New data shows 10 year decline in the Atlantic Conveyor

Laurence Hecht writes:

This review in the 19 Jun 2015 issue of Science  reported solid observational evidence (not proxies and modeling à la Rahmstorf and Mann) of a 10-year trend of decline in the Atlantic Conveyor. It’s not long enough to prove anything, but it’s solid data based on a string of tethered buoys stretching across the Atlantic from the coasts of Florida to Africa at 26.5ºN.

I found the whole 7-page report most enlightening. Especially interesting was the correlation of the huge 30% reduction in the AMOC flow rate with the record drop in the Arctic Oscillation in the winter of 2010. That was the Snomageddon that  drove every northeastern state into record or near-record cold temperatures for the month of February.
And now there’s new data in from another array across the North Atlantic,  called OSNAP, which also shows a declining AMOC.


With the solar decline underway (perhaps a new grand minimum) and a shutdown in the AMOC, we may not need to be worrying much longer about the blind ideologues who see anthropogenic CO2 behind every raindrop. But we shouldn’t ignore the data from these new strings of buoys.

It is important to note that 10 years worth of data really isn’t long enough to make any long term prediction about the state of the AMOC. Other recent studies have suggested that there is no long term declining trend:

New NASA measurements of the Atlantic Meridional Overturning Circulation, part of the global ocean conveyor belt that helps regulate climate around the North Atlantic, show no significant slowing over the past 15 years. The data suggest the circulation may have even sped up slightly in the recent past.
0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
Tom Halla
September 6, 2016 8:51 am

It looks like a decline, but there is so much noise in the system, it it hard to tell if it means anything. Anyone calculate the standard deviation of the variation in temperature?

Peter Sable
Reply to  Tom Halla
September 6, 2016 10:36 am

Standard deviation doesn’t mean anything when unless it’s a gaussian time-independent series. Which it is not. There’s likely periods with energy above the meager 10 year window. We’ll know something useful when we’ve seen two 75 year cycles…

Peter Sable
Reply to  Peter Sable
September 6, 2016 10:37 am

ugh I meant “time independent data set”. std deviations on time series are nearly useless.

Reply to  Peter Sable
September 6, 2016 11:35 am

Peter , when you say “which it is not” , should one infer that you , yourself, or with others have already looked at the distribution either side of the mean or mode and made a note of its assymmetry and shape?
I think that Tom Halla was interested in getting some idea of the width of that “deviation”, and I am sure that his curiosity is shared by several readers .

Reply to  Peter Sable
September 6, 2016 9:01 pm

Mike, it isn’t realistic to assume the distribution, certainly anyone can though if they want, then put the series into excel or some other package. I think Peter’s point (which was a good one) is just that unless the distribution is known, the variance is meaningless (literally) :).

george e. smith
Reply to  Tom Halla
September 6, 2016 6:55 pm

Well ten years is almost exactly on the button about one third of a climate observation period.
So give it another twenty years or so, and we maybe will have some idea if something is going on.
Of course, since those buoy strings haven’t been there very long, we don’t have any damn idea what was going on before. We don’t have any buoy data from a million years ago, or even from 5,000 years ago.
It is a bit difficult reading that graph above as someone has scribbled a distracting red line across the plot of the data.
That is NOT noise in the system.
That IS the data !
But yes it is interesting.

george e. smith
Reply to  Tom Halla
September 6, 2016 7:03 pm

I don’t think that is a plot of variation in Temperature. I think it is water flow; is it not ??
What would a calculation of the standard deviation in Temperature tell you other than tell you what the standard deviation in the Temperature WAS for that data set ??
What do YOU think that earth’s climate will do in response to your calculation of the standard deviation in the Temperature for that period of data ??
Who cares ??

george e. smith
Reply to  george e. smith
September 6, 2016 7:18 pm

Just as an aside: If that red scribble actually represented some actual real physical system response to the original black line data; which is to say the physical planet is somehow low pass filtering the actual real data, then that physical system response, would actually be DELAYED in time by some sizable fraction of the apparent time constant of the system. Low pass filters with zero time delay are not physically realizable.
I see NO apparent time delay between the black data and the red scribble, so I conclude that the red scribble does NOT actually represent anything that is real.
So it is simply real fiction.
Numerical Origami !

