Oakland Geology

The first two posts in this series were about the Hayward fault itself in Oakland and about the different types of ground in Oakland that earthquakes on the fault will affect. This post goes into some details about the kinds of earthquakes we can expect in Oakland. I’m going to try and ignore all the interesting complexities — the subject is full of rabbit holes to go down — and keep things really simple.

Earthquakes are a release of energy that is stored in the rocks along and near the fault surface. The energy comes from the movements of the great plates of rigid rock that form the outermost skin of the Earth. Oakland sits in the middle of a wide boundary between the Pacific plate on the west and the North America plate on the east. In the grossest terms, the Pacific plate is sliding quickly northwest and the North America plate is moving slowly west, pushing against it like a semi merging onto a busy freeway.

The input of energy into the fault from plate tectonics is extremely steady, while the output of energy from the fault — in earthquakes — is pretty much random. That’s the essential mystery at the heart of earthquake science.

At our latitude, that wide boundary between the Pacific and North America plates extends from the San Andreas fault, across the Bay, to the Concord and/or Calaveras faults over the hills. Each of those faults takes up part of the overall motion between the plates. They’re basically long, vertical cracks that extend downward about 13 kilometers — below that depth (8 miles), the rocks are too hot and soft to crack and instead they just deform like modeling clay.

^{From the USGS Earthquakes Map}

The energy that earthquakes release builds up steadily in the Earth’s crust along the two sides of the fault. Friction keeps the two sides from sliding past each other. We say that the fault is locked. Instead, the rocks slowly warp, exactly as if they were great blocks of rubber. And at some point the friction is overcome, the fault ruptures, the rocks spring back into shape, and that elastic energy is released as an earthquake.

(This is where the math kicks in! Various well-known laws of physics allow us to turn my word descriptions into actual equations, and science can exert its superpowers. Careful measurements and creative mathematics give us ever-better answers to ever-deeper questions, and seismology, the study of earthquakes, has progressed into a vibrant field of science with important problems to explore. This paper by Rundle and Donnellan on earthquake clusters is a fresh example of research on the leading edge.)

An earthquake’s magnitude is based on how much the ground moves back and forth (or up and down), as measured by seismographs. Each unit of magnitude represents a factor of ten: if a magnitude-4 shaker moves the ground by a millimeter — not much but definitely perceivable — a size-5 event would move it a centimeter. The same unit of magnitude represents a 32-fold difference in the total amount of energy released, because the geometry is different. (The U.S. Geological Survey has a page with more detail about magnitudes.)

The amount of energy in an earthquake depends on how big a patch of the fault gives way. The largest possible earthquake, then, would happen if the whole thing rips. And how big is the whole thing? We used to measure the Hayward fault as 120 kilometers long, from Point Pinole to the hills east of San Jose, enough to generate a magnitude 7.0 quake. In 2016 we learned that the fault is directly connected to the Rodgers Creek fault in the North Bay, which extends up to Santa Rosa. The combined fault is roughly twice the length of the Hayward fault alone, and it could produce a magnitude 7.4 quake. That’s getting close to the size of the 1906 San Francisco earthquake (about 7.7), and it would tear right through Oakland.

That’s the largest possible earthquake — and also the least likely. There’s a strong element of chance in the way that ruptures grow. A rupture starting out does not “know” how big it will become. The growth of a rupture is more like a run of luck at a casino. Our earthquakes generally don’t rupture the ground surface unless they’re larger than magnitude 6 or so. Those, we have a hope of detecting in the ancient sediments along the fault.

Studies made by digging trenches across the Hayward fault have given us a fair idea of its history of large earthquakes. There seem to have been six in the last 900 years and a dozen in the last 2000 years. Age estimates are fuzzy, but these events aren’t very rhythmic. Their average rate is one every 160 years, but the time between these major ground-breaking earthquakes has ranged from 95 to 225 years.

It’s been 152 years since the last one on 21 October 1868, which had a magnitude estimated at around 6.8. While that matches the average rate, these things aren’t like clockwork. Earthquakes aren’t scheduled. Therefore the next Big One is not “overdue.” Nevertheless, it appears that enough stress has accumulated on the fault since 1868 to power another event of about the same size. The scientific authorities say that a large earthquake (magnitude 6.7 is the size they picked) has a two-in-three chance of happening by 2043 somewhere in the Bay area. Half of that probability comes from the Hayward-Rodgers Creek fault.

What about smaller quakes, like magnitude 5 or greater? These would be strong enough to knock down things like chimneys and crack walls and windows. The area around the Hayward fault has had maybe 20 since 1850, as shown in the map below.

