Hayward Fault: The Most Dangerous Fault in America

Introduction

California has no shortage of dangerous faults. The San Andreas Fault gets the headlines and the Hollywood treatment. But seismologists at the USGS have consistently pointed to a smaller, less famous fault running through the East Bay hills as the single most hazardous fault in the United States. The Hayward Fault earns that distinction not because of its size or its slip rate, but because of what sits on top of it: one of the most densely populated and infrastructure-rich corridors in North America.

The Hayward Fault runs approximately 75 kilometers through the heart of the East Bay, from San Pablo Bay in the north through the cities of Richmond, El Cerrito, Berkeley, Oakland, San Leandro, Hayward, and Fremont to the south. Roughly 2.4 million people live within 10 kilometers of the fault trace. BART — the Bay Area's commuter rail system — runs directly across the fault. Major water supply lines, freeways, and tunnels cross the fault at multiple points. UC Berkeley's Memorial Stadium was literally built straddling the fault trace.

The fault's last major earthquake struck on October 21, 1868, producing a M6.8 event that destroyed much of the town of Hayward and was felt across the entire Bay Area. At the time, it was known as the "Great San Francisco Earthquake" — a title it held until the far more destructive 1906 earthquake on the San Andreas Fault. Now, more than 155 years have passed since 1868, and the paleoseismic evidence suggests the fault ruptures on average every 150 years. The Hayward Fault is, by geological standards, ready.

Geology and Tectonics

A Branch of the San Andreas System

The Hayward Fault is a right-lateral strike-slip fault — meaning the far side of the fault moves to the right relative to an observer standing on either side. It is part of the broader San Andreas Fault system, which accommodates the motion between the Pacific Plate and the North American Plate. The total relative motion between these two plates is approximately 50 mm/year according to the USGS, and this motion is distributed across several parallel faults in the Bay Area, including the San Andreas, Hayward, Calaveras, and Rodgers Creek faults.

The Hayward Fault itself accommodates approximately 8–9 mm/year of slip, making it one of the more active faults in the system. It is the most prominent fault in the East Bay and connects at its northern end with the Rodgers Creek Fault in Marin and Sonoma counties. Seismologists have long been concerned that a large earthquake on the Hayward Fault could trigger a rupture on the Rodgers Creek Fault (or vice versa), producing a longer rupture and a larger earthquake than either fault would produce alone.

Fault Geometry and Depth

The Hayward Fault extends to a depth of approximately 12–15 kilometers, below which the crust becomes ductile and deforms by slow creep rather than sudden rupture. The fault dips nearly vertically, typical of strike-slip faults. Its surface trace is well-mapped and cuts through urban areas with remarkable precision — running through backyards, across sidewalks, and beneath buildings throughout the East Bay.

The fault's trace is visible in many locations as offset curbs, cracked walls, and misaligned structures caused by surface creep — the slow, continuous movement of the fault between earthquakes. This creep is one of the Hayward Fault's most distinctive and well-studied characteristics.

Sources: USGS, UCERF3, HayWired Scenario (2018)

Fault Creep: A Deceptive Calm

What Creep Means — and What It Doesn't

The Hayward Fault is one of the best-known examples of a creeping fault in the world. Surface creep — measured by the USGS at approximately 4.6 mm/year in the central section of the fault — causes slow, continuous deformation that is visible in offset curbs, cracked building foundations, and misaligned railroad tracks throughout the East Bay.

This creep might seem reassuring, as if the fault were slowly releasing its accumulated stress without the need for a large earthquake. But the math tells a different story. The long-term geological slip rate of the Hayward Fault is approximately 8–9 mm/year. If surface creep accounts for only 4.6 mm/year, then the remaining 3.5–4.5 mm/year of slip is not being released — it is being stored as elastic strain on locked portions of the fault. Over decades and centuries, this deficit accumulates into a strain budget that can only be released by sudden slip: an earthquake.

According to USGS research, the sections of the Hayward Fault that are locked (not creeping) are the sections most likely to produce the strongest shaking in a future earthquake. The creeping sections will also rupture, but the locked sections will release their accumulated energy more violently.