Tom Halla
Reply to  george e. smith
September 6, 2016 7:42 pm

It was that the decline in temperature was so much less than the extreme variation in the graph. I know very well that the probability calculations from a standard deviation only apply to a Gaussian distribution, but I was trying to get a measure of how much variation there was in standard deviations, which is applicable as a measure of variation.

Billy Liar
Reply to  george e. smith
September 7, 2016 5:41 am

George, yes it’s water flow in Sverdrups:

September 6, 2016 8:52 am

The other ‘recent studies’ are from 3/25/2010, which may have data from as late as 2009, half the ten year period looked at here.

September 6, 2016 8:56 am

if we had more data I am sure that it would show a general decline since 1971
and it will start to creep again from 2014
[as per the magnitude of the solar polar magnetic field strengths]

Reply to  HenryP
September 6, 2016 8:58 am

was there perhaps a specific reason not to show the measured data after 2014?

Peter Sable
Reply to  HenryP
September 6, 2016 10:38 am

Data collected are analyzed to derive transport estimates of the Deep Western Boundary Current over the continental slope southeast of New England. The moored array time series, started in 2004 and completed in 2014, was maintained by the Woods Hole Oceanographic Institution (USA),


Peter Sable
Reply to  HenryP
September 6, 2016 10:41 am

BTW the shear insanity of measuring climate metrics for only 10 years and then ending the program boggles the mind. When you see something that stupid it is usually for reasons of politics.
I bet we’ll see the day when satellites that measure temperatures will no longer be replaced, ending that highly useful data set before the 100+ years needed to actually conclude anything.

Reply to  HenryP
September 6, 2016 11:59 am

I am sure the data after 2014 would show a slow increase
I don’t trust the satellite data too much either
nothing [except our atmosphere] can protect the sensors against degradation by the scorching sun

Joel O'Bryan
Reply to  HenryP
September 6, 2016 5:22 pm


“• It is very likely that the Atlantic Meridional Overturning Circulation (AMOC) will weaken over the 21st century. Best estimates and ranges18 for the reduction are 11% (1 to 24%) in RCP2.6 and 34% (12 to 54%) in RCP8.5. It is likely that there will be some decline in the AMOC by about 2050, but there may be some decades when the AMOC increases due to large natural internal variability. {11.3, 12.4}
• It is very unlikely that the AMOC will undergo an abrupt transition or collapse in the 21st century for the scenarios considered. There is low con dence in assessing the evolution of the AMOC beyond the 21st century because of the limited number of analyses and equivocal results. However, a collapse beyond the 21st century for large sustained warming cannot be excluded. {12.5}”

2009 looks like an abrupt transition.

Joel O’Bryan
Reply to  HenryP
September 6, 2016 5:49 pm

The IPCC statements on AMOC make no sense in a warming world they forsee.
From the cyclical seasonal data, it is clear the AMOC speeds up in around NH summer solstice. The AMOC abruptly slows down in the NH winter solstice. So simply from that 1st order behavior standpoint, a warming world would suggest AMOC acclerates and thus speeds up the heat transfer circulation.
What am I mising? Why do warmunists believe it will slow down?

Reply to  HenryP
September 6, 2016 5:56 pm

All the Warmunistas know is that Man (especially men) is bad and Big Oil worse. That is all they know on earth and all they need to know.
Don’t confuse their simple, blind faith with facts.

Paul Penrose
Reply to  HenryP
September 6, 2016 12:06 pm

Yeah, and if my grandma had wheels, she’d be a wagon.

Reply to  Paul Penrose
September 6, 2016 12:25 pm


Reply to  Paul Penrose
September 6, 2016 1:49 pm

Well, not necessarily. If she only had two wheels she’d be a hand, ox, dog or donkey cart. Now if you were to specify that she had four wheels that would go far toward placing her in the wagon category. Grandmother’s superstructure would also influence her category. For example if her superstructure involved a permanent enclosure of her cargo space she could be a stagecoach or a hansom cab. And I think because we’re dealing with science here greater precision in her specifications is required. That’s just my peer review anyway.