^{Magnitude 5 earthquakes since 1850 within the box including the Hayward fault, from the USGS catalog search}

But since 1889, only three were actually on the Hayward fault (5 September 1955, 13 June 1988 and Halloween 2007). Historically (aside from 1868), Oakland has suffered more from large earthquakes elsewhere in the Bay area than from homegrown ones. The most recent of those were in October 1989 (Loma Prieta, magnitude 6.9) and August 2014 (Napa, magnitude 6.0).

We’ve been lucky. But luck is a human concept, not a geological fact. There is a hint in the Bay area historical record (documented in this 2002 paper and elsewhere) that big quakes have been preceded by clusters of middle-sized ones. Or rather, big quakes appear to quiet down activity on Bay area faults for several decades.

Time keeps on ticking into the future. Our fault will reawaken. My next post will look more closely at that. In the meantime, I recommend that you bookmark Temblor.net in addition to the usual sites at the U.S. and California Geological Surveys. In a Web long plagued by armwavers, alarmists and frauds, these guys are quick on their feet, know what they’re talking about and know what to say.

In my last post I talked about the Hayward fault itself, the actual crack in the ground. I argued that the extra hazard of living right on the fault doesn’t add much to the risk, which is already high for other reasons. Those reasons are mainly two: the size of likely earthquakes and the type of ground that they shake. This post goes into the ground and how the ground can fail during earthquakes.

The ground is a BIG factor in how the same quake affects different places, as we learned from the 1989 Loma Prieta earthquake. Vibrations from an event way down below San Jose damaged a number of buildings in downtown Oakland, some fatally, but none of them collapsed outright. By contrast, the same earthquake, at the same distance, singled out one type of ground — landfill in West Oakland, formerly wetlands of Bay mud — and brought down the Cypress Street Viaduct, crushing 42 people to death. Soft ground slows down seismic waves, which pushes their energy into a smaller space, and lowers their frequency. In a word, soft ground amplifies shaking.

The Oakland flats consist of alluvium and have no solid rock anywhere near the surface. They’re deep dishes of mud, sand and clay, saturated with groundwater. When you shake this kind of material, the water in it starts disrupting the sediment into a slurry. Think of the times you’ve played near the water at a sandy beach, patting the damp sand until it turns wet, shiny and loose. The water transmits the energy of your hand very efficiently and drives apart the sand particles.

A moderate quake in alluvial ground will start triggering little eruptions of sandy water — sand boils.

_{Sand boils formed by the 2011 Virginia earthquake; photo by Mark Carter, US Geological Survey}

Larger quakes raise the risk that the ground will fail en masse, losing its strength in the phenomenon called liquefaction. If this happens near a vertical boundary, like a streambank or a harbor channel, the ground may spread out sideways (lateral spread). Not just buildings but anything on or in the ground, like pipelines and railroad beds, can fail during liquefaction.

To help guide building designers and city planners, maps showing liquefaction-prone areas are published by the US Geological Survey and by the California Geological Survey. Here’s a sample of the state’s map, which shows the suspect areas in green. These are not areas of guaranteed damage, but zones where we should be aware of the likelihood of damage.

Notice that flat, wet areas aren’t limited to the coast — green areas extend up every stream valley, even in Montclair.

The high ground has its own geologic vulnerabilities. First, of course, is that it’s already prone to landslides: the slopes are steep, our rocks are generally soft and like to disintegrate into soil, and their bedding is steeply tilted, which helps rainfall to infiltrate them. The map above shows landslide-prone areas in blue, in the upper right corner east of the Hayward fault (the black line). Strong shaking will trigger landslides in these areas — not everywhere, but anywhere.

Notice that there’s blue immediately west of the fault too, in Piedmont and west Montclair. The blue areas are smaller there because the rocks are different: harder Franciscan rocks as opposed to softer Great Valley Sequence rocks.

The second geological factor that affects how this ground responds to earthquakes is its topography. Steep ridges tend to direct earthquake waves upward in a sort of whip-crack effect, so that the highest ridgetops shake most violently. This is an area of active research, because it’s complicated.

Finally there’s the middle ground, the low hills that sweep from Pill Hill through Grand Lake and Haddon Hill and Maxwell Park to Evergreen Cemetery. These hills have no bedrock, but they’re made of alluvium that’s older and firmer than the flats. The problems lie around their edges, where the slopes are steepest and prone to landslides, like those at Jungle Hill and McKillop Road and Wallace Street, among others.

If you live in parts of Oakland that consist of these types of ground, I won’t say “you are warned,” because I’m not licensed to practice geology, but you are hereby informed. These facts are part of what makes up your overall picture.

