Energy Policy xxx (xxxx) xxx
Contents lists available at ScienceDirect
Energy Policy journal homepage: http://www.elsevier.com/locate/enpol
Unmanaged climate risks to spent fuel from U.S. nuclear power plants: The case of sea-level rise Lisa Martine Jenkins, Robert Alvarez, Sarah Marie Jordaan * Johns Hopkins University, School of Advanced International Studies, 1740 Massachusetts Ave. NW, Washington, DC, 20036, United States
A R T I C L E I N F O
A B S T R A C T
Keywords: Climate change Nuclear power Cooling pools Spent fuel management Independent spent fuel storage installations Sea-level rise
Climate change and its accompanying sea-level rise is set to create risks to the United States’ stockpile of spent nuclear fuel, which results largely from nuclear power. Coastal spent fuel management facilities are vulnerable to unanticipated environmental events, as evidenced by the 2011 tsunami-related flooding at the Fukushima plant in Japan. We examine how policy-makers can manage climate risks posed to the coastal storage of radioactive materials, and identify the coastal spent fuel storage sites that will be most vulnerable to sea-level rise. A geo spatial analysis of coastal sites shows that with six feet of sea-level rise, seven spent fuel sites will be juxtaposed by seawater. Of those, three will be near or completely surrounded by water, and should be considered a priority for mitigation: Humboldt Bay (California), Turkey Point (Florida), and Crystal River (Florida). To ensure policymakers manage such climate risks, a risk management approach is proposed. Further, we recommend that policymakers 1) transfer overdue spent fuel from cooling pools to dry casks, particularly where located in high risk sites; 2) develop a long-term and comprehensive storage plan that is less vulnerable to climate change; and 3) encourage international nuclear treaties and standards to take climate change into account.
1. Introduction As of 2017, nuclear power continues to contribute 20 percent of U.S. electricity supply with 805 Gigawatt-hours of electricity over the year (US Energy Information Administration, 2018). Despite its high capital costs, nuclear power promises a technologically-proven, low-carbon source of electricity, capable of mitigating greenhouse gas emissions from the power sector while simultaneously providing for much of a grid’s needs (Kopytko and Perkins, 2011). However, nuclear power faces questions of economic viability and public perception, especially in the post-Fukushima world. Many existing large light water reactors have already closed in the United States, largely due to economic pressures and an aging fleet (Morgan et al., 2018). Some pundits are pushing for small modular reactors to be used as part of the solution, but this narrative – and the narrative of nuclear energy in the U.S. more broadly – ignores the back-end of the process. Nuclear power necessitates risk management over the life cycle of projects, from materials extraction and processing through use and waste disposal. Even over 60 years since the first nuclear power plant (NPP) became operational, the United States still lacks a long-term nuclear waste management solution, and therefore must bear the risks of the present management regime – a
regime that was developed as a temporary solution without consider ation of longer-term risks from climate change (American Society of Mechanical Engineers, 2019). The need to develop a long-term nuclear storage plan for the United States is well recognized by many experts, institutions, and organiza tions, including the U.S. Department of Energy (DOE) and the Nuclear Energy Institute (NEI) (Department of Energy, 2019; Nuclear Energy Institute, 2019). The total inventory of used fuel in the US is approxi mately 80,000 metric tons, which – lacking a central repository – is stored at NPPs and independent spent fuel storage installations (ISFSIs) across the country (Nuclear Energy Institute, 2019). And in many nu clear power plants, spent fuel is stored onsite in cooling pools. According to the DOE, as of the end of 2016, all operating NPPs are storing spent fuel in onsite spent fuel pools, and over half are also storing it in ISFSIs (Department of Energy et al., 2017). Those pools were meant to be a short-term solution; after cooling (which typically takes five years), the fuel was to be transferred to dry casks (U.S. Nuclear Regulatory Com mission, 2012). However, in many cases, fuel is left in cooling pools for much longer than anticipated, potentially imposing risks near genera tion sites. Collectively, these cooling pool sites contain some of the largest concentrations of radioactivity on Earth (Alvarez, 2012).
* Corresponding author. E-mail addresses: [email protected]
(L.M. Jenkins), [email protected]
(R. Alvarez), [email protected]
(S.M. Jordaan). https://doi.org/10.1016/j.enpol.2019.111106 Received 18 May 2019; Received in revised form 6 November 2019; Accepted 10 November 2019 0301-4215/© 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: Lisa Martine Jenkins, Energy Policy, https://doi.org/10.1016/j.enpol.2019.111106
L.M. Jenkins et al.
