Earthquake Resilience of Southern California’s Water Distribution Systems

The tortured rocks in the Coachella Valley’s Painted Canyon are evidence of California’s earthquake history (Photo by Maven)

All the major water infrastructure bringing in water to Southern California cross the San Andreas fault at least once, sometimes multiple times.  What would happen in the event of an earthquake?  UCLA’s Dr. Jon Stewart (and others) have been conducting research to find out.

California is earthquake country, renowned for being one of the most seismically active regions in the world. There are more than 300 faults criss-crossing the state that we know about, and an untold number we know nothing about.  Every year, there are more than 10,000 quakes in Southern California, most too small to be felt.  The USGS says there is a 99.7% chance a magnitude 6.7 quake or larger will strike in the next 30 years.

California relies on a network of dams and aqueducts to store and transport water throughout the state. Southern California, in particular, relies on this infrastructure for 60% of its water, receiving imported water from the Owens Valley, the Colorado River, and the Bay-Delta Region.  Much of the state’s infrastructure crosses these faults.  How will our water infrastructure fare when the big one hits?

Dr. Jonathan P. Stewart is a Professor and Chair of the Civil and Environmental Engineering Department at the University of California-Los Angeles, specializing in geotechnical engineering, earthquake engineering, and engineering seismology.  In this presentation, Dr. Stewart defines the meaning of resilience as applied to water systems, describes the seismic threats to California’s water systems and gives his thoughts on critical system components with and without suitable resilience.

This post is derived from a presentation given by Dr. Stewart for Distinctive Voices, a public lecture program sponsored by the National Academies of Sciences, Engineering, and Medicine.

Californians have a general awareness that earthquakes are a threat, but despite the general awareness, there isn’t necessarily an awareness of what the most serious risks are, said Dr. Stewart. “As an earthquake engineer, I worry about things like that as that’s my job,” he said.  “At the top of my list is the risk to the water system that we have.   So I’m going to try to explain why I’m worried about this, and explain also some of the things that the state and local governments here in Southern California and elsewhere are trying to do to address these things.”

Dr. Stewart acknowledged his presentation will draw on the work of many people; their work has been supported from the National Science Foundation, the state, DWR, and others over the years.  Several faculty members and post docs have contributed, as well as many professionals and practicing engineers, both nationally and internationally.


Resilience is defined as the capability of the system to maintain some functionality after a stressing event.  Dr. Stewart noted that with few exceptions in earthquake engineering, infrastructure isn’t designed to be undamaged as damage in severe events is to be expected.  The goal is to recover quickly from the damage without losing functionality for an extended period of time.

The graph on the slide depicts functionality with 0 being not working at all and 1 being fully functional.  If there are no earthquake or serious events, we muddle along at the functionality of one.  When there is a serious event or disaster, we lose functionality; there will be some losses.

While disasters are bad, from a resilience point of view, recovery can occur in a matter of weeks; there are effects, but society can go on without massive long-term impacts.  The Northridge earthquake 20 years ago was a disaster.

Catastrophes, however, are different.  Catastrophes are unacceptable.  “We should not be having catastrophes; we should be engineering our systems to avoid that, because where as you may have the same level of initial damage, it takes a lot longer to recover,” Dr. Stewart said.  “So what resilience is really about is the area underneath the curve [on the graph].  The larger that area is, the less resilient our system is.”

There are a lot of ways this functionality can be measured.  The metrics the LA Department of Water and Power uses are:

  • Water delivery: does anything come out of your tap after the earthquake?
  • Water quality: can you drink it? Poor quality water limits the usefulness.
  • Quantity: Does everybody get what they need?
  • Fire protection: there must be some degree of pressure to the fire system to deal with post-event fires
  • Functionality: is it working at the full capability that we would like?  This is usually going to take a lot longer to achieve.

Dr. Stewart than gave an example of a disaster using the 1994 Northridge Earthquake that struck Southern California at 4:30 am on January 17th.   He presented maps showing the faults, noting that the worst shaking was in northern San Fernando and Santa Clarita.  The quake was a blind thrust fault that produced some intense shaking.  Fifty-seven people died; there was over $10 billion in damage from this event.

