Climate change and earthquakes: It's complicated

In the wake of the Tohoku earthquake and tsunami last March, I started seeing a lot of headlines like this:

"Does climate change mean more tsunamis?"

"Did climate change cause the Japanese earthquake?"

In those stories, environmentalists and climate science deniers went head-to-head, with one side pointing out yet another unintended consequence of fossil fuel consumption, and the other side pointing and laughing at what it saw as patently ridiculous fear-mongering. Missing: The nuance. And you know how much I love the nuance.

This is a story that contains a whole lot of yesbut. Yes, it really does make sense that climate change could trigger earthquakes. But it's very, very unlikely that that effect is responsible for any of the monster quakes we've experienced recently. And behind that apparent contradiction lies some really, really interesting science.

Let's start with a quick overview of why scientists think climate change and earthquakes are connected.

On the surface, this does sound pretty insane. Climate change is about the greenhouse effect increasing the global average temperature. The impacts of climate change tend to be things that are linked, somehow, to weather and climate—droughts, storms, changing habitats, melting ice caps. Earthquakes, on the other hand, are about landmasses bumping up against one another. That's plate tectonics, not El Nino. But the basic theory actually does make a lot of sense. And it's really just a logical extrapolation of some well-established natural phenomena.

It begins with the forces that cause earthquakes. The surface of this planet, what we see, is actually the crust—just the crispy coating on a ball of nougat. The crust is broken up into large pieces and those pieces move over the surface of the gooey mass beneath. At the borders, the pieces of the crust are riddled with faults. These are places where the crust has broken and different pieces are moving in different directions—away from each other, towards each other, or slipping past one another.

These faults can get stuck on one another and, over time, build up tension like a rubber band being pulled back. Earthquakes happen when the tension gets released and the pieces of the fault move suddenly with the pent-up force of many decades.

The Earth naturally forms these tense spots. That's just how the movement of the crust works. But things that happen on the surface of the crust can affect when and where the tension gets released.

The crust of the Earth seems like a big, mighty, un-moveable thing when you're walking around on it, but it's actually relatively sensitive. Over decades, scientists have amassed evidence that the application of a heavy weight to the surface of the crust (or the removal of that weight) can trigger earthquakes.

For instance, when we build major reservoirs, we look for places that don't have a lot of seismic activity. For obvious reasons. But, sometimes, after the dam has been built and the reservoir has filled, the area will start to become seismically active. The big-name example here is the earthquake at Koyna Dam in India's Maharashtra state. The reservoir was filled in 1963. People began reporting small shakes soon afterwards and, in 1967, a magnitude 6.3 quake struck the region. The epicenter was very near the new reservoir. "There's a really compelling association with a reservoir here," says Susan Hough, a seismologist with the United States Geological Survey. "In this case, it was a part of India that was not all that active until the reservoir was built."

These time/place associations between earthquakes and reservoir building have happened reliably enough that scientists are confident that there is a connection. This isn't a controversial theory.

Nor is it controversial that the buildup and melting ice and snow can trigger earthquakes. The evidence for this comes from echoes of Earth's last Ice Age, when glaciers (heavy glaciers) stretched all the way down from the North Pole into places that are now quite decidedly temperate. When the Ice Age ended, the glaciers slunk back to to the Pole. And there's good evidence of a major increase in seismic activity that corresponded to the time and place where those glaciers were receding.

"The evidence includes a big increase in earthquake and volcanic activity in previously glaciated areas, such as Scandinavia and Iceland, increased volcanic activity in oceanic areas, and a greater prevalence of large submarine landslides," says Bill McGuire, professor of geophysical & climate hazards at University College London. "Some of [the landslides], the Storegga Slide off Norway 8,200 years ago, for instance, triggered major tsunamis that left their mark in the UK and elsewhere in the North Atlantic."

The effects aren't limited to things that happened thousands of years ago. Hudson Bay in Canada is basically just a divet where a patch of crust sunk under the weight of a glacier. It's slowly rebounding, in a way that we can measure today. In fact, in another 10,000 years, Hudson Bay won't exist at all, says Henry Pollack, professor of geophysics at the University of Michigan. Meanwhile, small earthquakes that happen today in Eastern Canada are thought to be associated with the Bay's slow spring-back.

