gulf Stream. Anyone looking for that in those 20K papers?

  16. 16
    Keith Woollard says:

    Are you sure you got the final version of SR15? I have just looked at the one you link to, dated prior to your post, and it makes no mention of the 30%. It also references three papers, not two. Here is a direct copy/paste

    It is more likely than not that the Atlantic Meridional Overturning
    Circulation (AMOC) has been weakening in recent decades, given
    the detection of the cooling of surface waters in the North Atlantic
    and evidence that the Gulf Stream has slowed since the late 1950s
    (Rahmstorf et al., 2015b; Srokosz and Bryden, 2015; Caesar et al.,
    2018). There is only limited evidence linking the current anomalously
    weak state of AMOC to anthropogenic warming (Caesar et al., 2018). It
    is very likely that the AMOC will weaken over the 21st century.

  17. 17
    Mr. Know It All says:

    Would summer temps in northern Europe be cooler if the Gulf Stream stopped pumping heat northward from the tropics? Wouldn’t that be a plus for northern Europe?

    And if the GS stopped pumping heat to the north Atlantic in winter, perhaps we might get more ice in winter, increasing reflection of the sun rays? Wouldn’t that be a plus for the planet?

  18. 18
    Al Bundy says:

    Mr FatalInaction,
    It has been found that obesity is harmful. It has also been found that cancer cures obesity. Therefore, shouldn’t we do everything possible to give everyone cancer? Wouldn’t that be good for our bodies?

    Dude, just because certain metrics improve via death and destruction does not mean that death and destruction are grand.

  19. 19
    SecularAnimist says:

    Related reading:

    An international research programme has uncovered data that could transform scientists’ understanding of the Atlantic Ocean current – a circulation pattern that plays a central role in determining weather across the world.

    The research, published in Science, challenges the long-held view that the strength of the “Atlantic Meridional Overturning Circulation” (AMOC) is primarily driven by processes in the Labrador Sea, which is in the north-west Atlantic.

    Instead, the project finds that – over a 21-month period – the strength of the AMOC was most linked to processes in waters between Greenland and Scotland, more than 1,000 miles away in the north-east Atlantic.


  20. 20
    John D. Wilson says:

    I would like to be the first to comment on this thread about the amazing statistic that 20,000 peer reviewed studies are published every year on climate change. Considering the complexity of those studies, let’s assume that it takes an average of one FTE to manage 40 research projects and papers, including supervision, training, fundraising, etc. That means that the secret organization that manages the Greatest Hoax of All Time (TM) must have around 500 employees dedicated to this task alone. That information should help narrow the search for the secret organization.