Story and 2014 Napa earthquake photos
David Bolling


99 percent chance of 6.7 quake within 30 years.

63 percent chance of 6.7 or greater quake in Bay Region by 2036.

31–33 percent chance of 6.7 or greater quake on Rodgers Creek Fault.

7.2–7.4 maximum estimated earthquake on Rodgers Creek Fault.


This is not the news you wanted to read.

Not if you lived through the wildfires of October, not if you’re weary of disaster, uncertainty and total lifestyle disruption.

But here’s the simple set of seismic facts: A consensus of earth scientists agrees there is a 99 percent chance of a 6.7 or greater magnitude earthquake somewhere in California in the next 30 years. And of all the fault systems in Northern California on which that quake could occur, scientists are equally certain the top candidate—by far—is the Rodgers Creek Fault, that begins in San Pablo Bay and runs up along the western flank of Sonoma Mountain, just a ridge away from Temelec, Sonoma, the Springs, Glen Ellen, Kenwood, Oakmont, into and through Santa Rosa and all the way up to Healdsburg.

But that’s just part of the story. Scientists are now convinced that the Rodgers Creek Fault is connected to the Hayward Fault which notoriously runs along the East Bay metropolitan sprawl, through the UC Berkeley campus and directly beneath Cal stadium.

That subsurface connection means that the maximum quake possible on either extension rises from 6.7 to 7.2 or even 7.4, an order of magnitude so much greater that one study described the potential outcome as likely to “cause extensive damage and loss of life with global economic impact.”

On its own, the Rodgers Creek Fault has been given a 33 percent likelihood of delivering a 6.7 or greater jolt within 30 years. And in probability-speak, that means it could as easily happen tomorrow as in 2048.

Californians are remarkably sanguine about earthquake risks, even after witnessing the 6.9 Loma Prieta quake in 1989 (62 deaths, $6 billion damage), the 6.7 Northridge quake in 1994 (57 deaths, $13 billion to $40 billion damage estimates) and the milder, 6.0 Napa quake in 2014 (1 death, $362 million to $1 billion damage estimates).

By comparison, the current estimate of insured losses in the North Bay wildfires is more than $9 billion, and true losses are considered to be much higher.

In the aftermath of those wildfires, many of us learned belatedly what we could have done—and should still do—to prevent them. Most of us discovered we were woefully unprepared for such an epic event.

And yet, with all the evidence of earthquake damage around us, we are equally unprepared for the next big quake—a quake more likely to be in our backyard than anywhere else in the state.

To that end, we have compiled the following information—drawn from well-informed expert sources, providing guidance on what to expect from a big event on the Rodgers Creek Fault, and how to prepare for it.


How Bad
Will It Be?

Damage Projections For
Rodgers Creek Fault

Now that we know the odds and the reach of a major earthquake on the Rodgers Creek Fault system, the next obvious question is how much damage will be done and what will be the toll in human life.

The questions are methodically answered in the Sonoma County Hazard Mitigation Plan, which relies on intricate computer modeling to produce estimated damage and casualty figures. The plan was adopted by the Sonoma County Board of Supervisors in March, 2016, so the picture that emerges below is reliably up to date. What follows is largely the language of the report itself, addressing an earthquake scenario of a 6.7 event on the Rodgers Creek Fault.

What isn’t discussed, and remains unclear, is whether a Rodgers Creek earthquake would accompany, or be triggered by, a similar event on the Hayward Fault, to which USGS seismologists now know it is connected.

Earthquakes on the Rodgers Creek Fault will vary in intensity and resulting damage depending on a number of factors, including the depth of the epicenter and duration of the event. The likely consequences of earthquakes affecting Sonoma County can be estimated by using computer modeling software developed for FEMA. The California Geological Survey (CGS) has developed additional methods to model ground shaking and ground motion from seismic events and to characterize different earthquake scenarios.

HAZUS is a regional earthquake loss estimation model that was developed by FEMA with the National Institute of Building Sciences and the California Geologic Survey. Its primary purpose is to provide a methodology and software application to project earthquake losses at a regional scale.

The estimates of social and economic impacts contained in this report were produced using FEMA’s HAZUS–MH loss estimation methodology software, which is based on current scientific and engineering knowledge. There are obvious uncertainties inherent in any loss estimation technique and these modeled estimates are just that.