Paul Penrose
Reply to  Paul Penrose
September 6, 2016 4:18 pm

That’s a dismissive way of saying “pure speculation”. You are so sure that “if we had more data” your pet theory would be confirmed. I’ve heard that a lot in observational sciences such as climate, solar studies, etc., but it usually doesn’t take long for nature to prove them wrong.

Reply to  Paul Penrose
September 6, 2016 7:57 pm

Or, as one would say in England, “if my aunt had b0llocks she would be my uncle”.

Frederik Michiels
September 6, 2016 9:03 am

something has to “drive the AMO” and i am pretty sure AMOc and AMO are pretty closely related.
I wouldn’t be surprised to see the conveyor “slow down” to create the negative state of the AMO and a “speed up” to make it positive.
note that this is what i call a “logical guess” i desperately tried to look for some science reportsd that do link the AMO with the AMOC. but untill now no success

Pop Piasa
Reply to  Frederik Michiels
September 6, 2016 5:30 pm

That seems logical. but which is cause and which is effect?

Jeff Alberts
Reply to  Pop Piasa
September 6, 2016 6:33 pm

Anthropogenic CO2 is always the cause, everything else are the effects – including toenail fungus, Goiter, and Herpes.

Reply to  Pop Piasa
September 8, 2016 3:41 pm

you left out baldness, that hits men more than women due to men being more evil…

george e. smith
Reply to  Pop Piasa
September 12, 2016 10:56 am

Grass doesn’t grow on busy streets.

Reply to  Frederik Michiels
September 6, 2016 7:02 pm

“I wouldn’t be surprised to see the conveyor “slow down” to create the negative state of the AMO and a “speed up” to make it positive.”
Negative NAO/AO drives a slow AMOC, which drives a warm AMO and Arctic.

Reply to  ulriclyons
September 6, 2016 7:04 pm

Think about it, if the overturning slows, the warm gulf stream has to still flow somewhere, so it spills into the N Atlantic and Arctic instead of overturning.

Reply to  ulriclyons
September 10, 2016 6:22 am

Hi Ulric – lets talk about this over coffee soon….I want to get my head around what really happens if the AMOC slows right down… then fails to REMOVE heat from the surface to depth in the downwelling phase, and you are right, that heat then gets dissipated somewhere – hence the Arctic warms as it has just done (a sign of heat LEAVING the planet!!!)….but at the same time there is less northward transport of surface waters containing heat, and hence overall the northern Atlantic starts to cool – which it has done in the last six years, and subsequently less heat is transported to Western Europe. and also into the Arctic, which should now begin to cool. I suspect the shift in the AMOC – which looks here like a step-change in 2010, follows a shift in winds (ie a shift in the tracking of the northern jetstream). This shift could be gradual – giving western Europe first mild and wet years, to be followed by cold and dry if the shift progresses. Anyway – much to discuss!

Reply to  ulriclyons
September 10, 2016 9:11 am

“that heat then gets dissipated somewhere – hence the Arctic warms as it has just done (a sign of heat LEAVING the planet!!!)…but at the same time there is less northward transport of surface waters containing heat”
That is a contradiction, there has to be increased warm flow to the North Atlantic and Arctic ocean for them to warm.
“and hence overall the northern Atlantic starts to cool – which it has done in the last six years”
Cooled marginally around this last sunspot maximum, which is the typical pattern during a warm AMO, because the solar wind is stronger around the sunspot maxima and generally weaker in between. Around the last two sunspot maxima are the only positive NAO/AO regimes since 1995. Predictably the AMO and Arctic will experience a strong warming phase again through to around the next sunspot maximum, which is already underway as the N Atlantic is very warm currently. image