The map (above, right) shows the pipes in the water distribution system that were broken as a result of the earthquake.   Dr. Stewart pointed out that the Northridge Earthquake was the most damaging event for a water supply/water distribution system in the United States since the 1906 San Francisco Earthquake.  There were 74 breaks in the large trunk lines which led to major losses of pressure in parts of the San Fernando Valley, and over 1000 breaks in distribution lines throughout the city.

A couple of the breaks were particularly dramatic.  Balboa Blvd in the northern San Fernando Valley was displaced about a meter to the south, as a result of softening of the soil caused by the earthquake and movement on a very mild slope down the hill.  The displacement broke the gas line as well as the water line; the gas line was ignited when the owner of that truck came out and tried to start it, with led to an explosion.  “You have this juxtaposition that could only happen in LA with fire and water all together in the same place, and that was an image that was on the cover of the LA Times at that time,” he said.

The break happened because the pipe was old and brittle and couldn’t sustain the displacement of the ground around it, so it broke in several places.  “We have pipes like that throughout the area,” he said.  “This is not a unique occurrence.  We can absolutely expect things like this to happen when we get other Northridge type events in the future.”

Dr. Stewart presented a graph showing how the functionality of system was affected after the earthquake.  The y axis is time in days, from one day to two weeks.  Water delivery was essentially restored to full capacity within about a week; they did not have all the lines fixed by that time, but they were able to tap into alternative sources to get water into the system downstream of where they breaks were in the system, he said.

Initially, they didn’t have the quantity, but just a few days later, they more or less had the quantity they had before.  Fire protection was restored reasonably soon.  Quality took longer; there were boil in place for a while.  Getting to full functionality took many months, even more than a year, as they had to repair all the lines to restore the system to where it had been prior to the earthquake.

The Northridge event to an earthquake engineer is not all that extraordinary,” Dr. Stewart said.  “In fact, while that particular fault we don’t expect to be producing earthquakes on a decades or even a century-type of timescale, there are so many faults like that throughout southern California.  When you look collectively across the region, we can basically expect Northridge type earthquakes every few decades, and the record more or less bears that out.  That’s pretty much what we get.”

To an earthquake engineer, a mid-6 earthquake is small, so 6s to 7s we can expect to occur on approximately that time interval, and we can expect those kinds of losses,”  he said.  “That was a disaster; it was not a catastrophe.  LA went on.”

A catastrophe, I would argue, was Hurricane Katrina.”

Although he acknowledged Hurricane Katrina was not an earthquake, it still makes the point about the impacts of a catastrophic loss of resources.  Katrina started out as a category 5 hurricane over the Gulf; when it came onshore near New Orleans, it was category 3.  The levee system did not hold up in New Orleans; there were 53 breaches and 85% of the city flooded with over $108 billion in losses, as compared to $10 billion in Northridge.  One of the consequences was substantial loss of water service for quite a long time – not until over a year later; there were a lot of other problems in New Orleans not related to water as well.

Looking at the resilience for the city as whole, the catastrophic nature of this can be measured through the population loss and the related economic impacts, he said.  He presented a graph [above, right] showing employment data for the U.S. and for New Orleans, noting that they are scaled so it’s the same percent change from top to bottom.

During those three years prior to Katrina, New Orleans was more or less tracking the national trend, and then you had this dramatic loss,” he said.  “Now there’s been some recovery, but over a ten year period; there was this huge loss of people who left and never came back.  All the economic activity that goes along with population was lost, and has not recovered a decade later.  That’s catastrophic, and that’s the kind of risk I will argue we could face here in Southern California if we don’t get our act together with some of our water supply systems.”


Next, Dr. Stewart focused on the resilience of California’s water systems, noting that he won’t be discussing the distribution systems that distribute the water within the urban areas, but more about the major aqueduct systems that bring water into Southern California.

