‘How Many Ways Can the Potter Valley Project Fail?’: Friends of the Eel River

This is a press release from the Friends of the Eel River:

Scott Dam which is part of the Potter Valley Project. [Photo cropped by one from PG&E]

The Eel River dams will never produce hydropower again. The unreliable water supply they allow to the Russian River will fail as geology and physics proceed.

Pacific Gas and Electric’s (PG&E) Potter Valley Project has failed, forever, at the primary purpose for which Cape Horn and Scott Dams and the diversion to the East Branch Russian River were built. After the transformer at the Potter Valley powerhouse failed in 2021, PG&E declined to buy a new one. So the dams will never produce another watt of power.

Fortunately, this benefits PG&E’s embattled ratepayers. Between 2005 – 2016, the PVP generated less than ¼ of a percent of all PG&E’s hydroelectric production, yet the PVP cost PG&E (or rather, ratepayers) more than twenty dollars for every dollar’s worth of electricity it made. PG&E cites the significant economic losses of the project as a primary factor in their 2019 decision to withdraw their relicensing application, which led directly to current plans for dam removal.

But, while the economic losses started PG&E on the path toward decommissioning, it’s clear that dam safety issues at this century-old, high-hazard facility are what is really motivating PG&E to act quickly to remove this massive liability.

Seismic Risk is Amplified by Structural Issues

Scott Dam, which impounds the Lake Pillsbury Reservoir, sits nearly atop the Bartlett Springs Fault Zone, capable of generating a M7 earthquake. Many elements of Scott Dam’s design and current condition compound the seismic risks of its location.

Congressman Huffman recently shared at a public meeting that an engineer with PG&E told him that of all the utility’s projects “Scott Dam is the one that keeps me up at night”. Here is a summary of the most concerning elements at Scott Dam fueling the nightmares of PG&E’s engineers:

a. Design v Construction

Scott Dam was originally designed to go straight across the river. During construction, however, builders discovered that what they thought was bedrock on the southern abutment was actually a giant boulder. This boulder shifted during construction and required a seat-of-the-pants redesign to build the rest of the dam in front of the boulder (nicknamed “the knocker,” it’s the dark purple blob in the image below), at a sharp angle to the rest of the dam.

b. Sediment accumulation

The Eel “has the highest recorded average annual suspended-sediment yield per square mile of drainage area of any river of its size or larger in the United States. This yield, in tons per square mile, is more than 15 times that of the Mississippi River and more than four times that of the Colorado River”. This sediment has been piling up, in the Lake Pillsbury Reservoir, for over a century, and is now placing significant pressure on the upstream face of Scott Dam.

These walls of sediment will at some point collapse and block the only low-level water outlet. When that water outlet, controlled by a needle valve, stops working, PG&E will only be able to release water when Scott Dam is full. The risk of sediment collapse is greater when sediments are exposed, and when reservoir levels are drawn down rapidly. Since this problem was identified, PG&E has managed the reservoir to always maintain at least 12,000 acre feet of water.

c. Aging infrastructure

Although any century-old infrastructure is going to degrade and pose an increasing risk of failure, dams are particularly subject to decay and especially vulnerable to catastrophic failure. However, the Federal Energy Regulatory Commission (FERC) flatly refused to address questions about the seismic safety of Scott Dam in the relicensing process, insisting that its Dam Safety division’s inspection process keeps all federally licensed dams safe, by definition.

Similarly, California has its own Division of Safety of Dams (DSOD), with significantly more capacity and authority than most state level bodies. This system gives the outward assurance of regular review by experts, but it is nearly impossible for the public to track this process.

Nearly everything about dam safety in the PVP FERC docket is classified as Critical Energy Infrastructure Information (CEII), a designation that keeps all such information out of public view. Over years of reading the unclassified parts of correspondence between PG&E and regulators on the FERC docket, we caught some glimpses through the CEII fence of Scott Dam’s deeper challenges.

d. Foundation

As Scott Dam was being built, contemporary observers raised what appeared to be serious questions about the materials and methods used to construct the dam’s foundation, but no investigation was ever undertaken. Again, much of the information about the stability of Scott Dam’s foundation is concealed behind CEII classification. What we do know is that many of the piezometers installed to measure uplift pressure have failed. FERC recommended that PG&E install new ones during routine dam safety evaluations in 2018, but we are unsure if that ever happened.

e. Seismic risk

Many dams built before we understood plate tectonics sit on fault lines because where river channels cross side-slip faults, they are pinched into configurations ideal for a small dam to create a large reservoir. This is exactly the case for Scott Dam. The Bartlett Springs Fault is a part of the San Andreas complex has been the focus of decades of work by USGS geologists, which by 2015 had begun to show that it is capable of generating up to a M7 earthquake. That new information slowly percolated into FERC’s systems, ultimately generating a new estimate of the potential maximum earthquake for Scott Dam.

In addition to Scott Dam’s precarious location atop the Bartlett Spring Fault system, there is also an active landslide above to the southern abutment of the dam (where “the knocker” is located). When Miller Pacific conducted a slope stability analysis in 2018, they concluded that the landslide, with a mass of over 8 million cubic feet, weighing over 520,000 tons, presents a significant geologic hazard. In PG&E’s 2016 safety review they state that the “susceptibility of these slopes to seismic events is not known and has not been studied.”

Failure as water supply infrastructure

When PG&E received an assessment of the seismic risk to Scott Dam in 2023, they quickly decided to mitigate that risk by lowering the radial gates atop the dam and keeping them down, reducing the capacity of Lake Pillsbury Reservoir by about 20,000 acre feet. Combine that with the sediment accumulation which has both reduced storage capacity and requires PG&E to maintain at least 12,000 acre feet to prevent blockage of the only water outlet, and that leaves relatively little water to spare for diversions. What had previously been a significant asset for water users is now at best uncertain, and very much at risk of complete failure.

When PG&E attempted to auction the PVP, they received no qualified bids. Representatives of water users in Sonoma and Mendocino county are clear that this is due to the liability of the dams and the annual operating losses. They say, “there is no legal basis for requiring PG&E to maintain the dams…and we cannot operationally or fiscally take ownership of or fix both dams.”

Thanks to the cooperation of stakeholders in both river basins, transforming the Eel into California’s longest free-flowing river will not end diversions into the Russian, but rather allow for a change that will be more ecologically appropriate for both watersheds. Dam removal is the inevitable solution to the public safety and environmental hazard that the current project presents. And let’s not forget, dam removal is also the single most important action we can take to support recovery of the Eel’s native salmon and steelhead.