A recent study by Stanford and MIT argues that there are potential benefits to extending the operating life of PG&E’s Diablo Canyon Nuclear Power Plant beyond its scheduled 2014/2025 retirement. The Stanford/MIT study finds benefits in Diablo Canyon’s largely carbon-free generation of electricity, and its potential to provide power for water desalination and hydrogen production. The study tries to address many of the physical and financial obstacles to extending the operating life of Diablo Canyon, but my initial impression was that it had flaws or omissions that rendered its conclusions too rosy.
Before publicly pointing out the problems I saw, I contacted the authors of the study to make sure I wasn’t missing anything. They responded promptly and politely to my questions, but their answers only partially reassured me. That said, there are definite limits to my expertise and to the depth of my analysis of the study. First, I am trained as an attorney, not as an engineer or scientist, and my knowledge of Diablo Canyon comes from my work at the California Public Utilities Commission (CPUC) relating to PG&E’s application to retire Diablo Canyon. Second, I was the Administrative Law Judge (ALJ) that heard PG&E’s application and wrote the CPUC decision approving the retirement of Diablo Canyon. I believe that decision was correct at the time, and that it is still correct. Things have not changed so significantly since that time; climate change was as real then as it is now. Finally, please note that my comments here are solely my own in my personal capacity – they are not on behalf of the CPUC or any division of the CPUC, and I have had no assistance from CPUC technical staff. I have not addressed every potential issue raised by the study, nor have I analyzed any of the cost figures in the study, but I have just focused on a few areas where I have some knowledge.
When I read the study, three potential problems immediately jumped out. 1) The study includes scenarios in which Diablo Canyon is assumed to be able to ramp up and down by as much as 50% to 100%. This is inconsistent with PG&E’s testimony before the CPUC. 2) The study did not appear to account for the capital cost expenditures necessary to continue operating the plant for another 20 years. 3) One of the water desalination scenarios used by the study did not appear to account for the true cost of installing a large water pipeline to the Central Valley.
I asked the authors of the study about these three issues; they clarified that the capital costs of continuing the plant in operation were included, in the amount of $145 million per year. This answered my question regarding issue #2 above, but I found their answers on my other two issues to be less satisfactory.
The biggest flaw I noticed was that the study used three long-term scenarios that used varying assumptions about how Diablo Canyon would operate going forward. As the study describes it, “The variation is based on the possibility that Diablo Canyon will provide electricity to the grid more flexibly in the future, either by ramping its generation up or down, or by allocating some fraction of that generation to such non-grid applications as H2 production or desalination.” The three operating patterns used by the study are “must run, 50% flexible, and 100% flexible.” The must-run scenario is how Diablo Canyon operates now, as a baseload facility that does not ramp up and down. The study describes the other two scenarios as follows: “The 50% flexible scenario allows Diablo Canyon to operate as low as 50% minimum load, with the ability to ramp up or down by 50% of its full capacity in an hour. The 100% flexible scenario allows Diablo Canyon to run without a partial load limit or ramp-up limitation.” (Study at 23-24.)
My recollection was that PG&E stated to the CPUC that Diablo Canyon could not feasibly ramp up and down, so I conveyed that information to the authors of the report and asked them where they got the idea that Diablo Canyon could ramp up and down. Their response was that Diablo Canyon’s current license provides for ramping, and that other nuclear plants worldwide could also do that. PG&E’s position before the CPUC did not cause the authors to change their conclusion.
Here is what PG&E told the CPUC:
“CGNP [Californians for Green Nuclear Power] states that flexible operation of DCPP is “certainly possible” and “might directly help mitigate overgeneration” but it offers no evidence to support this conclusion. The evidence in the record is that Diablo Canyon was licensed and is designed to operate at full power (e.g., baseload). Operating in load-following mode would take Diablo Canyon outside of the currently authorized NRC license conditions and would require extensive technical feasibility studies, redesign of procedures, processes and systems, maintenance practices and nuclear fuel redesign. While there are a few pressurized water reactors (“PWRs”) in Europe that perform daily load-following, these plants were specifically optimized and designed for this purpose. The United States fleet of PWRs do not operate in this mode (with a few exceptions where plants are capable of making minor ramping changes in plant output). It is unclear if Diablo Canyon could be retrofitted to safely and reliably operate in a different operating mode, whether the NRC would approve it, and whether it would be cost-effective to do so given the reduction in capacity factor that would result if Diablo Canyon were to be frequently ramped down to minimum operating levels during the daytime hours when solar power is prevalent.” (PG&E Reply Brief at 7, in proceeding A.16-08-006, footnotes omitted.)
