Earthquake Threats to
Transportation and Utilities
“We rarely see in full the cities that we live in. Focused on our daily lives, urban dwellers are often only dimly aware of the numerous, enmeshed layers of critical infrastructure that quietly hum in the background to make modern life possible.” (Macro City, 2014)
It is when infrastructure fails that we become keenly aware of our reliance, and the cascading impact a single failure has across multiple systems, sectors, and processes. Degrading infrastructure systems and future large earthquakes with epicenters near critical regional infrastructure could result in system outages that last weeks for the most reliable systems, and multiple months for others.
This report maps airports, passenger rail, roadways, fuel, electric, and water systems, and highlights their interaction with seismic hazards. We used publicly available information to describe how each system operates, and the consequence of system damage. The key findings warrant a transparent public discussion of the reliability the region desires for its vital infrastructure systems.
Want to skip the highlights and dive into a detailed report? Download the full report to learn more about the impacts of infrastructure failure on the Bay Area.
Bay Area Earthquake Hazard
We studied three earthquake faults that could cause damage to infrastructure systems and impact the entire Bay Area. San Andreas, Hayward, and Concord scenarios produce strong shaking across large areas that are dense with regional infrastructure systems. Other faults can have significant local impacts, but are not explored in this report.
It’s not just ground shaking and fault rupture that can damage buildings and infrastructure; liquefaction is often a much more damaging earthquake effect for linear infrastructure systems. Explore liquefaction susceptibility and scenario earthquake ground shaking maps in the graphics below. The USGS has liquefaction hazard maps (which include ground shaking potential) for Northwestern Alameda County, and Northern Santa Clara County.
The Bay Area’s 26 airports are well distributed throughout the region; however, in San Andreas and Hayward scenario events, the three international airports will simultaneously experience strong to violent shaking. A 2013 liquefaction report suggests that in both events SFO and OAK will experience a few inches of runway settlement in either San Andreas or Hayward events. SJC is in a susceptible liquefaction zone, but has completed a mitigation project to greatly reduce the risk of significant settlement.
Bay Area airports provide residents and businesses the ability to travel and conduct business across the globe. The airports support the regional economy by providing airport sector jobs, economic access to domestic and global markets, air cargo services, and tourism access. Commercial travel out of the three international airports will be tested by San Andreas and Hayward earthquake events. Four of the region’s five airports that can handle large aircraft experience strong to violent shaking in both the San Andreas and Hayward scenarios. In these scenarios Travis Air Force Base in Solano County is the only large runway outside of the strong shaking zone./p>
Large-scale seismic retrofit programs have resulted in a much more resilient transportation network. Still a single failure along non-redundant corridors can severely disrupt travel.
The busiest highway corridors in the region are parallel networks (a good thing), but are subject to simultaneous hazards in single scenario events (a bad thing). In a San Andreas event I-280 will experience violent shaking while US 101 will likely experience liquefaction. The same experience occurs in the East Bay in a Hayward event. I-580 zig-zags over the fault three times, while I-880 passes through very-high liquefaction hazard zones. In each case it is possible for the network to brought to a standstill if the redundant pairs are damaged simultaneously.
An extensive network of both road and rail infrastructure provide the Bay Area region with multiple modes of travel across most of the region. There are four main intra-regional and inter-regional passenger rail services. The figure shows the map of these systems and their respective ridership levels along each section of track. BART expects the majority of their system to be operational very soon after a large earthquake. The figure shows their expected system restoration after a M7.0 Hayward event both before and after their mostly completed seismic retrofit program, which began in 2002 (BART 2002a). The other rail systems are primarily at-grade lines that should be quickly repairable. Altamont, Amtrak, and Caltrain all have at-grade platforms, and for the most part have fewer bridges than most of the highways. In a Concord event, the rail bridge that crosses parallel to the Benicia – Martinez Bridge is only two miles from the Concord fault. In a Concord event, the shaking and/or liquefaction could cause significant or complete damage to the rail bridge.
In the nine county Bay Area region there are over 1,400 miles of state highways, and another 20,000 miles of local roadways (Caltrans, 2011). California road networks have had catastrophic failures in both the 1989 Loma Prieta and 1994 Northridge earthquakes. Since 1989, Caltrans has spent over $12 billion to seismically strengthen over 2,200 of 12,000 bridges state-wide. Over the past twenty five years since Loma Prieta, the region has seismically retrofitted all bridges that cross the Bay. In 2013, Bay Area Toll Authority (BATA) and Caltrans completed all planned seismic retrofits of bay crossings, including the replacement of the eastern span of the Bay Bridge. The Golden Gate Bridge, which is operated separately, has continually completed seismic retrofits since 1997 and has work scheduled until at least 2018.