Energy Policy xxx (xxxx) xxx
Per the Nuclear Waste Policy Act of 1982 (NWPA), the U.S. Department of Energy is responsible for developing a geologic re pository for used nuclear fuel (Environmental Protection Agency, 1982). The original plans were for a centralized geological storage facility, proposed to be developed at Yucca Mountain in Nevada. In January 2010, the Obama administration canceled the plan. In Yucca’s stead, a Blue Ribbon Commission was established to find a storage site with local consent (as Finland, France, Spain, and Sweden managed to do) (Wald, 2012). Subsequently, a coalition of multiple utility industry groups released a statement urging the commission to make quick work of their task, especially after Fukushima. The impasse had caused tangible set-backs for the nuclear energy industry; in fact, nine states (including California) banned constructing new reactors until progress has been made on spent fuel management (Wald, 2012). The Fukushima disaster led to increasing scrutiny on the yet-to-be-resolved U.S. nuclear indus try’s spent fuel regime. In the nine years since Yucca was cancelled, concrete long-term plans have yet to be developed. Policy-makers must now address a persistent question: how will the U.S. manage its increasing stock of radioactive waste in a way that accounts for the long-term residence of nuclear waste while mitigating unanticipated longer-term risks? Previous research has indicated that coastal sites present greater climate change-related safety concerns than their inland counterparts, and that they will need to adapt to avoid costs to the environment and public health and welfare (Kopytko and Perkins, 2011). Most of the spent fuel is stored near electricity generation sites in the United States, our analysis focuses on thirteen which are located near seacoasts (using the sea’s cold water to cool the reactors). The goal of this research is to evaluate whether coastal nuclear power plants may be subject to increasing risks due to climate impacts, namely sea-level rise. Climate change is already causing coastlines to change due to sealevel rise, but—to our knowledge—no research has been completed about the impact of that change on coastal spent fuel storage. According to the Intergovernmental Panel on Climate Change (IPCC), global tem peratures are likely to rise anywhere from 1.9 to 5.4 � C by 2100, relative to 1850–1900 levels (Masson-Delmotte et al., 2018). This could cause a range of consequences, but we focus specifically on those related to sea-level rise. The U.S. National Oceanic and Atmospheric Administra tion (NOAA) explains that the most realistic of these warming estimates capture anywhere from 0.3 m (approximately 1.0 foot) to 2.5 m (approximately 8.2 feet) of sea-level rise by 2100 (Sweet et al., 2017). In parts of the Northeast Atlantic and the western Gulf of Mexico, sea-level rise is anticipated to be higher, and in much of the Pacific Northwest and Alaska, it is expected to be lower. Compared to the rest of the world, the U.S. is especially vulnerable in the scenarios that capture the higher range of sea-level rise (Sweet et al., 2017). According to the latest data from the IPCC, most humans live near the coasts. Coastal NPPs and ISFSIs are examples of vulnerable infra structure that form just one small part of a complex coastal system that requires systematic evaluation and management of future risks. The IPCC emphasized that, while some sea-level rise is an inevitability, slower climate change means slower sea-level rise, which would reduce these risks and allow for adaptation in the human and ecological systems impacted (Masson-Delmotte et al., 2018). This research examines one important facet of that adaptation: how can policy-makers manage the risk posed by coastal storage of radioactive materials? Our analysis begins with an evaluation of whether sea-level rise may create new risks for spent fuel management. We then discuss the im plications and propose solutions for mitigating these and other climate risks. We aim to present policy-makers with a framework for this eval uation, but also recommend a long-term solution to mitigate climate risks. Three key questions are central to this paper:
� What are policy options for managing the risk posed by these potentially exposed spent fuel management (SFM) facilities? � What are broader policy options for the U.S.’ spent fuel management regime? To do so, we used geospatial mapping software (QGIS) and govern mental datasets to overlay maps of present coastlines, maps of sea-level rise, and satellite images of coastal nuclear facilities and measured the change in coastline to determine water encroachment on the sites. Sites were then categorized as low, medium, and high risk using a qualitative ranking system. In light of the findings, we have developed recom mendations for site-level management but also more rigorous consid eration of climate risks in the development of spent fuel management. 