From my perspective, given the unequivocal statement of PG&E to the CPUC and the age of Diablo Canyon, I am not persuaded that Diablo Canyon could ramp up and down by 50% or 100% just because its license might allow that and because other plants might have that capability.
The other issue I noticed relates to the proposal to build a large desalination plant at Diablo Canyon. The study looked at two scenarios; one scenario called for installing a pipeline to connect the desalination plant to the Coastal Branch of the California Aqueduct near San Luis Obispo, not far from Diablo Canyon, which seems potentially feasible. The other scenario, however, is described as: “[A] shallow buried pipeline will convey product water from Diablo Canyon all the way to California’s Central Valley, where water demand from farmers is outstripping supply. This pipeline would be about 100 km long, with a net elevation gain of 100 m.” The pipeline would have an inside diameter of 3.5 meters, and based on quotes from two pipe vendors, the study estimates a total cost of $2.2B for the 100 km pipeline. That cost figure includes, “manufacturing, transportation and installation.” (Study at 69-71.)
Given the topography between Diablo Canyon and the Central Valley, I questioned the study’s apparent omission of the costs of permitting, acquiring land or rights-of-way, and more detailed construction and installation costs, which would appear to be significant. In response, the authors of the report stated: “[I]f a very large desalination plant was deployed at Diablo Canyon and its freshwater was to be delivered to the Central Valley, the new pipeline would follow the exact same course of the existing Coastal Branch pipeline. In fact, it is likely that the very same pipes could be used (in reverse flow direction) for large segments of the pipeline. Therefore, we estimated that the permitting costs would be low.”
While this approach of using the existing route in reverse (which was not disclosed in the study) could mitigate some right-of-way acquisition issues, it is not clear that it would be as simple or low-cost as the authors appear to be assume, especially taking into consideration the potential ramifications of reversing the flow of part of the state water system. I remain unconvinced that this would be as cheap or easy as the study seems to assume.
There are few other barriers that the study either doesn’t address or that it glosses over. Renewing the NRC license for Diablo Canyon would obviously be essential; the study believes that this process could be done relatively quickly (less than two years) because of the license renewal work previously done, and that the plant could stay in operation while that review is underway.
This seems a bit optimistic, particularly since PG&E has not re-started the NRC relicensing process or shown any inclination to do so. I believe that PG&E is unlikely to initiate the NRC relicensing process absent a clear indication from the CPUC that PG&E will be able to get rate recovery for those costs. This would require a re-opening and modification of the CPUC’s decision approving the retirement of the plant, which included a settlement of the amount of PG&E’s recovery of NRC licensing costs (signed by PG&E). Modification of that decision would be vigorously litigated by consumer and anti-nuclear groups (and probably PG&E itself), and would not likely be a quick process. Until that process is completed with a clear outcome, PG&E will be unlikely to want to spend any money on NRC relicensing.
There is also a potentially problematic state statute. California Senate Bill (SB) 1090 (2018) put into place Public Utilities Code section 712.7, and required modifications to the CPUC’s first decision approving the retirement of Diablo Canyon. While SB 1090 does not expressly codify the retirement of Diablo Canyon, the reality of that retirement is the basis for the substance of the bill, which required the Commission to provide substantially more ratepayer funding for the local community and plant workers to compensate for the retirement of the plant. Ratepayers are already paying those costs. Reversing course from the reality relied upon by SB 1090 would likely result in increased litigation risk.
The study acknowledges the significant financial and management challenges currently facing PG&E, and how those might make PG&E reluctant to take on another major project. In response, the study then proposes a number of possible other plant ownership models, such as a sale to another utility or a public entity, but without evaluating the feasibility of those options or the potential transaction costs. Any sale of Diablo Canyon would require prior approval of the CPUC under Public Utilities Code section 851, and would involve multiple contentious issues, which the study does not address. For example, a proposed sale of Diablo Canyon could trigger a review of the environmental impacts of that sale under the California Environmental Quality Act (CEQA), and the construction of additional facilities (such as those necessary for water desalination or hydrogen production) would definitely require a CEQA review. But the study does not address CEQA.
Accordingly, the study paints the potential for achieving its preferred outcome in too optimistic of a light. Even though the study acknowledges that there are substantial challenges to its proposal, it understates those challenges, including many that I have not addressed here. As a theoretical exercise, the study raises some interesting ideas, but as someone once noted: “The gap between theory and practice is wider in practice than it is in theory.” That gap looks particularly wide here.
Posts
