Not salty enough, not strong enough

The implications of recent AMOC news for the US East Coast

News stories about the Gulf Stream and the AMOC (Atlantic Meridional Overturning Current) have probably whizzed past you over the last few weeks. As with a lot of climate information, the details of these stories may have been absorbed into an overall fug of dread. But there's something meaningful here that you need to understand.

The key messages of recent weeks are: 

(1) that scientists now have enhanced physics-based understanding of what processes will cause the North Atlantic current to suddenly slow down,

(2) when they look at observed (actual) data about saltiness and strength of the ocean at key places on the globe, it looks as if these processes are actually underway, and

(3) although they don't know exactly when that slowing will happen, when they compare these actual real-life measurements to the computerized climate prediction models used by the IPCC it looks as if we are seriously underestimating the risks—meaning that the tipping point may be much closer than we had thought. 

I'll explain what underlies these three points in a moment. But this news matters enormously to the U.S. East Coast. As the AMOC slows down, it will do two things. It will stop moving warm water up our coast to the north. It will also stop deflecting water to the right, or east, out into the Atlantic. Three feet of water will pile up along our coast as a result. Those three feet will be on top of already ferociously accelerating global sea level rise. Along the southeast coast, we'll see even more hurricanes, because the warmer waters left behind by the slowing AMOC will be more conducive to stronger storms. Again, we don't know exactly when this will happen, but the risk of it happening is much greater than we have been assuming.

Dennis Quaid told everyone this was going to happen in The Day After Tomorrow (2004).

Just kidding. Not really. Although the movie is sensational, the underlying science is well-accepted. 

You may wonder what drives the oceans to move around the world. Is it the rotation of the earth? Is it the wind? Henry Stommel and others figured out in the 1960s that water moves around the globe by way of a complex system of conveyor-belt-like currents driven by differences in salinity and heat. Those currents are one of the main ways heat gets distributed around the world, and they affect weather patterns everywhere. 

These currents have been pretty stable over the course of our civilization, because there's been a delicate balance of cold/warm water and fresh/salt water. When those differences encounter one another, things happen, and that interaction drives motion: Cold, salty water sinks in the North Atlantic due to its higher density, while warmer, fresher surface water exists elsewhere. This density difference drives a large-scale circulation where the deep water travels southward, eventually rising and warming near the surface before returning northward, carrying salt that then drives the system onward. The Gulf Stream is part of this system, taking massive amounts of warm water from the Gulf of Mexico and shuttling them north. Here's a picture of thermohaline circulation (heat and salt differences).

Scientists have been directly measuring the AMOC since 2004, using a boatload of sensors to track the strength, salinity, and temperature of currents. They've been able to distill these observations into something they call the FovS, which measures the saltiness of the water going into the Atlantic ocean right at its southern border at the 34thS parallel. Here's a picture of the 35thS parallel, just so you see where this border is.

When that FovS indicator is negative, they know that the whole system is driving less saltiness north than needed to keep the AMOC moving along at the rate we've been accustomed to for thousands of years.

I'm sure you see the link: The ocean circulation transports salinity and heat from the southern hemisphere to the northern hemisphere. If less-salty water is going north, it won't be present to turn cold and sink in the North Atlantic, or not in the quantities and with the vigor on which the system depends. The buzzword for all this is the "salt-advection feedback." The AMOC's delicate balance of water qualities, the feedback loop that keeps it moving along, is being destabilized by increasingly fresh water in places where water used to be salty, particularly in the North Atlantic.

Why is this happening? Well, amazing amounts of ice is melting at the poles. This happened before, before our civilization emerged, and caused a cataclysmic climate shift. This time, it's happening extremely quickly because of human activities. That melting ice is driving a lot of fresh water into the system. The initial weakening of the current up north (because it's less salty, less dense water is sinking up there) means salt quantities being brought up to the north are lower, which in turn further amplifies the weakening and makes it worse. This is a positive feedback loop—but the word "positive" from a human perspective, here, is bad.

Even though the oceans are vast, the recent study that caused all the news stories showed that a very sophisticated computationally-heavy simulation of an enormous amount of freshwater being dumped into the North Atlantic (at 80x the rate it is happening now) triggered an abrupt tipping point after a simulated 1750 years. The conveyor belt stopped. 

The study is important because it's the first time we have seen a simulated physics-based, reliable set of processes leading to the AMOC stopping. Before this publication, scientists could still quibble about whether this stoppage was possible. Now they can't.

But wait, there's more. Eighty times the rate of melt, you'll say. Almost 2000 simulated years of that. Why should we care about such an outlandish result?

We need to care because the study incorporated currently-accepted climate models. It simulated all those generations of change using assumptions built into those models. It turns out, however, that those climate models are based on inaccurate assumptions about the saltiness of the water flowing into those Atlantic Ocean borders—they're wrong about how salty the water is on the surface of the Indian Ocean, apparently, and they're wrong about how salty the water is over the Atlantic Subpolar Gyre region.

Scientists know the models are wrong because they've done actual measurements of the salinity there, using sensors. Here's the truly dramatic summary: "many CMIP6 models [widely-accepted models used globally, including by the ICPP] have large biases [are wrong in their assumptions] in surface freshwater fluxes [about the salinity of water on the surface] which lead to an AMOC with an Atlantic freshwater transport that is in disagreement with observations." Kneebone-connected-to-the-thighbone sort of thing: if you're wrong about the Indian Ocean and the Atlantic Subpolar Gyre, you've got the wrong inputs going into your models. Operatic stuff.

Dr. René van Westen, a scientist in Utrecht who is on top of all this, told an interviewer last month

that 

[B]ecause these models have these climate model biases, if we correct for them, we probably expect that the tipping point will be much more closer. So then we're not talking about 80 times [current ice melt], so maybe then only 10 times or five times. We cannot say anything about that because we are not sure how all these biases influence each other. 

Because it's a very complex system. You can imagine that the atmosphere is part of that, sea ice, land ice. So it is therefore a very urgent effort to the modeling community to reduce these biases and push the AMOC in the real observed regime in this kind of models. 

In other words, it might be that just five or 10 times the amount of melt we're seeing today, and there's already a lot of melting going on now, would get the AMOC to this tipping point. We should continue to simulate these processes, says van Westen, but we should do it based on real observations.

Enormous consequences will follow from the slowing and possible stopping of the AMOC for the East Coast (and of course for many other places, but let's be parochial for a moment). 

Because of the way the earth rotates, in the Northern Hemisphere water deflects to the right off the East Coast. To the east, out into the Atlantic. The wide, rushing Gulf Stream is accordingly also tilted, deflected, keeping water off the coast. As the AMOC slows down and stops moving that water north, it will also stop deflecting it to the right. Where will that water go? It will slop up on our coast, on Charleston, Norfolk, New York, and Boston. This process isn't adding any new water—it's happening because of the Gulf Stream's angle flattening. So the three feet of water this process dumps on us will be added to the sea level rise we're already facing. 

Look, we still don't know when this happens. It may be happening already—the AMOC has slowed by 15 percent since 1950. van Westen and his colleagues are not screaming. They're just doing good work. They just want the public to know this could happen, and that it's more likely than we used to think.