I'm not completely clear on what Western intensification means, but it looks like a pretty powerful (combination of) forces that presumably wouldn't be likely to change on a timescale of years (or even decades, centuries).

But I recall reading some time ago that global warming is already known to be causing a significant increase in the amount of fresh (as opposed to salt) water being introduced into the system at the European end. I assume that's "one-time-only" melting of glaciers and snow/ice in areas that were previously permafrost, but for all I know there could be much more simply arising from a (potential) massive increase in precipitation across north-western Europe.

I freely admit I don't know the relative importance of warm/cold salty/fresh water in the thermodynamics of the Gulf Stream system, which is probably relevant. But what I really want to know is...

Do we understand the Gulf Stream well enough to answer the question posed in my title?

  • $\begingroup$ Suddenly as in a matter of seconds? Can you please clarify what timescales you mean? $\endgroup$ Commented Jan 8, 2015 at 5:53
  • $\begingroup$ @Isopycnal: The timescale of "suddenly" is implied by my "wouldn't be likely to change on a timescale of years". At 60, I'm probably well into the second half of my life. If the Gulf Stream were to have stopped before me, I'd consider that relatively "sudden". (If I were still around to have an opinion, which obviously I wouldn't be! :) $\endgroup$ Commented Jan 8, 2015 at 13:10

2 Answers 2


There's a pretty good rundown of this issue over at RealClimate, and a wikipedia article dedicated to the topic.

A summary of the main points from there:

  • The Gulf Stream is part of the Meridional Overturning Current (global ocean conveyor), which is primarily wind-driven, and secondarily thermohaline-driven. The Gulf Stream itself is also primarily wind-driven.
  • There has been some reduction of the thermohaline drivers in the north Atlantic, due to increased temperatures, and influxes of fresh water, largely from melting of the Siberian permafrost. This part of the system does fluctuate (as do most parts), but there does appear to be a trend.
  • These changes are not enough threaten the Gulf Stream, and there hasn't really been any observations of an over-all slow down that couldn't be otherwise explained by natural variation.
  • It is likely that the Gulf Stream will slow down over the next few centuries (slow downs appear in some of the IPCC models out to 2100).
  • A complete shut-down is possible (it has happened before, notably during the Younger Dryas), but it is hard to predict, as causes are not very well understood. But it seems unlikely to happen any time soon.
  • 1
    $\begingroup$ That's a good summary of the RealClimate article. In principle I did actually know the Gulf Stream is primarily wind-driven. But over the decades since I was first taught that in school, I seem to have heard much more about meltwater/preciptitation (because they're changing faster?), so I've ended up being confused about what really counts. Anyway, I'm heartened by the last line in that link: While continued monitoring of this key climatic area is clearly warranted, the imminent chilling of the Europe is a ways off yet. $\endgroup$ Commented Nov 28, 2014 at 12:47
  • 1
    $\begingroup$ Well, that article is 9 years old, and much of the wikipedia page contents aren't much newer. There could be newer studies that have changed things, but I doubt they would have made enough difference that anyone should start worrying, or it'd be all over the news :) $\endgroup$
    – naught101
    Commented Nov 29, 2014 at 12:13
  • $\begingroup$ @naught101 don't forget that "the news" is not a reliable source about climate change. $\endgroup$
    – Gimelist
    Commented Dec 6, 2014 at 8:41
  • $\begingroup$ @Michael: no, but it is liable to exaggerate any mildly exciting or concerning scientific finding. If any papers had been released indicating greater likelihood of Gulf Stream shutdown, there would almost certainly have been a few catastrophe articles. $\endgroup$
    – naught101
    Commented Dec 9, 2014 at 2:13

In some capacity, differential heating is responsible for convective motion of the ocean due to buoyant forces. However, it is possible to show that the net (vertically integrated) horizontal transport in most of the ocean is solely due to the action of the wind stress on the water surface (Sverdrup, 1947). In other words, the pattern of major ocean currents is a direct mirror of the pattern of the wind stress.

The main reason the current is intensified in the western boundary is because of the $\beta$-effect, or the strengthening of Coriolis with latitude. This effect causes the slope of the ocean surface to be steeper on the western side because the equator easterlies are stronger than the mid-latitude westerlies and therefore surface water piles up in the western boundary. Due to conservation of potential vorticity the steeper slope results in faster geostrophic flow on that side of the gyre (i.e. western intensification).

Another cause for the westward intensification are the density gradients, namely the sign of the gradient of zonal density. For example, if the density in the polar regions suddenly became less than the density in the equator regions, the Western boundary current would be considerably weaker. Another argument explains the phenomenon of western intensification via the accumulation of energy due to westward propagating planetary Rossby waves.

If we assume that the current velocity in the Gulf stream is $U \sim 1$ m/s on average (its maximum may be twice that) then the momentum per unit volume of a fluid parcel will be $\rho U \sim 1000$ kg m$^{-2}$ s$^{-1}$. Typical estimates suggest the total mass transport across a cross section of the Gulf stream is on the order of $90 \cdot 10^6$ m$^{3}$ s$^{-1}$ of water. By comparison, the momentum of a fluid element in the jet stream in the atmosphere is one order of magnitude smaller than that in the Gulf stream, despite the fact that the wind speed is one hundred times larger than the water current speed.

Assume the gulf stream has a total volume $V = L \cdot W \cdot D$, where the length $L = 2500$ km, the width $W = 75$ km and the depth $D = 800$ m then, as a crude approximation, we can estimate the total force $F$ that must act on the fluid over (lets say) $2$ years to cause a sufficient change in the bulk fluid momentum $m \Delta v = \rho V \Delta v$ so that the Gulf Stream comes to a stop. From Newton's Second Law and plugging in some typical values:

$F = m \frac{\Delta v}{\Delta t} = 2.5 \cdot 10^9$N

Suppose the force is applied through a surface wind and suppose its action extends all the way to the bottom of the current. Then one would require a wind stress $\tau = 0.01$ Pa acting for at least two years, of course that number would need to be higher since in reality the wind stress accelerates the top layer of the water column, and its effect weakens with depth. Note that typical oceanic values of wind stresses are about $0.1$ Pa. This indicates that, purely from physical principles, stopping the Gulf stream within two years is realistic.

However, it is not possible without a complete reversal of well established wind patterns combined with a drastic reduction in the density near the poles compared to near the equator. The latter is why some models predict future slowdown of the Gulf Stream, as the anthropogenic heating of the earth causes the melting of polar ice, the ocean water becomes fresher and less dense. Fortunately, as @naught101 pointed out, changes of that magnitude are unlikely to take place in the near future.

Sverdrup, Harald Ulrich. "Wind-driven currents in a baroclinic ocean; with application to the equatorial currents of the eastern Pacific." Proceedings of the National Academy of Sciences of the United States of America 33.11 (1947): 318.

Pedlosky, Joseph. "Geophysical fluid dynamics." New York and Berlin, Springer-Verlag, 1982. 636 p. 1 (1982).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.