He presented a map of the state’s water infrastructure.  There are three major aqueducts that provide water to Southern California:   The Los Angeles Aqueduct, built by Mulholland, which brings water from the Owens Valley to Los Angeles for the LA DWP; the Colorado River Aqueduct which is operated by the Metropolitan Water District; and the State Water Project’s California Aqueduct, which comes from the Delta and is owned and operated by the California Department of Water Resources.

The water quality of the three systems is worth mentioning, he noted.  The water quality from the Los Angeles Aqueduct is excellent, and the water quality through the Delta is also pretty good; it doesn’t have many salts in it so it requires fairly minimal treatment before it can be used.  However, the water quality from the Colorado River Aqueduct is the worst of the three; there’s quite a bit of salts in there and it’s not usable, he said.  “Colorado River water can’t be wheeled out into the distribution system.  The reason they can use it is it gets mixed, so you mix the poor quality Colorado River water with the higher quality Delta water, and then with a little treatment, it goes out.  That’s kind of how the system works.  You wouldn’t want to have only Colorado River water.”

The major earthquake fault in California is the San Andreas fault, and he noted that each of the three water systems crosses the fault line at some point.  The Los Angeles Aqueduct crosses at a place called Elizabeth Lake; the tunnel underneath it is called Elizabeth tunnel.

That particular segment of the San Andreas fault last ruptured in 1857 during the Fort Tejon Earthquake; the return period, or the average time between major earthquakes of about that size, is estimated to be between 140 and 160 years, so we’re actually a little bit overdue for that event to happen again, he said.

The Elizabeth tunnel is about 3 meters in diameter.  One thing earthquake engineers do is forecast risk and what’s going to happen when the earthquake occurs.  They have models for how much faults move when earthquakes of different magnitudes occur, he said.

We know that by going out and looking.  We measure it.  If you have a 7.8 earthquake on a strike slip fault, there’s a whole range of displacements of course, but the average is about 5 meters,” Dr. Stewart explained.  “So if you have a diameter tunnel of 3 meters and you have 5 meters, you don’t have anything left. You have a tunnel crossing a fault and basically it gets offset by the rupture of the fault, the water can’t go through anymore.”

The Colorado River Aqueduct crosses at the San Giorgonio Pass.  The last time this part of the fault ruptured was in 1680, so an earthquake here is really overdue. “It’s been over 300 years, and that’s the one I and all other earthquake specialists are the most worried about,” he said.  “Because it’s so primed, we’re worried about a major rupture starting on the southern segment of the fault, perhaps near the Salton Sea and rupturing up through this area.”

He presented a slide showing where USGS has mapped the fault in the area; the three triangles represent the three points at which the aqueduct crosses the fault.  “One thing to appreciate if you haven’t studied faults is that they are complicated,” Dr. Stewart said.  “There are multiple branches.  The San Andres Fault is not single plane, and this is a place where there are number of splays.  There are actually three major fault crossings of this one aqueduct and they are all parts of the San Andreas Fault.”

At this particular point along the aqueduct, it is a shallow cut and cover conduit, so it is more accessible then the Elizabeth tunnel.  “But as you can imagine, Metropolitan is very worried about this,” he said.  “They consider this one of their greatest threats.”

Lastly, he turned to the State Water Project.  The California Aqueduct has substantial hazards at fault crossings at the southern portion, which is why he drew the circle so big, he explained.

It basically comes to the fault and it kind of goes around the fault and it’s crossing at multiple times as it goes along,”   he said.  “It turns out to be a nice route for having a good gradient for keeping the water flowing, but at the hazard of being next to and crossing the fault multiple times.  It’s obviously accessible, so if it was damaged, you could go in and fix it more easily than Elizabeth tunnel, but there are many kilometers that you have to deal with.  The channel is basically formed by compacted fills along here, and we could have many kilometers of these things failing.  It’s fixable, but it would be a pretty substantial effort with a lot of delay to go in and do all of that.”


Dr. Stewart then turned to the intake of the State Water Project and the risks to the Delta and Clifton Court Forebay.