“Each [bay crossing] retrofit is designed to a level that, at a minimum, will ensure that the bridge will remain standing in an earthquake. The California Legislature has designated the San Francisco-Oakland Bay Bridge and Benicia-Martinez Bridge as “lifeline structures” since they are located along transportation corridors determined to be crucial to both emergency relief and economic revitalization of the region following a major earthquake. Based on this distinction, the retrofit strategies for these two bridges incorporate some design elements that exceed standard seismic bridge design,” (BATA, 2013).
The Bay Area and all of Northern California are reliant on the five refineries and the Concord pumping station. Because these refineries are located near one another, built on similar soils, and constructed with similar standards, their performance is likely highly correlated. If there is damage to one refinery in an earthquake, it is likely other refineries are also damaged, interrupting a large percentage of the fuel refinement capacity in the Bay Area. If refineries are damaged a conservative restoration estimate is months.
Each studied scenario event will cause significant shaking across a majority of the refineries. These facilities are assumed to be extremely sensitive, as seen in the 2012 Richmond refinery fire when a single pipe failure led to a much more damaging fire. The damage from the fire required eight months to repair. In past earthquakes in Turkey (1999) and Chile (2010), refineries in the shaking region were completely shut down for three months, with limited capacity for over a year.
In addition to the risk of refinery damage, the export of product could be interrupted. All of the refineries export their refined fuel through Kinder Morgan’s Concord station. This facility is responsible for pumping fuel across the northern half of the state. The Richmond Chevron refinery also has separate refined fuel pipelines that service Brisbane, and San Jose; however, these pipelines represent a small share of the regional fuel. In Hayward and Concord scenarios, the Concord Station experiences strong and very strong shaking respectively. Additionally, in the Concord scenario there is potential for surface fault rupture that could damage both the station and incoming and outgoing pipelines. Severe damage to the Concord Station or multiple refineries would impact all of Northern California and Northern Nevada. Transporting a normal fuel demand by truck after a disaster simply is not feasible beyond service to the most critical facilities.
Damage to the region’s power generation facilities along the Carquinez Strait, or interruption in the natural gas system could result in long power supply interruptions.
No publicly available data source gives insight into the expected performance of substations, but historic earthquake have shown that substations represent the most fragile portion of the electricity distribution system. There are over 425 substations in the Bay Area with varying degrees of age and investment. There is no publicly available source on the varying age or retrofit status of these substations. No analysis could be completed on Bay Area substations.
In 2011, the Bay Area consumed 55,000 GWhrs of electricity, 60% of which was generated inside the nine county region (CEC, 2013a; CEC, 2013b). The remaining demand was met by power imports generated elsewhere in the state, the Pacific Northwest, and Southwest. Ninety-eight percent of the regionally produced power is generated at 25 large facilities with the remaining 2% generated at 44 small facilities with less than 50MW capacity. The 25 larger facilities are mapped in the figure.
Based on past earthquake damage and technical report documentation, only the energy generation and substations are likely to cause disruptions for a significant length of time. Of the regionally-generated power, two-thirds is produced by natural gas facilities, which are mostly located along the Carquinez Strait, an area that is bisected by the Concord fault. An interruption of natural gas would impact a large portion of electrical generation.
The Bay Area’s water supply comes from a portfolio of sources. The Mokelumne and Hetch Hetchy systems supply the Bay Area exclusively, while both the Central Valley Project and State Water Project supply water to regions across California. The Bay Area’s water supply is distributed by 89 different water providers (districts, agencies, and cities). Eleven providers distribute water to 93.7% of the Bay Area’s population. This research focuses specifically on the reliability of the region’s water transmission systems and the capability of the local water storage to meet water needs if outside sources are interrupted.
Most of the 11 Bay Area water districts studied have multiple water sources or have invested in robust, redundant, and repairable systems that contribute to system resilience. When reservoirs and groundwater reserves are above half full there is significant regional water storage available locally if regional systems require repair. Agencies dependent on Delta water would be significantly impacted if levees failed, causing flooding and salt water intrusion into State Water Project (SWP) and Central Valley Water Project (CVWP) sources.
SFPUC and EBMUD assessed the seismic performance of their own transmission supply systems and have since mitigated their transmission system to be more reliable. Both recognize that their distribution systems remain vulnerable. There is no record of the Central Valley Project (CVP) and State Water Project (SWP) taking comparable action to ensure their systems are functional in an appropriate time scale following a Bay Area Earthquake. Additionally the CVP and SWP systems capture water from the Sacramento-San Joaquin Delta, which is subject to salt water intrusion if levees that hold back water fail, resulting in a long term shut down of the CVP and SWP systems that supply the Southern half of the state (DWR, 2008). “A moderate to large earthquake in the San Francisco Bay region could cause major damage to Delta and Suisun Marsh levees, and could cause many of them to fail…Seismically induced levee failures would be expected to extend for thousands of feet if not miles and impact many locations simultaneously… For example, there is about a 40 percent chance that 20 or more islands will flood simultaneously as a result of an earthquake sometime over 25 years of exposure.” (DWR, 2008)
If interruption to out-of-region water sources were to occur, local sources and storage would be relied on until repairs were made to restore the transmission supply for districts reliant on imported water supplies. In communities and economic centers located on the bay margins water distribution pipelines may require weeks or months to repair liquefaction damaged pipes.