Reply to  ulriclyons
September 10, 2016 9:12 am
September 6, 2016 9:07 am

The AMOC is doing what it always does.
The salinity-downwelling positive feedback gives the AMOC the nonlinear instability that causes it to abruptly switch on and off every few millenia.
Thus the frequent spikes in the Greenland temperature record compared to much smoother more gradual changes in the Antarctic where there is no such feedback.comment image
It is known from the study of the classic nonlinear pattern formation systems such as platinum-catalysed oxidation of CO, that ramping up feedback causes intermittent chaotic regimes leading to oscillation.
BTW what is the AMOC’s salinity-downwelling positive feedback? It goes like this:
Warm high-saline water from gulf stream is transported to north-east Atlantic – in Norwegian Sea above-average salinity water gets cooled to near freezing – cold plus more saline means dense so this water sinks all the way to the bottom and flows south – Southward bottom water flow impels surface flow in the other direction, reinforcing the gulf stream – back to beginning and replay …
In warmist linear-world positive feedbacks lead to run-away dystopian apocalypse and a path to political power.
But in the real nonlinear-chaotic climate system positive feedbacks are self limiting just leads to oscillation.
In the same way that the Bjerknes feedback (positive) is the driver of the intermittent chaotic oscillations in the Pacific that are el Nino and La Nina.
So what self-limits the salinity feedback of the AMOC?
It is freshwater from Greenland melt which eventually interferes with the Norwegian Sea downwelling and stops the gulf stream. This leads to big-time climate cooling.
So warmistas who gleefully point to evidence of Greenland melt and freshwater runoff should be careful what they wish for.

Reply to  ptolemy2
September 6, 2016 9:58 am

But what about the effect of the N. Equatorial Current? I thought that the NE Trades pushed equatorial water into the Caribbean and then this water was squirted out into the Atlantic between Cuba and Florida, becoming the Gulf Stream.

Reply to  Oldseadog
September 6, 2016 1:35 pm

You are right, the gulf stream as far as I understand starts life south of the equator. My all time favourite phrase in climae science and oceanography – “heat piracy” – is where it starts. A current crosses the equator south to north – cant remember its name. This turns into the Caribbean current which indeed then feeds the Caribbean with equatorial warm water. Then it circles the Caribbean and as you say gets squirted north-east as the gulf stream

Charlie O.
Reply to  ptolemy2
September 6, 2016 10:01 am

I’m by no means an expert, but seems to me that, based on your BTW2, increasing temperatures in the arctic (i.e. global warming) will inadvertently cause cold temperatures and increased snowfall in the UK and Europe? Just askin’.

Reply to  Charlie O.
September 6, 2016 3:23 pm

It is possible that a Greenland melt pulse could choke off Norwegian sea deep water formation, so yes, warming could lead to cooling. But many natural oscillations operate in the NH – as examined by Wyatt and Curry in their “stadium wave” hypothesis:
So it’s not a simple oscillator in isolation (gulf stream on-off switching), but one of a network.
William McClenney has posted here a number of times evidence that several interglacials have ended with a warming excursion not too dissimilar to our current one. AMOC cutoff is the most frequent cause of abrupt NH cooling episodes during the lifetime so far of the Atlantic Ocean.

Reply to  ptolemy2
September 6, 2016 6:03 pm

That’s a neat graphic. Couldn’t find it via the link because the paper is behind a paywall. Please confirm the citation for the graph. (NGRIP_NEEM_EDC_Global_135kya_png)

Reply to  Frederick Colbourne
September 6, 2016 8:51 pm

The graphic came from a Bill Illis post, I don’t have the reference. He may have made it himself from original data.

george e. smith
Reply to  ptolemy2
September 6, 2016 7:25 pm

Well let us suppose that the oceans contained NO dissolved salts of any kind. Just pure H2O.
So all of that water is just going to sit there and go nowhere but look silly.
Nonsense !
So long as the earth continues to rotate from West to East about once in 24 Earth hours, those equator to polar flows will just continue unabated.
What could change them; except a sizable change in the viscosity of H2O ??