The Delta serves a diverse array of interests, including agriculture; levees were built in the 1800s and the land reclaimed for agriculture.   He noted that it’s a very sensitive environmental area, with several endangered species, such as the Delta smelt, living there.

The Delta is at the confluence of the Sacramento River and the San Joaquin River, where the two rivers join together and then flow west to the San Francisco Bay.

This is a major node in our water system,” said Dr. Stewart.  “You can’t overstate the importance of the Delta to our water system.  Basically everything west of the Sierra crest, all of the precipitation falling in the mountains west of the crest, which is most of it, is draining into the Central Valley and going out to the sea through the Delta, so most of California’s water is going out this way.”

He noted that there’s the balance between seawater and freshwater that in the Delta that water resource managers deal with every day.  The fresh water coming in from the rivers meets the seawater pushing in from the San Francisco Bay, creating a natural transition zone from freshwater to seawater that has to be delicately maintained at all times; the  seawater cannot be allowed to penetrate inland to the south Delta where the export pumps are located or else the water would be too salty to be used.

Prior to human development of the Delta, there were natural channels bringing water from the Sierra down to the Bay much as they do now, Dr. Stewart explained.  Every once in a while, there would be a flood event and the water would go over the top of these natural levees, and as it would flow over the top, the coarsest sediments would fall out by the edge, and the finer sediments would flow onto the landscape.  This caused natural levees to form at the edge of the channel with the interior portions with very rich material and all kinds of vegetation naturally growing.  As that vegetation dies, the ground was replenished, so there was a natural cycle of release of methane and carbon dioxide from the dead vegetation, along with the replenishment with new sediments which created a sort of balance in the ecosystem.

Then there was the Gold Rush.  Many who could not make enough money mining gold turned to farming, and thus farming was started in the Delta on some of the richest soil around.  To prevent their farms from flooding, they starting building levees on top of the natural levees.  The federal government encouraged the reclamation of the Delta and other marshes through the Swampland Act; Reclamation Districts were formed where owners got together to build and maintain levees on their islands.

Reclamation of the Delta started around 1850 and continued through about 1930.  Dr. Stewart noted that the picture is a clamshell bucket that was used to grab the soil, whatever was in the soil, and then dump it.

There wasn’t a lot of engineering going on here,” he said.  “There’s no compaction, just taking the soil and basically dumping it and not worrying too much about tree limbs and things like that that might be sitting there.  Needless to say, we would not build it that way today.”

He reminded that it was the overtopping of the levees that was replenishing the interior and maintaining a sort of ecosystem balance, but once levees were built to prevent the flooding, the interior was no longer bring replenished by the decay of the organic matter.  “The soil there is peat and it’s organic soil, so the decay still happens,” he said.  “It dries out, releasing carbon dioxide, and the wind comes along and literally blows the soil away.  The interior would naturally fill up with water, but they pump the water out to have that be usable as farmland.”

As farming continues, the subsidence of the land continues, basically making the heights of the levees taller and the gravity stresses greater, and in the unengineered systems, it’s not unusual to have a failure, Dr. Stewart said.  The subsidence rates can be as much as 10 centimeters per year, with the interiors of some island being below sea level as much as 7.5 meters.  The subsidence contributes to the instability of the levees, he said.

Here’s a number to keep in mind,” he said.  “If you add up all the volume that is below sea level in the interior of these islands, it is 3 billion cubic meters of volume in the interior of these islands.  That is actually a huge number and a very significant number in understanding the risk.  The thing about it is this isn’t static.  This keeps getting worse year after year.  We’re pumping out those islands, the wind blows, and the soil lowers, so those levees are getting taller relative to the interior island every year.

In the Delta, the levees are owned by many different entities: some are owned by the Army Corps of Engineers and the state, but most of them are owned by individual reclamation districts, he said.  And while levees are used all over the world for flood protection on rivers everywhere, normally the levees are not continuously impounding water.  These levees are different; they are more like dams because they are always impounding water due to the lowering of the material in the interior of the islands.

