Faults
Seismic activity resulting from the sudden release of energy along the surface of faults is inevitable in this part of California, and the potential for significant earthquakes is present. In general, understanding fault systems is difficult, especially without seismic surveys. California's strike-slip system is especially challenging with components of motion along every axis. Moreover, faults may exist both onshore and offshore without surficial expression.
In short, faults are hard to see, hard to understand, hard to predict, and hard to characterize.
Measurement
Most of us are familiar with the Richter scale, a log scale where anything above a 7 is damaging and anything above an 8 is potentially catastrophic, depending on proximity and other factors.
Important to understand is the relationship (or lack thereof) between the Richter scale and another measurement called peak ground acceleration (PGA, expressed in units of acceleration as a percentage of gravity, g), and how those create further uncertainty in safety margins.
Higher magnitude events tend to product mode ground acceleration, but factors such as sediment composition and other geological factors muddle the relationship. Engineering parameters designed around PGA have a loose correlation to magnitude. A shortcoming of PGA is that it only records a value, not direction or duration. Structures can only withstand a certain PGA for a certain duration.
Furthermore, structure damage is better characterized by PGV, peak ground velocity, measured in units of cm/s. Below is a chart adapted from two Wikipedia charts to relate PGV to Richter.
Magnitude | Description | Peak Ground Velocity (cm/s) | Effects | Avg global freq. |
7.0–7.9 | Major | 20 – 41.4 or higher | Causes damage to most buildings, some to partially or completely collapse or receive severe damage. Well-designed structures are likely to receive damage. Felt across great distances with major damage mostly limited to 250 km from the epicenter. | 10 to 20 per year |
8.0–8.9 | Great | 41.4 – 85.8 or higher | Major damage to buildings, and structures likely to be destroyed. Will cause moderate to heavy damage to sturdy or earthquake-resistant buildings. Damaging in large areas. Felt in extremely large regions. | One per year |
9.0–9.9 | Extreme | Up to and beyond 178 | Near total destruction – severe damage or collapse to all buildings. Heavy damage and shaking extend to distant locations. Permanent changes in ground topography. | One to three per century |
Geoscience and Engineering Intersect
Diablo Canyon can
supposedly withstand a PGA of 0.75g and magnitude of 7.5. However, information is lacking on what duration it can sustain, and in which component directions. Information is also lacking on PGV.
There are significant discrepancies in PG&E's methods of determination of Diablo Canyon's PGA ratings, and the assumptions and methods are more relaxed than Federal Nuclear Regulatory Commission (NRC) guidelines (
NRC 2011). There has in fact been no rigorous regulatory validation of 0.75g or 0.83g. The processes of these ratings were not applied according to NRC standards, and the NRC has allowed the plant to continue in spite of this fact. Moreover, a larger magnitude or lower-PGA/higher duration earthquake is possible in this area, regardless of the actual nominal PGA rating of the facility.
The strike-slip fault system near to Diablo Canyon is potentially insufficiently understood given the proximity of several of the faults, notably Hosgri and Shoreline, the latter of which was only discovered less than a mile away in 2008, a full 40 years after the construction of the facility.
Hardebeck et al. (2010):
The Central Coast is bounded on the east by the San Andreas Fault, the major plate boundary fault, and lies between the greater San Francisco and Los Angeles areas. This coastal region is not as densely instrumented or as tectonically well understood as the San Andreas Fault or the major urban areas.
The identification of new faults, and the reinterpretation of known faults, suggests that further work is necessary to better constrain the seismic hazards of the Central Coast. While the locations and focal mechanisms (the direction of slip in an earthquake and the orientation of the fault on which it occurs) for aftershocks of the 2003 M6.5 San Simeon and 2004 M6.0 Parkfield earthquakes are similar to those found in previous aftershock studies, the seismicity features in the offshore region near San Luis Obispo are sharpened considerably by this study.
The most prominent newly-observed feature is the Shoreline Fault, a ~25 km-long vertical strike-slip fault running parallel to the coastline just offshore of Point Buchon. Several smaller strike-slip seismicity lineations are also observed in Estero Bay, along with a deep reverse structure at the depth of the top of the remnant subducted slab. Strike-slip faulting is observed along the Hosgri-San Simeon Fault system, up to ~10-15 km inland from the Hosgri Fault in Estero Bay and near Point Buchon, and on the onshore Rinconada and West Huasna Faults.
The Shoreline Fault in particular requires further study to better constrain its geometry, how it may connect to the Hosgri Fault or other faults to its east, its slip rate and whether it has produced large earthquakes in the past.
The operating company, PG&E conducted an
audit and reported to the NRC in 2011 that the magnitude of potential earthquakes posed by the newly-discovered Shoreline fault were below the threshold of concern for the facility.
We are partnering with the United States Geological Survey (USGS) to update the earthquake hazards along the Central Coast and throughout our service territory. We also study significant global seismic events and apply those lessons learned against the design criteria of Diablo Canyon to verify the basis of its design. Those efforts resulted in the discovery of the Shoreline Fault in 2008. Using updated models of the ground motions gives a ground motion of 0.56 g from the Shoreline fault. PG&E’s evaluation concluded that the existing plant design is adequate to accommodate Shoreline Fault ground motion.