My next post will get into earthquakes themselves, and how they’ll affect Oakland.

The topic of this set of posts arises from the terrible wildfire season we’re experiencing in 2020, on top of the terrible pandemic, on top of the unemployment crisis, and other more distant events that may be affecting our friends and family. We’re all learning a lot about these topics because directly or indirectly they personally affect each one of us.

When we think about them, we each come from our own place, and my place is geology. I find myself thinking about the ways these new disastrous events feel compared to the old familiar geological risks that have been on my mind all along.

So I thought I’d talk about earthquakes and their risks more deeply and more frankly than I usually do on this blog. I’m doing this because the catastrophes of 2020 have given us all a lot to think about, and if your perspective on those things has evolved — for instance, the way you react to masks — maybe you can think about earthquakes from a different perspective too. Let me stipulate right off that this is only my earthquake advice, and that I speak for no one else and do not pretend to supplant the US Geological Survey, the state Office of Emergency Services or any other authority, although I endorse everything they say.

Oakland is a special case when it comes to earthquakes. Here the risks are different from most other cities because the Hayward fault runs through the whole city, including some important parts. And the long-anticipated large earthquakes — not just one Big One like the 1868 earthquake but about ten times as many damaging Pretty Big Ones — will hit the whole city hard, not just along the fault.

Every so often, someone writes to me with a question about buying or living at a homesite on the Hayward fault. For them it’s always about the fault itself. Something about a rip in the ground on a million-dollar piece of property brings out these sharply felt concerns about “The Fault.”

I’m not a licensed geologist, so what I can say is legally limited. I always say that first. Practicing geology in California without a California geologist’s license is a crime. If someone really wants specifics, they need to engage the correct professional — architects, contractors, geotechnical engineers, home inspectors, lawyers — and all of those experts are legally limited too. The expertise of specialists is partial and does not add up to certainty or wisdom.

The ways I try to help questioners are to provide data about the hazard and correct errors in their thinking about the risk. Here’s an example: a person wrote me with a question about a house that was apparently right on the fault. They’d looked up all the maps they could find, but wanted newer maps, maybe unpublished maps, that showed the fault’s location more precisely. They were asking for more knowledge about the hazard, namely a rip in the ground that damages the house.

Before answering, I had to step back, because better maps aren’t available (if they even exist) and aren’t necessary for making a decision. The thing about the fault is that we don’t precisely know where the rip in the ground is, except in a few well-defined places. And even where we do, the fault is not just a sharp line on the ground but a zone, sometimes tens of meters wide, that will warp and crumble when the fault gives way. In other places, like the south end of Redwood Heights, we aren’t sure where it even is.

Earthquakes happen deep underground, where most of the energy is, and the surface where we live is near the outer edge of most quakes. Up here along that outer boundary, being “on the fault” gets fuzzy. Every large earthquake is different. You can’t count on the ground breaking in the same place each time.

The state’s official solution to this uncertainty is the Alquist-Priolo Act, under which the authorities do their best to map active faults and then establish a wider zone around the faults (usually 50 feet) that effectively has the same hazard. Insurance companies, for instance, rely on these. So my answer to questions about better fault maps is that (1) this is as good as they get and (2) this is as good as they need to be.

There is one way to get more certainty, which is to look at the ground very carefully, and call in an expert to confirm any suspicions that arise. That’s because the Hayward fault doesn’t only rip the ground; it also pulls the ground, very slowly, all the time — the process called aseismic creep, or just “creep” for short. If creep on the fault is already pulling the house apart, the signs may be there, although deep landslides can create the same signs.

This is about the best one can do in gauging the hazard for a particular site. As for the risk, well, being in an Alquist-Priolo zone presents a relatively high risk and being directly affected by creep presents a risk that is absolutely high, not just relatively high. But like I said, the next rip in the ground may come somewhere else, maybe across the street, so the risk is still not 100 percent certainty.

This is where I step even further back, back to where knowledge may edge into perspective. From here, all natural threats are alike at their core. I can tell you this: Earthquakes are inevitable threats, but homebuyers roll the dice and most of them come out winners, because they survive without experiencing a Big One and their houses sell at a profit, and in the meantime they have enjoyed years of pleasurable life in their homes. Even a house directly on the fault, if it’s well sited and well made, will not kill you in a major earthquake.

I think that’s the attitude most people around here have, a California attitude. It strikes outsiders as odd, as I recorded in a story at the end of this post from 2017.

But life and death are not the only risk considerations. There’s also injury, damage and inconvenience, matters that mix concern with money. I’ll get into that topic in posts to come.