Over 200 reservoirs store water in the Bay Area, all with varying owners and operation goals. The 11 main water districts rely on 39 large local reservoirs with a maximum storage capacity of 3 million acre-feet. In addition to surface storage SCVWD, ACWD, and Zone 7 rely on local ground water for a large percentage of their storage and emergency supply. The graphic shows the relationship between a district’s average weekly water use and how much water is available when reservoirs are at 50% their total storage capacity. It also includes the addition of local groundwater reserves for the four districts with large aquifers. Within the region, there is capacity for the water system to operate in isolation from the water sources outside the region if local reservoirs are (1) more than half full, (2) ground water reserves are near current levels, and (3) inter-regional systems can be repaired in a few months. In a drought, it is possible that local reserves will not be sufficient to supply water while regional systems are repaired.
To increase redundancy, many agencies have constructed interties, or links, between systems. The interties can be used to share water during the interruption. The capacity of these interties supplies a fraction of the normal demand, but could be used effectively to provide emergency water to some locations.
This study only examines the vulnerability of the regional portions of water systems. An earthquake can cause severe damage to aged distribution pipes, requiring weeks if not months to restore water to all customers.
Damage caused by the earthquake will be only one source of failures. The failure of one infrastructure system will lead to the failure of other systems and slow the restoration of services.
In 2014, the City and County of San Francisco’s Lifeline Council, a pioneering council made up of utility operators that service the City, published its first Lifelines Interdependence Study. For the study, past research and utility interviews were used to roughly qualify the interdependence between systems. Figure 13 shows the matrix of interdependence between twelve important systems for the City and County of San Francisco. This information was then taken and displayed with lines in a scallop diagram. It is clear from both graphics that fuel is the system most relied on by all other systems. Roads, electricity, telecom, and water were also main systems relied on by others.
The San Francisco study was completed for the City and County of San Francisco. The specific relationship between systems may be different for other cities, but the overall interactions are likely to be fairly similar for the Bay Area region as a whole. The study is an example of the work a Lifelines Council can achieve. The Council has already worked to designate priority routes through the city that are critical for multiple systems restoration, and is currently magnifying its study of cell sites, fuel supplies, and utility staging sites. The Council should be used as a model to address issues of infrastructure vulnerability and interdependence for the Bay Area region.
Functional infrastructure systems are necessary for achieving community resilience. The consequence of infrastructure damage cascades well beyond the costs to repair the immediate damage. The failure of one system limits the functionality of other key regional assets, and will cause interruption for both households and businesses. While it is unrealistic to expect systems to be earthquake proof, knowing what to expect provides the users of infrastructure systems the information they need to take measured preparedness actions, or advocate for greater reliability. Currently, the vulnerability of many infrastructure systems is not well known or not well communicated to the public. With a lack of information, stakeholders have no baseline for predicting the benefits of possible preparedness or mitigation strategies. Going forward, the region must understand and communicate the vulnerability of infrastructure systems to inform stakeholders on what to expect so that they can make informed decisions to reduce impacts to their home or business should systems fail.
This study is a first step in understanding the risks to transportation, fuel, electric, and water systems. The report should be used to inform actions in the present, and also as a call for greater study and transparency of the region’s infrastructure systems.
Aagaard, B., Lienkaemper, J., Schwartz, D. (2012). Probabilistic Estimates of Surface Coseismic Slip and Afterslip for Hayward Fault Earthquakes. Bulletin of the Seismological Soceity of America, Vol. 102, No.3 pp. 961-979. June 2012.
Amtrak Capitol Corridor Joint Powers Authority (2013). Monthly Station Ridership Activity. Capitol Corridor Train Service.
BART (2002a). BART Seismic Vulnerability Study. San Francisco Bay Area Rapid Transit District. Prepared By: Bechtel Infrastructure Corporation
Lienkaemper, J., Baldwin, J., Turner, R., Sickler, R., Brown, J. (2013) A Record of Large Earthquakes During the Past Two Millennia on the Southern Green Valley Fault, California Seismology Society of America, Vol. 103, No. 4 p 2386-2403. August 2013.
Macro City (2014) Macro City. Brava Theater, San Francisco. 31 May 2014.
Thatcher W., Marshall, G., Lisowski, M. (1997) Resolution of Fault Slip Along the 470-km-long Rupture of the Great 1906 San Francisco Earthquake and its Implications. Journal of Geophysical Research. Vol. 102, No. B3, p. 5353-5367. March 10 1997. Altamont Corridor Express. (2013). Daily Summary. Herzog Integrated Transportation System