Note the citation of a single PGA value without a duration or direction, and no PGV.
However,
Hardebeck (2013) revealed that neither the Shoreline fault nor the greater Shoreline-Hosgri system were sufficiently understood, adding that there is the potential a 7.5 magnitude event, which would potentially exceed even PG&E's optimistic PGA ratings:
The Shoreline fault is a vertical strike‐slip fault running along the coastline near San Luis Obispo, California. Much is unknown about the Shoreline fault, including its slip rate and the details of its geometry.
Here, I study the geometry of the Shoreline fault at seismogenic depth, as well as the adjacent section of the offshore Hosgri fault, using seismicity relocations and earthquake focal mechanisms. The Optimal Anisotropic Dynamic Clustering (OADC) algorithm (Ouillon et al., 2008) is used to objectively identify the simplest planar fault geometry that fits all of the earthquakes to within their location uncertainty.
The OADC results show that the Shoreline fault is a single continuous structure that connects to the Hosgri fault. Discontinuities smaller than about 1 km may be undetected, but would be too small to be barriers to earthquake rupture. The Hosgri fault dips steeply to the east, while the Shoreline fault is essentially vertical, so the Hosgri fault dips towards and under the Shoreline fault as the two faults approach their intersection. The focal mechanisms generally agree with pure right‐lateral strike‐slip on the OADC planes, but suggest a non‐planar Hosgri fault or another structure underlying the northern Shoreline fault.
The Shoreline fault most likely transfers strike‐slip motion between the Hosgri fault and other faults of the Pacific–North America plate boundary system to the east. A hypothetical earthquake rupturing the entire known length of the Shoreline fault would have a moment magnitude of 6.4–6.8. A hypothetical earthquake rupturing the Shoreline fault and the section of the Hosgri fault north of the Hosgri–Shoreline junction would have a moment magnitude of 7.2–7.5.
PG&E's own
latest assessment admits a larger earthquake is possible which would exceed their own limits:
The largest earthquake considered in the SSC model is a magnitude M 8.5 on the Hosgri fault source, representing an extremely rare, but plausible, rupture between the offshore Point Arguello south of DCPP and the Mendocino Triple Junction offshore Cape Mendocino in northern California. The postulated rupture would include the entire 410 km (255 mi) length of the Hosgri-San Simeon-San Gregario fault zone and an additional 330 km ( 205 mi) of the northern San Andreas fault north of San Francisco.
Diablo Canyon was scheduled in 2016 to shut down starting in 2024, but due to the energy needs of California, it is planned to remain online. As recently as 2023 a
5-year extension was accepted:
The NRC has determined that the granting of the exemption request involves no significant hazards consideration because allowing the submittal of the license renewal application less than 5 years before the expiration of the existing license and deeming the license in timely renewal under 10 CFR 2.109(b) does not (1) involve a significant increase in the probability or consequences of an accident previously evaluated; or (2) create the possibility of a new or different kind of accident from any accident previously evaluated; or (3) involve a significant reduction in a margin of safety.
The reasoning here is essentially, "it's very unlikely anything bad will happen in the next 5 years, and at least we're not making it less safe." This reasoning is not altogether unsound, but is borne of necessity, and brushes aside the geological evidence that the risk is higher than previously thought.
A
thorough report from the Union of Concerned Scientists summarizes the issue:
Dr. Michael Peck, then an NRC resident inspector at Diablo Canyon, pointed out numerous deficiencies in PG&E’s evaluation of the Shoreline fault. Peck concluded that more analysis and likely additional modifications would be necessary before anyone could honestly claim that Diablo Canyon was adequately protected from an earthquake originating along the Shoreline fault.
Even if the Diablo Canyon reactors can in fact withstand the level of earthquakes PG&E asserts they can (0.75g), NRC analysis shows that there is roughly a 1-in-6 chance that the reactors will experience an earthquake larger than that over their 40-year lifetime. This suggests that even if the reactors are capable of withstanding 0.75g of ground motion, that may still be inadequate to ensure public safety.
It is also
reported that the Federal Nuclear Regulatory Commission recognizes the relatively recent geological hazards and will likely require further upgrades to the plant's mitigation strategy:
"To continue operating Diablo Canyon beyond 2025 would have required a license renewal from the Federal Nuclear Regulatory Commission," CPUC spokesperson Terrie Prosper said in an email. "As part of the renewal PG&E would need to make seismic upgrades. Those upgrades combined with required changes to the cooling systems to comply with state and federal water quality laws would likely cost more than $1 billion."
Summary so Far
There is a lot more to touch on, but I am running out of time tonight. I personally think there is indeed cause for significant concern at Diablo Canyon, and that the necessity of the plant is allowing regulatory laxity in the face of increasing realization of geological risk.
Note, I am not opposed to nuclear in general. I believe all energy sources come with risk that must be mitigated and tolerated to some degree. There is always a trade-off when it comes to energy, and the tolerance for risk has to be a balance between the benefits and the costs. Accurate assessment of risk and cost, however, can be extremely challenging in the face of economic and pragmatic reality.