Visible Evidence of Creep

The Hayward Fault's creep is visible in everyday infrastructure across the East Bay. Among the most famous examples:

The Hayward Fault Creep exhibit at the Fremont Central Park, where offset curbs and sidewalks have been preserved and interpreted for public education, demonstrates several centimeters of accumulated offset.

UC Berkeley's Memorial Stadium, built in 1923, sits directly on the fault trace. Over decades, creep caused visible cracking and deformation of the stadium structure. A $321 million seismic retrofit completed in 2012 addressed both the creep damage and the stadium's vulnerability to a large earthquake, though the fault continues to creep beneath the structure.

Curbs and sidewalks along Rose Street in Hayward, Mission Boulevard, and numerous other streets show offsets of several centimeters accumulated over decades.

The 1868 Earthquake: A Forgotten Catastrophe

The "Great San Francisco Earthquake" Before 1906

On October 21, 1868, the Hayward Fault produced a M6.8 earthquake that was — at the time — the most destructive earthquake in San Francisco Bay Area history. The earthquake ruptured a section of the fault at least 30 kilometers long, producing severe shaking across the East Bay and significant damage in San Francisco, then a booming city of roughly 150,000 people.

The town of Hayward was nearly destroyed. Every major building in the town suffered severe damage or collapse. In San Francisco, numerous buildings collapsed, including the city's new county courthouse. The earthquake killed approximately 30 people — a toll limited by the relatively small population of the East Bay at the time.

The 1868 earthquake was widely known as the "Great San Francisco Earthquake" for nearly four decades, until the far more devastating M7.9 earthquake on the San Andreas Fault on April 18, 1906. That event, which killed approximately 3,000 people and destroyed much of San Francisco, eclipsed the 1868 earthquake in public memory. The 1906 earthquake ruptured 477 kilometers of the San Andreas Fault, dwarfing the 1868 rupture in scale.

But the 1868 earthquake should not be forgotten. It demonstrates that the Hayward Fault is fully capable of producing damaging earthquakes — and the next one will strike a metropolitan area with millions of people rather than thousands.

Paleoseismic Record

Paleoseismic trenching studies — in which geologists excavate across the fault to identify and date past earthquake ruptures — have identified approximately 12 surface-rupturing events on the Hayward Fault in the past roughly 1,800 years. This gives an average recurrence interval of approximately 150 years for significant earthquakes, though individual intervals have ranged from roughly 95 to 220 years.

According to USGS paleoseismologist James Lienkaemper and colleagues, the most recent events prior to 1868 occurred around AD 1715, AD 1630, and AD 1470, giving intervals of roughly 150, 85, and 160 years. The current interval — now 155+ years since 1868 — is within the expected range but at the longer end of the observed distribution.

The HayWired Scenario: Modeling the Next Big One

USGS 2018 Study

In 2018, the USGS published the HayWired earthquake scenario, a comprehensive scientific study modeling the effects of a M7.0 earthquake on the Hayward Fault. The scenario was developed by a team of more than 100 scientists, engineers, and emergency planners, and represents the most detailed assessment of the Hayward Fault's potential impact ever produced.

The HayWired scenario models a rupture beginning beneath the city of Oakland and propagating along the full length of the fault. The results are sobering:

• Fatalities: 800+
• Injuries: 18,000
• Displaced persons: 411,000
• Buildings damaged or destroyed: 152,000
• Direct economic losses: $82 billion
• Fire following earthquake: potentially hundreds of simultaneous ignitions, with limited water for firefighting due to pipeline breaks
• Aftershock sequence: hundreds of felt aftershocks over months, including several M5+ events that cause additional damage to weakened structures

The $82 billion figure represents direct physical damage. The total economic impact — including business interruption, supply chain disruption, lost wages, and rebuilding costs — could far exceed this figure. For comparison, the 1994 Northridge earthquake (M6.7) caused an estimated $44 billion in damage (in 2018 dollars), and the HayWired scenario models a larger earthquake in a denser urban environment.

