Why Aren’t We Afraid Of A Tsunami Hitting San Francisco?

When a 7.3-magnitude earthquake struck off Japan's eastern coast early Friday morning, we all feared a tsunami. But San Francisco gets earthquakes all the time, and we're not scared of a tsunami there. Why?
Wikimedia Commons

When news broke of a 7.3-magnitude earthquake off the eastern coast of Japan early this morning, our first reaction was to fear a tsunami. The devastating earthquake that hit Japan last March and left 15,000 dead was in large part so damaging because of the ensuing tsunami, massive waves of ocean water which crashed up to six miles inland and over a hundred feet high. Luckily, today’s earthquake and its aftershocks seem to have had minimal adverse effect, and the waves are not high enough to be damaging.

But it got us wondering. The San Francisco Bay Area, perhaps the most famous earthquake zone in the continental United States, is hit by dozens of small earthquakes every year–and only a century ago, pretty much the entire city of San Francisco was flattened by an earthquake. Yet we never worry about a San Francisco tsunami. Is that lack of foresight, or is something else going on entirely?

Tsunamis–the word derives from the Japanese characters meaning “harbor” and “wave”–are not like regular waves. Though they can be triggered by underwater landslides or even meteorological conditions, typically they’re the result of immense energy being transfered by displaced ocean water. And what displaces ocean water? Earthquakes.

There are a few kinds of earthquakes, and the differences between the ones triggered by earthquakes is where we’ll find our answer to this question. Most earthquakes are caused by the movement of the plates of Earth’s crust moving against each other, which you all know, because you are all very bright. So there are a few kinds of “plate boundaries,” and there are a few different types of “faults.” The differences are confusing, because they’re very very similar, but you can think about them generally that a plate boundary describes in a large general sense the movement of plates, and faults describe how chunks of plates interact with each other. Since the question of San Francisco vs. Japan is a large-scale question, we’re going to talk with the large-scale language of plate boundaries.

So, plate boundaries! There are three main types: convergent, divergent, and transform. Convergent is when two plates smack into each other, divergent is when they move apart from each other, and transform is when they rub laterally against each other. We’re putting aside divergent for now, because neither San Francisco nor Japan have to worry about them.

The movement of plates isn’t smooth and it isn’t clean; the plates stick to each other, pop loose, bounce backward of forward like a rubber band. (Yep, when you’re talking about this much rock, it has a tendency to act elastic.) The most important term you need to know about convergent plates is subduction. Subduction is when an oceanic plate smashes into a land plate (or “continental” plate). Oceanic plates are made of heavier rock, so when the Pacific Plate smashed into Japan, it slipped under the Japanese land plate. That’s not how all convergent boundaries work; sometimes, two continental plates will smash into each other and built a mountain range at the smack-point, which is what’s happening between India and Asia right now. But subduction doesn’t give us the Himalayas–just a ton of trouble.

Subduction does crazy things to the seafloor. Remember that earthquakes aren’t smooth; these two plates have been smashing together for eons, sticking in places, being jammed down into the Earth’s mantle, and all of a sudden, POP! Parts of the seafloor behind the contact point are forced up, in weird, non-uniform spots. If you had a piece of cardboard hanging off the edge of a table, and folded the edge down, it’d force some of the cardboard still on the table upwards. That’s kind of what’s happening here; it’s not only where the two plates smash together that sees the impact. But a whole mess of the seafloor very suddenly pops up.

That displaced an absurd amount of water, and water carries energy very efficiently. Imagine jumping into a bathtub–the water reacts pretty violently. So you’ve got tons and tons of water, moving towards shore. As it gets closer to shore, it picks up speed for awhile, because there’s less room for the water to be in, like when you put your thumb partly over a spurting hose. But eventually it starts to slow down due to friction with the ocean floor. Here’s where things get ugly: the water may be slowing down, but its amplitude is increasing. (Amplitude refers to the height of this slow-brewing wave or series of waves.) So it slows down a bit, but that doesn’t make this any less dangerous, because the wave is getting taller at the same time it’s scrunching together. Then it his the shore, and it’s a tsunami. A huge goddamn wave.

That’s what happens in Japan. But it’s not what happens in San Francisco.

* * *

San Francisco and Japan are both at risk of earthquakes because they lie close to plate boundaries. But they’re not the same kind of plate boundaries. Remember those three types, convergent, divergent, and transform? Japan’s dealing with a single convergent boundary, but San Francisco is staring down multiple faults, and the ones that matter are transform faults.

San Francisco’s seismic situation is incredibly complicated and, frankly, kind of a mess. We’re not dealing with one fault line and two plates here, even though it’s commonly referred to simply as the San Andreas Fault. The San Andreas Fault is the line between the Pacific Plate (which is oceanic) and the North American Plate (which is a land or continental plate), and it’s a transform boundary. That means instead of smashing into each other, those two plates are sliding past each other, violently scraping and getting stuck and popping free. The last time the San Andreas really let loose was in 1906, which any Northern Californian will know as the Big One. The 1906 earthquake destroyed 80% of San Francisco and killed thousands.

But if you follow the San Andreas northward, you’ll eventually get to a fork, where three faults all meet. It’s called a “triple junction,” and its where a small oceanic plate called the Gorda enters the fray in an everyone-loses tectonic brawl. The Gorda’s boundary with the North American plate is actually a convergent boundary, just like the one going on in Japan, but the Gorda plate is so small that it has very little leverage to cause all that much turmoil. It’s driven so far down into the earth that it’s mostly locked there, stuck unmoving. But that’s definitely not a safe place to be; that fault line between the Gorda and North American plates is called the Cascadia Subduction Zone, and it’s one of very few subduction zones that are capable of delivering a “megathrust” earthquake of more than 8.5 magnitude. There hasn’t been one there for three hundred years, but there certainly will be another eventually (although it probably will hit the Pacific Northwest and British Columbia harder than San Francisco).

So, San Andreas and Cascadia, that’s two of the three. The last one, where the Gorda and Pacific plates meet, is called the Mendocino fault, which is also a transform (slipping and sliding) boundary; jolts along this fault regularly cause earthquakes in northern California.

Of the three main fault lines that would affect San Francisco, two out of the three are transform faults, and one is currently (but ominously) inactive. Transform boundaries are just as dangerous to people on land as convergent boundaries; they still shake the hell out of the land, which can lead to fires and floods and all kinds of disaster. But one thing they don’t do is abruptly displace ocean water, because they’re moving laterally rather than up and down. So, no displaced water, no tsunami. And that’s why nobody’s worried about a giant wave turning Golden Gate Park into a swamp.

Emily Elert contributed a LOT to this article.