2. Background Recent events have informed this analysis, providing perspective on the risks that coastal NPPs may experience. Regardless of whether disaster actually strikes, the task of managing spent fuel has proved intractable; despite the mandate of the 1982 NWPA, the current regime has left long-term risks for spent nuclear fuel management. This has long-term implications for both environmental and human health if left unmitigated on the decadal time scale and, crucially, developing a long term spent fuel management plan has experienced delays on these time scales. While the most prominent example of a seacoast event – the March 2011 Fukushima disaster – is is not a direct consquence of climate change, it is related to the combination of a NPP’s proximity to the ocean confounded by inadequate risk management of environmental risks. 2.1. Fukushima The fate of the Fukushima nuclear site illustrates a case where an environmental event snowballed, illuminating unforeseen and thus un mitigated environmental risks. When a 9.0 magnitude earthquake struck off Japan’s coast, it prompted a 46-foot tsunami, which crashed directly on the Dai-Ichi nuclear power site. Within a few hours, offsite power and backup diesel generators were both rendered inoperable, and the cool ing infrastructure – wiring, pipes, and pumps for four reactors and fuel storage pools – was severely damaged (Alvarez, 2012). This in turn prompted a chain reaction of reactor explosions over subsequent days. Radioactive contamination was an outcome of the event. Even miles away in the Tokyo metropolitan area, a citizens’ group found Cesium137 hot spots in the soil, with radiation levels comparable to those in the Chernobyl exclusionary and radiation control areas (Alvarez, 2012). Fukushima was the first major nuclear accident caused not by human or technological error, but rather by natural disaster. The consequences of the event prompted reflection on infrastructural preparation for natural disaster, and on whether a Fukushima-scale disaster could happen again. The crisis presented nuclear scientists and policy-makers with the question of whether present spent fuel management adequately ac counts for not only low probability, high consequence events, but also over long enough time horizons to account for infrequent or unantici pated risks (such as those that might arise from persistent climate change). In its aftermath, researchers pointed out 1) that the interim method of storing spent fuel before reprocessing or disposal – i.e. in the spent fuel pools damaged in the case of Fukushima – merits reconsid eration and 2) that a long-term plan for disposal is a necessity, if we want to avoid overreliance on the interim solution (Macfarlane, 2011). Addressing these lessons requires that policy-makers understand the nuances of multiple interlocking systems, which has consistently been a barrier to change in storing nuclear waste. However, proactive – rather than reactive – planning for spent fuel management has consistently received less attention and fewer re sources than merited, leaving the system vulnerable to environmental shocks (Macfarlane, 2011). This vulnerability can lead to unmitigated risks; while the Fukushima disaster impacted the entire power plant, it
� Could sea-level rise pose new risks for coastal nuclear plants and spent fuel storage? 2
L.M. Jenkins et al.
Energy Policy xxx (xxxx) xxx
combination of these factors has resulted in public and political concern. While the statement has yet to be confirmed with a comprehensive risk assessment, the Los Angeles Times reported that a local councilwoman called the plant a “Fukushima waiting to happen” (Chapple, 2018). San Onofre’s canisters were loaded in 2003, and the canisters at other coastal plants like Calvert Cliffs and Maine Yankee were loaded in 1993 and 2002, respectively (Chapple, 2018).
was the nuclear waste (specifically the waste stored in spent fuel storage pools) that proved to be the greatest liability. 2.2. Infrastructural challenges Compounding the problem is the fact that U.S. spent fuel pools are not required to have the same level of protection as reactors precisely because they are assumed to be interim, for before the rods were reprocessed or stored more permanently (Alvarez, 2012). In most U.S. cases, this assumption has not been the case. Spent fuel remains in cooling pools, with a long-term solution yet to be developed and implemented. Prolonged flooding, as with climate change and sea-level rise, could interfere with spent fuel pool heat removal further, blocking ventilation ports with water and silting air passages (U.S. Nuclear Regulatory Commission, 2010). Heat from the radioactive decay in spent nuclear fuel is also a safety concern. Several hours after a full reactor core is offloaded, it can give off enough heat from radioactive decay to melt and ignite the fuel’s reactive zirconium cladding, and over time to destabi lize its geological disposal site. After a century, decay heat and radio activity drop substantially but remain dangerous. The risk of a station blackout like that of Fukushima to pool cooling has not gone unnoticed in the United States. For