Fire Following Earthquake

One of the most alarming findings of the HayWired scenario is the potential for catastrophic fire following the earthquake. The 1906 San Francisco earthquake demonstrated that fire can cause more damage than the shaking itself — roughly 80% of the destruction in 1906 was caused by fires that burned for three days.

According to the HayWired scenario, a M7.0 Hayward Fault earthquake could produce hundreds of simultaneous fire ignitions across the East Bay from ruptured gas lines, overturned appliances, and damaged electrical systems. Normally, fire departments can respond to these ignitions quickly. But in a post-earthquake environment, water mains are broken, roads are blocked by debris, communications are disrupted, and fire stations themselves may be damaged. The scenario models significant uncontrolled fires in Oakland, Berkeley, and other East Bay cities.

Critical Infrastructure at Risk

The Hayward Fault's path through the East Bay crosses an extraordinary concentration of critical infrastructure. Unlike faults in rural areas, where a rupture primarily affects open land, the Hayward Fault underlies and crosses systems that millions of people depend on daily.

Sources: USGS HayWired Scenario (2018), BART, EBMUD, Caltrans

BART: A System Built Over a Fault

BART, the Bay Area's commuter rail system, was designed and built in the 1960s and 1970s with limited seismic design standards by modern measures. The system's Transbay Tube — an immersed tube tunnel beneath San Francisco Bay — would experience severe shaking in a Hayward Fault earthquake, though it was retrofitted in the 1990s and 2000s to improve its seismic performance.

Multiple BART stations and tunnels in the East Bay sit directly on or near the Hayward Fault trace. BART has invested significantly in seismic safety in recent decades, including retrofitting the Transbay Tube and reinforcing elevated structures, but a M7.0 earthquake would likely shut down the entire system for an extended period for inspection and repair.

BART Seismic Safety Program

Water Supply: The Hetch Hetchy Lifeline

The Hetch Hetchy Aqueduct delivers water from the Hetch Hetchy Reservoir in Yosemite National Park to 2.7 million people in the San Francisco Bay Area. The aqueduct crosses the Hayward Fault, and a rupture could sever this water supply for weeks to months. The San Francisco Public Utilities Commission (SFPUC) has invested billions in seismic upgrades to the system, including the $4.8 billion Water System Improvement Program completed in 2016, but the fault crossing remains a critical vulnerability.

East Bay MUD, which serves 1.4 million customers in the East Bay, faces similar challenges. Its water distribution network crosses the Hayward Fault at numerous points, and pipeline breaks during a major earthquake would compromise both domestic water supply and firefighting capacity.

[MAP: Hayward Fault Through the East Bay] Data source: USGS Quaternary Fault and Fold Database, USGS HayWired Scenario. Features: Hayward Fault surface trace from San Pablo Bay to Fremont, fault creep rate measurements at key stations, BART alignment and stations, Hetch Hetchy Aqueduct crossing, East Bay MUD major pipelines, I-880 and SR-24/Caldecott Tunnel, UC Berkeley campus with Memorial Stadium marked, connection to Rodgers Creek Fault in the north and Calaveras Fault in the south. Major cities labeled: Richmond, El Cerrito, Berkeley, Oakland, San Leandro, Hayward, Fremont.

UC Berkeley Memorial Stadium: Built on a Fault

Perhaps no structure better symbolizes the Hayward Fault's risk than UC Berkeley's Memorial Stadium. When the stadium was built in 1923, geological knowledge of the fault was limited. Over the following decades, fault creep caused visible cracking in the stadium's structure, with the east side of the stadium moving northward relative to the west side at roughly 4.6 mm/year.

By the early 2000s, the stadium had accumulated roughly 36 centimeters (14 inches) of total offset from creep since its construction. The University of California undertook a $321 million seismic retrofit, completed in 2012, that included a new press box and structural improvements designed to allow the stadium to withstand a major earthquake while accommodating ongoing creep. The retrofit was one of the most expensive and complex seismic projects for any individual building in U.S. history.

