During the past five years, while Californians have been subjected to increasingly panicked predictions about an earthquake causing levee failures in the California Delta, the U. S. Bureau of Reclamation (USBR) has been doing a Corrective Action Study to figure out how to prevent an earthquake-caused dam failure that “could inundate hundreds of square miles including the town of Santa Nella, about 7 miles of Interstate 5, and numerous farms and houses along the San Joaquin River including some areas of Stockton.”
We’re thinking this dam failure couldn’t be too good for the California Aqueduct and the Delta-Mendota Canal either. It would definitely impact state and federal water project deliveries, causing major water problems in California.
The dam in question is B. F. Sisk – the dam that impounds 
the San Luis Reservoir. This dam, it turns out, “is located in a seismically active area, very close to the Ortigalita Fault.” Click here to see the letter from USBR to the Central Valley Project (CVP) water contractors served by the San Luis project. One reason we haven’t heard about this problem may be that USBR asked water agencies to sign non-disclosure statements, using the rationale that there is “procurement sensitive” information that could affect the project.
The Ortigalita Fault goes directly under the San Luis Reservoir. There are other active faults nearby, including the San Andreas and the Calaveras. USBR and the Department of Water Resources (DWR) have apparently used different modeling to predict risk of failure at San Luis than they used to predict levee failure in the Delta Risk Management Study. The modeling has given them a low risk of failure for B. F. Sisk Dam.
In the case of the Delta, USBR and DWR based predictions on movement of the Midland Fault, a blind fault under the Delta that has had no measurable seismic activity for thousands of years. Comparison of numbers in one U.C. Berkeley report suggests that the Ortigalita fault slip rate could be 5 times greater than the slip rate on the Midland fault.
Researchers have turned up other vulnerabilities with B. F. Sisk, a 300-foot-high, 3 ½ mile long compacted earthfill dam that is sort of like a really BIG levee. In September 1981, a rapid drawdown of the reservoir triggered a huge slide in which 400,000 cubic yards of embankment slid down 177 feet along a 1,100 foot section near the crest of the dam. Repairs were made, but three years later, a crack opened along the embankment, parallel to the dam’s centerline. It eventually stopped, and no other movement or cracks have been reported since.
Also in 1981, in May, about 1,000 feet of canal lining cracked and slumped into the California Aqueduct near mile 10 (south of Bethany Reservoir). 
Pumping at the Banks Pumping Plant had to be curtailed until the end of July. An emergency diversion permit was obtained from the State Water Resources Control Board on May 22 to allow the pumping of SWP entitlements at the USBR Tracy Pumping Plant while aqueduct repairs were made.
The takeaway message here: California’s entire water transfer infrastructure and much of its storage is vulnerable to seismic activity and various kinds of failures. If the point is really to protect water deliveries for urban users and economic activity, it is short-sighted and unwise for the State and federal governments to focus its attention only on the vulnerability of Delta levees.
Delta levees should be upgraded to the highest possible standards to protect people in the Delta and the Delta economy. The entire state water system, however, is subject to the threat of earthquake. That is why it makes more sense to support regional water projects that improved local water supplies instead of depending on water that is moved hundreds of mile over fault lines.