example, 2016 review by the National Academy of Sciences found that the loss of spent fuel cooling at the Fukushima site “should serve as a wake-up call to nuclear plant operators and regulators about the critical importance of having robust and redundant means to measure, maintain, and, when neces sary, restore pool cooling” (Nuclear and Radiation Studies Board, 2016). The members also urged the NRC to ensure that storage sites have measures in place to mitigate potential zirconium cladding fires from loss of cooling. But the alternative to spent fuel pools – dry casks – are not foolproof and require diligent implementation. The coastal San Onofre plant in Southern California has recently drawn attention to the infrastructural vulnerabilities of transitioning existing spent fuel from pools to dry casks. While the plant has been decommissioned, San Onofre reportedly still stores 3.6 million pounds of spent fuel onsite (Chapple, 2018). Southern California Edison is in the process of moving the fuel from spent fuel pools to dry casks, but has run into problems with the thin canisters used; in early 2018, work was halted after a worker found a loose bolt in a canister (McDonald, 2018). Evidence suggests that the thin canisters may begin to crack and leak after 20 years in use, and they are difficult, if not impossible, to inspect or repair (Gray, 2015). High levels of corrosion are correlated with ground chemicals and moisture, especially in marine environments. The
2.3. Status of spent fuel management in the United States and beyond As of 2009, 78 percent of the 62,683 metric tons of commercial spent fuel in the U.S. was stored in pools – just 22 percent were stored in dry casks (U.S. Nuclear Regulatory Commission, 2017). And as of the end of 2016, the total had swelled to an estimated 77,842 metric tons, with 34 percent stored in dry casks; some estimates put the rate of stockpile increase at 2000 metric tons per year (Holt, 2018). As indicated in Table 1, all operational nuclear power plants at present have at least one spent fuel pool. The only two facilities that do not have fuel pools are inactive, with just dry cask ISFSIs on site: Humboldt Bay and Maine Yankee. As discussed above, in 1982 the NWPA broadly mandated that the DOE develop a repository for used nuclear fuel. Evidently, this has yet to come to fruition. However, the U.S. has taken small steps toward greater safety in the years since, both in terms of technology and regulation. Due to the evidence that an emergency could lead spent fuel pools to leak, Fukushima prompted utilities and their regulators globally to weigh the benefits of the current storage approach and timeline (Gal braith, 2011). For example, in light of the crisis, the U.S. Nuclear Reg ulatory Commission, an independent government agency, researched how a spent fuel pool in the U.S. would react to an even more powerful earthquake (U.S. Nuclear Regulatory Commission, 2015a,b). In light of its findings, the agency directed licensees to 1) install additional in struments to monitor water levels in the pool, and 2) develop ways to easily maintain or restore pool cooling in an emergency (U.S. Nuclear Regulatory Commission, 2015a,b). This illustrates that – despite the high volume of fuel currently stored in pools in the U.S. – the storage requirements can and do evolve in response to potential threats. This challenge extends globally. Most international nuclear stan dards – such as those put forward by the International Standards Or ganization and the World Nuclear Association – have yet to complete standards that directly address increasing climate risks to nuclear and other energy operations (Jordaan et al., 2019). While the International Atomic Energy Agency (IAEA) did put forth a report in 2012 that ad dresses climate change, highlighting “extreme weather conditions” in particular (IAEA, 2012), climate change is not mentioned in the IAEA
Table 1 Seacoast SFM facilities and state locations. Site name
Spent fuel pools (þreactor type, pressurized or boiling water: PWR or BWR)
Nearest water body (þcooling source)
Decommissioning status (þdecommissioning date, if applicable)
Diablo Canyon Humboldt Bay San Onofre
Yes (2 PWRs)
Seacoast (Pacific Ocean)
Seacoast (Pacific Ocean)
Yes (2 PWRs)
Seacoast (Pacific Ocean)
Millstone Crystal River St. Lucie Turkey Point Maine Yankee Calvert Cliffs Pilgrim Seabrook
Connecticut Florida Florida Florida Maine
Yes (2 PWRs) Yes (2 PWRs) Yes (2 PWRs) Yes (2 PWRs) No
Seacoast (Long Island Sound) River (Gulf of Mexico) Seacoast (Atlantic Ocean) Seacoast (Canal systems) River (Black River)
Permanently shut-down, planned decommission (1992) Operational (Unit 1: 1998) Closed (2013) Operational Operational Decommissioned (1997)
Maryland Massachusetts New Hampshire New Jersey Virginia
Yes (2 PWRs) Yes (1 BWR) Yes (1 PWR)
Seacoast (Chesapeake Bay) Seacoast (Cape Cod Bay) Seacoast (Atlantic Ocean)
Operational Operational (2019) Operational
Yes (1 BWR) Yes (1 pool, 2 PWRs)
Seacoast (Barnegat Bay) Seacoast (James River)
Oyster Creek Surry
L.M. Jenkins et al.