Reply to  george e. smith
September 6, 2016 8:58 pm

Large lakes have their own complex hydrodynamics, so of course it’s not just salt. Fresh water has its own idiosyncrasy of having maximum density at 4 C and decreasing density at lower temperatures down to zero.

george e. smith
Reply to  george e. smith
September 7, 2016 9:06 pm

Well yes ptolemy2, you are of course correct about the fresh water maximum density. I was just going to ignore that in my model, whose intent was simply to point out that those currents are driven by earth’s rotation, and water viscosity long with the resultant tidal bulge and its KINETIC energy, which causes water to pile up above the hydrostatic equilibrium level when it runs into east coast shores, and all that momentum is converted into increased water height, until the pileup comes to a halt, and the now stationary water must escape from the equatorial region, so north and south are the only places to go.
So it has nothing to do with the salinity. And all of the oceans are sufficiently saline, that sea water increases in density all the way down to its freezing point, so it must sink as it travels towards the poles and cools, regardless of salinity or irregardless as the case may be.

george e. smith
Reply to  george e. smith
September 12, 2016 11:16 am

Whether you are talking about tidal bulges or simply the viscous drag between the solid earth and the liquid earth, and also the gaseous earth, there is always a net lag between the solid earth rotation and the liquid/gaseous components.
As a result, a Western shore line (East Coast) appears to be moving easterly relative to the oceans and the atmosphere, so in the rotating frame of reference of the solid earth, the oceans and atmosphere have a net westerly momentum.
That of course gets blocked eventually when the water/air runs into the coast line, and can’t go any further; well at least the water can’t. So that kinetic energy is converted into potential energy; meaning the sea level rises to a greater height than the hydrostatic equilibrium level, in a short term transient pulse. once that additional water is brought to a stop in the solid earth rotation frame, then of course that extra height buildup has to collapse basically at the speed at which water waves can propagate, and since it can’t go any further west, and can’t back up and go east, as more water excess is arriving, then it has to divide up and go north or south toward the polar regions.
So it doesn’t have anything to do with salinity or ocean pH or anything else besides the rotation of the earth.
Don’t even have to enlist the help of Coriolis, although that would certainly dictate some part of the motion.
So stopping the gulf stream would require stopping the earth from rotating.
Well I have always wanted to reverse the rotation so we could have a warm current going up the California coast, and have Tuna and mahi mahi and other decent fishes to fish for instead of those crummy salmon.

The other Phil
September 6, 2016 9:43 am

If anyone happens to look at the schematic and sees the explanation of a Sverdrup
(1 Sv = 106 m3 s–1).
note that it contains a typo. It isn’t 106 cubic meters per second, it is one million. It should be 10^6 not 106.
(Doesn’t affect the conclusions.)

Charles Higley
September 6, 2016 9:59 am

It would make sense for the flow to slow down as colder water is more viscous. Sediment studies in the passage between Florida and Cuba have suggested just this effect. Colder = slower, warmer = faster flow. This indeed the opposite of the predictions of the warmists, in which they claim that the AMOC should slow down with global warming—they like everything to be bad, but their prediction is actually a negative feedback as they state it.

September 6, 2016 10:10 am

If I read the graph correctly, the flow actually reversed twice.
I wonder if we have an actual slowing or is it that the current deviated from its path and we’re seeing the result of how the bouys are positioned.

Bloke down the pub
September 6, 2016 10:14 am

So it’s worse than we thought?

Reply to  Bloke down the pub
September 6, 2016 10:40 am

it is bad
it is bad as it gets

Peter Sable
September 6, 2016 10:34 am

Get your raw data right here, skip the paywalls:

September 6, 2016 11:02 am

another graph you can only see….if they blow the scale up really big

September 6, 2016 11:50 am

AMO = atlantic multidecadal oscillation.
You can’t tell much from a decade of data with something that oscillates over multidecadal periods.
Could be going up, down, or sideways.

Joel O’Bryan
Reply to  wallensworth
September 6, 2016 5:34 pm

AMOC = Atlantic Meridonal Overturning Circulation.
You suggest an equivalence between AMO phase and AMOC. Is there?

September 6, 2016 12:27 pm

“….a string of tethered buoys.” I’d be interested in more information, if any reader knows about these. How many, spaced at what intervals? How deep is the water and how are they tethered? Are they still viable after several years? Are they a navigational hazard? Or are they liable to be damaged by large ships? Not much to do with the subject under discussion.- I’m just curious!

Reply to  mikelowe2013
September 6, 2016 2:59 pm

@mikelowe , I am curious as well but your great questions are way to logical for the warmists, a ten year study is to me not enough basis to go on and ending it is just plain silly. If there was a trend that could have a major impact on climate in Europe this is one that should be continued.