One of the important spots in the Delta system are the pumps at the south Delta that export water to the Central Valley and Southern California.  “A lot of these levees, aside from protecting the islands, are channeling water through the Delta to the pumps,” said Dr. Stewart.  “So securing a proper flow of high quality water to the pumps is essential for the state of California.”

These levees are marginally stable,” he continued.  “Levees fail.  They fail on clear blue sky days without any earthquakes or any flood event.  We don’t always know why; maybe it’s burrowing animals or whatever.  They are often sitting there in a marginally stable state and so we can’t be too surprised when they fail.  The levee in Upper Jones Tract just failed one day in 2004, flooding the interior island.  They came in and fixed it at considerable trouble; it cost $100 million to fix that and restore the area.  If you look at the historic record, these types of things are not unusual at all.  In fact, between 1900 and 2000, there’s been over 150 failures from non-earthquake sources in the Delta.”

Dr. Stewart acknowledged that whether or not the Delta is vulnerable to earthquakes is subject of some debate in some parts of the state, particularly the Delta.  He presented a map of earthquake risk as peak ground velocity, explaining that the hotter colors indicate higher risk.  He noted that the San Andreas Fault is barely on the map; it is located west of the Delta.

The areas around San Francisco Bay have a very high risk; moving to the east are faults with much less activity.   “Some of the local faults in the Delta that are particularly important are Midland, Pittsburg, and Clayton,” he said.  “The activity of these faults is heavily debated.  There are interests within the Delta who do not want the seismic hazard to be a driver of decision making, and the arguments that have been put forward are that these faults are inactive, they are not going to produce earthquakes, and oftentimes the reasoning behind that is, “I’ve never felt an earthquake, I’ve been living in the Delta for 60 years, I haven’t felt anything.  You tell me there’s an earthquake problem, I don’t believe you.’”

We debate these things, and what we’re debating very often is, is the slip rate which is basically the rate with which one side of the fault moves relative to the other, high enough that it’s a problem or not?” he continued.  “The Napa earthquake which happened a few years ago on the west Napa fault; it turns out that the west Napa fault has the same slip rate, practically speaking, as the faults further east from the Delta, like the Pittsburg, Kirby Hills, Clayton, and so on.  So if you’re going to argue that there is no earthquake risk on faults on that type, then by extension there could not have been an earthquake on the west Napa fault.  From a scientific point of view, this makes no sense.  But this is the type of arguments that I and others have been involved with, with interests who do not want to recognize the seismic hazard in the Delta.”

Dr. Stewart noted that the map does not account for the effects of shallow soil conditions which amplify the ground shaking.

So if there is significant ground motion, what happens to the levees?  He said it’s useful to separate the cases where levees are continuously impounding water and acting more like dams from those that are intermittently impounding water.  He presented an image of the Delta that highlights the changes in relief to give a visual of how much retreat of the island interior has happened, noting that this is a continuous impoundment (lower, left).

In contrast, the levees along the American River in Sacramento are more typical (above, right).  In this case, the levee is high and dry most of the time; it’s only going to feel water during an extreme flood event.

What happens to levees when they are shaken?  “One possibility is that you have liquefaction; that’s where the soil is losing strength, and when it loses strength, you can have this kind of slumping or you could have a flow failure,” Dr. Stewart said.   “Keep in mind that the water level in the channel is usually about a meter below the crest, so if you’re losing height like this, you’re going to lose the island.”

That is by no means the only problem,” he continued.  “These levees sit on this very soft peat; that’s the organic soil.  That material can have varying failures: you can have levees partly sliding or you can even have the levee simply sinking into the ground as a result of volume change in that fill.”

Earthquake engineers predict ground motion and other risks, and once they have that, they next try to figure out fragility, Dr. Stewart said.  He presented a graph, noting that the x axis is the intensity of ground motion and the y axis is probability of failure.  Conceptually it’s pretty simple: If you don’t shake the ground very hard, the probability of something bad happening is small; if you shake the ground really hard, the probability of something bad happening is really high, maybe even 1.