The stadium seats approximately 63,000 people, and on game days it is among the most densely occupied structures sitting directly on an active fault anywhere in the world.

Earthquake Probabilities

UCERF3 and the 30-Year Forecast

The Uniform California Earthquake Rupture Forecast, Version 3 (UCERF3), published in 2015, provides the most comprehensive probabilistic assessment of earthquake hazard in California. For the Hayward-Rodgers Creek Fault system, UCERF3 estimates approximately a 31% probability of a M6.7 or larger earthquake in the 30-year window from 2014 to 2043. For the Hayward Fault alone, the probability is approximately 14.3%.

For context, the overall probability of a M6.7+ earthquake anywhere in the San Francisco Bay Area within the same period is 72%, according to UCERF3. The Hayward Fault represents a substantial fraction of that total Bay Area risk.

The UCERF3 model also considers the possibility of multi-fault ruptures — earthquakes that break across more than one fault simultaneously. A combined rupture of the Hayward and Rodgers Creek faults could produce an earthquake of M7.3 or larger, with significantly greater damage than a Hayward-only rupture.

The 1868 + 150 Year Question

The popular framing of the Hayward Fault as "overdue" is based on the simple arithmetic that 150+ years have passed since the 1868 earthquake and the average recurrence interval is approximately 150 years. While this framing is not technically wrong, it is misleading. Earthquake recurrence is not periodic like a train schedule. The paleoseismic record shows intervals ranging from roughly 95 to 220 years, and the current interval, while long, is within the observed range.

What can be said is that the Hayward Fault is in the latter portion of its typical inter-earthquake period, and the probability of a large earthquake is elevated compared to a fault that ruptured recently. Combined with the locked strain inferred from the creep deficit, the seismological evidence points to a fault that is storing significant energy for a future earthquake.

Living on the Hayward Fault

Preparedness in the East Bay

The Bay Area has better earthquake preparedness than most regions of the United States, reflecting decades of experience with seismic hazards and the institutional legacy of the 1906 and 1989 (Loma Prieta) earthquakes. However, preparedness for a Hayward Fault earthquake specifically remains uneven.

Retrofit programs for soft-story buildings — multi-unit residential buildings with weak ground floors, common throughout Oakland and Berkeley — have made progress. Oakland's mandatory soft-story retrofit ordinance, adopted in 2009, has required thousands of building owners to strengthen their properties. Berkeley has a similar program.

However, thousands of older unreinforced masonry buildings remain in the East Bay, many predating modern seismic codes. Residential preparedness — emergency kits, seismic strapping for water heaters, and knowledge of "Drop, Cover, and Hold On" — varies widely.

San Francisco Bay Area Earthquake History

San Francisco Earthquake Risk and History

What to Do During an Earthquake

California Earthquake Insurance Guide

Earthquake Retrofitting: What You Need to Know

Earthquake Early Warning

The ShakeAlert earthquake early warning system, operated by the USGS in partnership with state agencies, can provide seconds to tens of seconds of warning before strong shaking arrives. For the Hayward Fault, warning times would vary: communities directly on the fault might receive only 5–10 seconds of warning, while San Francisco could receive 10–20 seconds. Even a few seconds of warning allows automated systems to slow trains, open fire station doors, and alert individuals to take cover.

Sources

• USGS, "Hayward Fault — Is It Due for a Repeat of the Powerful 1868 Earthquake?" — USGS Hayward Fault
• USGS, HayWired Earthquake Scenario (2018) — USGS HayWired Scenario
• Lienkaemper, J.J., et al. (2010), "Evidence for a Twelfth Large Earthquake on the Southern Hayward Fault in the Past 1,900 Years," Bulletin of the Seismological Society of America
• Field, E.H., et al. (2015), "Long-Term Time-Dependent Probabilities for the Third Uniform California Earthquake Rupture Forecast (UCERF3)," Bulletin of the Seismological Society of America
• BART Seismic Safety Program — BART Seismic Safety
• San Francisco Public Utilities Commission, Water System Improvement Program
• USGS Quaternary Fault and Fold Database — USGS Fault Database