Reply to  Toby Smit
September 7, 2016 2:12 am

I seem to remember years ago reading about or being told about a series of NATO buoys tethered at or near the N. Atlantic sea floor to give information about passing submarines. I suppose the USSR might have had a similar installation. No danger to surface vessels.
Maybe this is the source of the data.

Reply to  mikelowe2013
September 7, 2016 5:09 am

“….a string of tethered buoys.” I’d be interested in more information, if any reader knows about these. How many, spaced at what intervals? How deep is the water and how are they tethered?
PERFECT QUESTIONS! And ones that are imperative to ask. Where are the buoys? In what part of the water column? Are they doubled, tripled or just singular? How far are they spaced? Are they moved as the currents move from season to season?
These are all questions that need to be answered first before any kind of “trend” or what have you can be drawn.
The ocean water column is not the same….it is a dynamic environment that we know very, very, very little about and measuring surface temp/salinity doesn’t do squat for what could be occurring 10M below the surface. While we understand it is all connected—exactly how is exceedingly complex and we are only at the beginning of our understanding.
You can’t just string a bunch of buoys along on the surface and figure it all out. Measurements need to be taken at regular intervals in the water column over a distance using the similar bottom features. And sets need to be placed over different bottom features as well. The bottom of the ocean is not a “constant” and alot of it’s topography has impact upon the currents–both deep, middle and surface.
It’s like the entire middle water of the ocean is completely ignored in these studies and it drives me insane!

Mike Borgelt
September 6, 2016 3:09 pm

Looks to me like a 4 year decline from 2006 to 2010, then steady.

September 6, 2016 3:31 pm

The Atlantic Conveyor is also a freighter carrying helicopters that was sunk during the Falklands War. That was at the other end of the Atlantic. This Atlantic Conveyor’s downtrend is 5 Sverdrups per decade. At that rate it will come to stop in 30 years.

September 6, 2016 7:15 pm

“With the solar decline underway (perhaps a new grand minimum) and a shutdown in the AMOC, we may not need to be worrying much longer about the blind ideologues who see anthropogenic CO2 behind every raindrop.”
Well if CO2 really had the clout that they reckon, we shouldn’t have seen the increase in negative NAO/AO since the mid 1990’s that has slowed the AMOC. And the reduced overturning is what has fed the warming of the AMO and Arctic since then. The Gulf Stream appears to speed up during low MOC events, maybe less resistance than overturning, though it doesn’t show a longer term trend:

Tom in Florida
September 6, 2016 7:27 pm

I wonder what the Atlantic Conveyor was doing the first 54 years of my life?

September 6, 2016 8:13 pm

Maybe the CAGW horrors depicted in the film The Day After Tomorrow will come true, after all.
I think it could be time for us all to run for the hills, shrieking and gibbering..!

September 7, 2016 1:12 am

Interesting that the decline coincides with the 10 year decline in the AMO as shown on Figure 6 at Prior to that, the AMO displayed a 32 year RISE.

Reply to  Dan Pangburn
September 7, 2016 6:56 am

This correlation between Barents Sea temperatures and the AMO:comment image
would suggest that the AMO could in fact be nothing other than a cyclical oscillation in the strength of the gulf stream – only this could bring changes in Barents sea temperatures down to 150m of up to 4 degrees C. Here is the reference for that figure:
Levitus, S., G. Matishov, D. Seidov, and I. Smolyar (2009), Barents Sea multidecadal variability, Geophys. Res. Lett., 36, L19604, doi:10.1029/2009GL039847.
It was from this previous WUWT post:

Reply to  ptolemy2
September 7, 2016 1:24 pm

ptol – Thanks for that.
That fits with what I discovered that works ( Blog shows 98% match between calculated and measured 1895-2015): Which includes simple approximation of net effect on global temperature of ALL ocean cycles (64 yr period). I have yet to see a rational explanation of why oceans cycle with such a long period.
I kind of perceive Arctic ice as melting from below due to water temperature competing with freezing from air temperature above.

Verified by MonsterInsights