This earthquake scenario would result in strong, damaging ground shaking in areas near the fault, including the communities of Santa Rosa, Larkfield, Wikiup, Windsor, Rohnert Park, Petaluma, Roseland, Sonoma and Healdsburg. Much of the expected damage described below would occur in incorporated cities that have adopted Hazard Mitigation Plans to reduce future risk. The shaking will cause ground failure in some locations. The fault may rupture at the surface in various locations between Petaluma and Windsor, causing buildings, streets, sidewalks, power lines, fences, and other structures along the fault to be displaced by up to several feet. Liquefaction and landslides may occur and may be exacerbated if the quake coincides with wet weather.

Building Damage

More than 11,030 single-family homes would be moderately to completely damaged, particularly wooden structures that are not bolted to their foundations.

More than 1,200 households in the county will be displaced from their homes immediately after this quake, seeking temporary shelter with friends, relatives or emergency shelters. Many will be able to return to their homes within a few days; others will need to temporarily relocate as their homes are reconstructed. Most of these displaced families will be residents of cities. It is likely that only several hundred residents of the unincorporated areas of the county will seek shelter.

The model projects that 11,427 multi-residential units, 3,615 commercial units, and 1,229 industrial units will be moderately to completely damaged. Most of these damaged structures are located in the major cities along the fault, but about a dozen commercial buildings in the unincorporated areas may experience partial or complete collapse.

Given the high number of county buildings on or near the fault, this earthquake scenario could have significant impact on the county facilities and its post-disaster operations. Considerable financial losses could be incurred to the county, in structural and nonstructural damage.


An earthquake on the Rodger’s Creek Fault would be most lethal if it occurs in the middle of a weekday. Under that scenario the model predicts that 42 people in the county could be killed, and 972 people would be injured such that medical care or hospitalization is required. Casualties will be slightly lower if the earthquake occurs at commute time, and significantly lower if it occurs in the middle of the night. One reason the lethality of the earthquake varies by time of day is that wood-frame buildings, the predominant structural type of residences in the county, are less likely to cause injury and death than
structural types commonly used for commercial buildings, such as reinforced concrete. Ninety percent of all casualties will occur within cities along the Rodger’s Creek Fault, with fewer casualties occurring in the unincorporated areas.

Hospitals may be damaged during a Rodger’s Creek earthquake. It could take several hours before emergency medical personnel from other areas of the region are able to assist overwhelmed facilities in the county. The speed of outside help arriving will be impacted by the condition of major roadways and the amount of damage sustained outside the county’s borders. Medical care could be further impeded by power failures and limited potable water.

Post-Earthquake Fire

Fires often occur after an earthquake. Because of the number of fires and the lack of water to fight the fires, they can often burn out of control. The HAZUS model estimates that two fires will break out under this scenario. These fires will be sparked by various sources, such as pilot lights on toppled water heaters, and they can be fueled by broken gas lines. If the earthquake occurs during a time of year with high wildfire risk, it is possible for small fires to ignite wildfires, and Sonoma County is now painfully familiar with what that can mean.


Water, power and phone services will likely be unavailable in many areas immediately after the earthquake. An estimated 19,372 households will be without potable water and about 50,000 households will be without electricity. Power to many areas will be restored within days, but areas of fault rupture or other ground failure may experience longer delays. Both land-based and cellular telephone systems will be affected by the shaking, but are likely to be largely restored within days. Communication systems used by emergency personnel, such as radios and microwave systems, may also be impacted and retain only partial functionality.

County infrastructure will be significantly impacted by the earthquake. Key roads in the county such as Highway 101, 12, 116 and 121 may be impacted and could be impassable due to liquefaction, landslides or bridge collapse, hampering emergency response efforts, including emergency medical care and fire suppression efforts. If the earthquake occurs when the ground is saturated after a rainstorm, landslide and liquefaction impacts will significantly increase.


An earthquake of this magnitude would have long-term economic impacts for Sonoma County. A major earthquake on the Rodger’s Creek Fault could cause total income losses of more than $565 million and total capital losses of about $2.9 billion in the county, including losses from structures, contents, inventory and nonstructural damage. The most expensive damage will occur within cities. Long-term business losses and indirect economic damage could significantly exceed direct damage figures.

Leave a Reply

Your email address will not be published. Required fields are marked *