It’s the whole relationship there as we go from we’re good to we’re bad, that’s called a fragility curve, and engineers want to be able to understand that relationship,” he said.  “In this case, we would be talking about what is the probability that the levee has enough deformation that it doesn’t retain the water anymore and it flows over the top?  So we and others are doing work to characterize things like that.

Japan has had a lot of recent earthquakes; there is significant documentation of how their levees have performed.  In particular, the percentage of time the levees do well and don’t do well is known, as well as how they responded to different levels of ground shaking.  He presented a fragility curve, noting that is based on actual field performance in Japan, although in this case, with intermittently loaded levees.

You can see that when you’re getting to ground motion levels of let’s say about 30 to 40 cm per second, you’re getting probabilities of failure of 20% or so, and that’s about what we’re looking at in the Delta, those ground motion levels.”

Dr. Stewart said they have done field tests and centrifuge tests to try to understand the mechanisms by which levees fail and to develop models for their performance.  One of those tests included building a model levee in the Delta and testing its strength by shaking it on top to see how the levee is deforming as the levee is developing inertia forces and moving back and forth and to understand the demands imposed upon the foundation soil by the vibrating levee.

Dr. Stewart said that they have been working to understand how the peat soil responds when it’s sheared.  When they imposed a deformation on it such as an earthquake would, what happens is that the peat generates water pressure, so the pressure within the voids is going up but much less than they thought.

As that pressure dissipates following the earthquake, there’s a little bit of volume change,” he said.  “But what surprised us in our research is that it went further; there was a lot more volume change than just the volume change from dissipating that water pressure.  That’s a mechanism that is a little bit complicated to explain, but it’s more or less a volumetric creep that is accelerating as a result of the disturbance from the earthquake.  And the significance of that is that while the levee might settle a little bit from the earthquake, it might then accelerate and settle catastrophically in the weeks following the earthquake.  And you only need to lose about a meter and the levee has failed.”

The outcome of all this is that we have pretty substantial risks to our levee systems in the Delta,” said Dr. Stewart.  “We’re going to lose some.  The question then is how resilient is it.  Remember that I started out saying it’s expected more or less that we’re going to have losses in earthquakes, but the question is can we recover?

Dr. Stewart then played a video from Metropolitan Water District that simulates what could possibly happen if an earthquake were to happen that would cause many levees to fail at the same time.  The video simulation shows eight days compressed to one minute, with the hotter colors indicating higher salinity.

That’s what those red dots are indicating, and they are all failing more or less at the same time, so if we look at a blowup of one of them, it’s settling, or having a stability failure, the water from the channel flows into the interior of the island,” he said.  “Now keep in mind we have three billion cubic meters of volume to fill up.  As the image pans out, you see that not just that island, but lots of other islands are failing because so many levees are failing throughout the system, and much of that 3 billion cubic meters of volume is going to be filling up.”

Only so much of the water to fill up the space will be freshwater coming from the mountains; you can open up the spigots in all the reservoirs but it can’t provide enough flow for that to be fresh water, he said.  The water will come in from San Francisco Bay, which is an endless source of water albeit saline water, he said.  If the saline water penetrates to Clifton Court and the export facilities, there isn’t a water source to flush the water out.

So what can you do?  “You can start repairing the levees, but you’ve got this giant lake in the middle of the island,” Dr. Stewart said.  “One storm comes along, you get some waves on that lake and it starts eroding the levee from the inside.  The levees don’t have any erosion protection in the inside; the water is supposed to be on the channel side, not on the interior, so now you have additional failures from erosion from the inside of all of these levees, so you repair one section, and you lose another one.

The problem becomes almost intractable,” he continued.  “Keep in mind, one failure took $100 million to fix, and now we’re looking at scores of failures, so the water managers for the state are petrified of this.  They are not sure they can ever get this system up and running, or at the very least, it’s going to take multiple years.  So this is pretty serious.”


Dr. Stewart then discussed what is being done to mitigate the risks.