All of this works because the movement of the crust can be influenced by the weight sitting on top of it. If you place a heavy weight, like a reservoir or a glacier, on top of a fault line, it might not move the way it otherwise would. Remove the weight, and it might go back to its original routine, or move in a different way. Either action, adding weight above a fault or removing it, could suppress or trigger an earthquake.

This is a simplified explanation. The reality gets a bit more complicated. For one thing, the weight of a glacier does more than just press down on what's immediately below it. "When you press your hand into the couch, [the fabric] doesn't just go down under your hand, it stretches out, too," Susan Hough says. Stretching the Earth's crust, and slowly letting it rebound back, could also trigger seismic activity. It's one theory—appealing, but still unproven—for why faults like North America's New Madrid experience seismic activity despite being hundreds of miles from a plate boundary.

By now, you should see where this is going. If naturally melting glaciers can trigger earthquakes, it stands to reason that a glacier that melts because of man-made climate change could do the same thing. Yes, it's a reasonable assumption. But …

There are a couple of caveats that you need to keep in mind. First off, while we know that changing climate has triggered seismic activity in the past, we don't really know yet how much seismic activity is likely to be triggered by contemporary, anthropogenic climate change. The effects still need to be quantified in this particular context.

Second, this effect might be happening already, but not in a way that has a big impact on most people. Bill McGuire of University College London and Patrick Wu, professor of geophysics at the University of Calgary, are two of the researchers really paying attention to the particular problem of seismic activity triggered by modern climate change. They both say the effect, so far, has been small—limited to low-level clusters of earthquakes in Alaska and around Greenland. In other words, where the glaciers, ice pack, and snow are melting. It's possible that we could see effects in other places—remember, weight on the crust stretches, it doesn't just compress—but we don't know that yet.

Third, Patrick Wu says it's unlikely that any really large earthquake, the magnitude 8's and up, would have a link to deglaciation. The earthquakes triggered by melting water tend to be smaller, he says, between magnitudes 5 and 7. Just because climate change can trigger earthquakes doesn't mean that every (or even most) earthquakes are triggered by climate change. Simple plate tectonics is still the primary force.

Nor can deglaciation trigger earthquakes in places that weren't at least somewhat earthquake-prone to begin with. "Deglaciation can't cause a crack. Tectonics can actually cause a crack, make a fault," he says. "Whenever there is glacial melting, faults can be reactivated. But it can't create the faults."

Finally, you can't look at the research being done by people like Wu and McGuire, look at the news, and go, "A-ha!" While these researchers are finding small increases in small, localized earthquakes in the far North, there's not actually been any dramatic, mysterious uptick in earthquakes that needs to be accounted for by this, or any other, theory.

The truth is, we've had several really big earthquakes in a relatively short period of time—Sumatra in 2004, Chile in 2010, and Japan in 2011. That's more than is normal. But not so many that the difference can't be accounted for by chance. And the broader frequency of earthquakes hasn't actually changed.

What has changed is our awareness of earthquakes, and their impact. Since the 1970s, we've had the technology and interconnectedness to reliably trace and report the vast majority of quakes that happen everywhere. Quakes that you'd have never known about 50 years ago are now on the evening news. And, more importantly, those quakes can appear to be more deadly because cities have gotten larger, allowing one earthquake to kill a lot of people in a relatively small area. All of that means that, watching the news, we perceive a bigger uptick in major earthquakes than there actually has been, according to the less subjective definition of "major" used by geologists.

Basically, it boils down to this: Climate change can trigger earthquakes. There's evidence that naturally occurring climate change did that in the past. There's some evidence that anthropogenic climate change might be doing that today. And there's evidence that we could see more climate change-related earthquakes in the future. But, if you're actually concerned about evidence (and you should be) then you can't go around, pointing to earthquakes that make the news today, and calling them consequences of climate change. And we can't oversimplify research to the point of forgetting all the yesbut.