With respect to the Los Angeles Aqueduct, LA Mayor Garcetti has undergone an effort to increase resilience in Los Angeles and has released the report, Resilience By Design.  The report addresses a number of earthquake problems within Los Angeles, and among them the LA Aqueduct and the risks to the Elizabeth tunnel.  The report proposes 15 different options for addressing the problem.

One option is to manage the water LA has locally better, including all of the wastewaters.  “We can take our wastewater, we can treat it, and we can basically pump it into the ground, usually near the coast, then we withdraw it further inland, the ground is naturally filtering it, we treat it again and we reuse it,” he said.  “There’s actually a lot of water that can be reclaimed that way, so we do have the ability with some expense to better use what’s already here.

The other option they are working on is putting an HDPE carrier pipe through Elizabeth tunnel; it wouldn’t be enough to supply LA but it’s better than nothing, he said.  “It’s extremely flexible so when the tunnel is offset by the earthquake, there can be at least some water still passing through there, so that is actually being done now.”

There is rather substantial water storage west of the fault in Diamond Valley Lake, which is owned by Metropolitan Water District.  Diamond Valley Lake and other reservoirs are meant to give months of supply, so that Southern California has something to live off in the event of an earthquake along the San Andreas Fault while the aqueducts are being repaired.

With the Bay Delta and the State Water Project, it’s trickier, he said.  “What the Governor has proposed is this twin tunnel system.  There is a good deal of political resistance to this from interests in and near the Delta.  There really is no viable fix, I would say, with the current levee system because of this continuous retreat of the interior islands, and we really can’t stop that.  You can try to repair the levees, but they keep getting worse year after year.

So the Governor’s proposal is to build tunnels that are draw water from Sacramento River and take the water down to Clifton Court.  These tunnels are not to be underestimated in their significance, they are 12 meter diameter tunnels, 48 kilometers long, costing estimated $15 billion and just to put that in perspective, these are bigger than the Big Dig and Boston and they are bigger than the tunnel going between England and France, so these are major engineering projects to try to deal with this problem.”


Dr. Stewart than gave his conclusions.  “We need a resilient water system,” he said.  “We can accept losses in the short term.  We do not want to see something like Katrina where we lose our water supply for an extended period of time, because we can’t survive as a society without it.  We can get by a few weeks; we can’t get by for a long time.  We could see a depopulation scenario unfold here, and that’s what we have to avoid.”

All our major water supply aqueducts cross the San Andreas Fault,” he said.  “It’s just the geography of the state.  There’s no way around it; that’s the reality we live with.  When we look at all those aqueducts and how they could be repaired after that event, there are varying degrees of resilience.  In every case, the owner recognizes the problem, and is trying to do something to prepare for it.”

There are additional hazards from local distribution systems, and we have pretty good experience from Northridge and other events that it’s damaging but we can recover,” he said.  “When we look at all these different aqueducts, I would argue that the most vulnerable is the California Water Project for the reasons I’ve explained; I should put in context that the San Andreas Fault earthquake has a 150 year return period, that’s pretty frequent in the earthquake world.  The earthquakes up in the Delta are much less frequent.  It could easily not happen in all of our lifetimes or our children’s lifetimes or our grandchildren’s lifetimes.  Or it could happen tomorrow, that’s the way risk works.  It’s a low probability event, it’s a lower probability than the San Andreas for sure, but it’s going to happen.  It’s just a question of how long.”

The State Water Project is essential, both from the volume standpoint as it provides a lot of our water, and from a water quality perspective, as the water quality is quite good from it.  As currently configured, the levees are highly vulnerable, not necessarily for San Andreas events but for the local events directly beneath.  The repair time is uncertain; it’s almost certainly very long.  They don’t even know how long it would take, and I think by any measure, it is not resilient, and this is the problem.”

There is some mitigation being planned,” he said.  “The Governor is trying to do something about it, but it’s costly and there is mixed levels of political support, so whether this project actually goes forward depends on who the next Governor is, what the priorities are, so we’ll see.  But at least now, I hope you have a better appreciation for the threat that we face.”

Thank you for your attention.”


Print Friendly, PDF & Email