Energy Policy xxx (xxxx) xxx
general safety standards. This research aims to address the aforementioned limitations of present spent fuel management in the U.S., to identify climate impacts which may compound present risks (focusing on sea-level rise and the potential consequences), and to develop a policy framework to manage these risks in the face of uncertainty.
rise. This left us with a list of thirteen facilities and their locations. In order to determine the decommission status of each of the plants in question, we used publicly available summaries of each plant’s website with the exception of Crystal River (which has been decom missioned and closed for long enough that no website exists), for which we used a copy of a letter describing the site’s condition that was uploaded to the NRC website (Elnitsky, 2013). We also used NRC data to determine which plants use pool storage, as well as both how many pools and what kind of reactors are used; the data was updated in 2013 and may not be entirely current in every case (U.S. Nuclear Regulatory Commission, 2013). To map these sites, we used Wikipedia’s GeoHack software to record the county, state, and geographical coordinates of every plant other than Humboldt Bay, for which no such information was available, and so we used Google Maps data. We verified that the resulting latitudinal and longitudinal coordinates are correct by visually geo-referencing the sites using satellite images from Google Maps and with a shapefile that in cludes power plant location from the U.S. Energy Information Admin istration (EIA) (Energy Information Administration, 2017). The National Oceanic and Atmospheric Administration (NOAA) publishes maps of different sea-level rise scenarios (US National Oceanic and Atmospheric Administration, 2018). These are undated, though six feet of sea-level rise is in keeping with 2100 predictions by both the IPCC and NOAA (Sweet et al., 2017). We downloaded map data for use in QGIS, an open source mapping software. The NOAA Sea-Level Rise Data for all counties where poten tially vulnerable sites are located were downloaded, resulting in 12 total maps covering the 13 total sites (both New Hampshire and Maine are contained within the same sea-level rise map) (Sweet et al., 2017). We converted the facility location data into an attribute table, and we then used the NOAA data to map both the present sea-level (zero feet of rise) and the future sea-level (six feet of rise) at each. A satellite hybrid map from a Google Maps plugin provided the backdrop for each of these maps. Results include 13 distinct maps, for which we completed a distance analysis to determine the distance between the potential sea-level (after six feet of rise) and the site’s edge. While distance and proximity ana lyses have been applied to situations like hazardous materials trans portation and malarial incidence near swamps (see Appendix 2), we are not aware of any U.S. studies that use these analyses to evaluate the risks presented by the geography of nuclear facilities (Panwhar et al., 2000; Staedke et al., 2003). To execute this analysis for our purposes, we delineated the land area using visual inspection for each plant through the satellite hybrid map. The NNJoin plugin was then used to measure the distance between the six feet of sea-level rise layer and the outline layer in decimal degrees. Using our previous latitude and longitude data, we converted the deci mal degrees distance into meters through a U.S. military conversion tool (National Geospatial-Intelligence Agency, 2018). For those SFM facil ities for which the new sea-level will encroach on the site borders, the distance given is zero. To validate these estimates, we visually approx imated each distance and then calculated them using the Measure tool on QGIS. We then applied a qualitative risk ranking to discern relative levels of risk, based on those distances. Our ranking classifies each site as low, medium, or high risk according to the criteria described below.
3. Methodology While there are numerous risk factors and other environmental threats involved with nuclear power generation, the focus of this research is to evaluate potential implications of sea-level rise to spent fuel sites. The initial intention of this research was to determine how close six feet of sea-level rise will bring the water to existing coastal NPPs, and from there to explore the implications of that information for coastal nuclear plants and spent fuel. While climate change is also likely to impact other water sources, we chose to focus on coastal sites because they will be most vulnerable to sea-level rise, our chosen climate variable. Our analysis includes a comparison of two situations: the present sealevel, and the six feet of sea-level rise scenario. We chose the six feet sealevel rise scenario due to the long time horizon associated with spent fuel management. NOAA reports that seas on the U.S. coasts are ex pected to rise by six feet in as few as 80 years (by 2100), at which point—pending a Yucca Mountain-type solution—much of today’s spent fuel will remain where it is (Sweet et al., 2017). The U.S. DOE plans include spent fuel storage for up to 300 years (Vinson et al., 2011). Similarly, radioactive isotopes decay to harmless materials over long time frames, thus requiring long-term management solutions. The time horizons depend on the nuclear materials in question: strontium-90 and cesium-137 have half-lives of about 30 years (meaning half the radio activity will decay in 30 years), while plutonium-239 has a half-life of 24,000 years (U.S. Nuclear Regulatory Commission, 2015a,b). A consideration of rising seas and changing coastlines is especially important for the process of choosing a location for new plant builds. However, it is also crucial for existing NPPs. Event at plants that have long been decommissioned—like Humboldt Bay (effectively inactive since 1976 in 1976)—spent fuel remains on site because the plants have ISFSIs. While coastal examples of these decommissioned plants use only dry cask storage, without a long-term storage facility, these sites may still remain vulnerable to rising seas and other climate change-related risks. This process required amalgamating data from a number of mainly government and international agency sources and using mapping soft ware to combine their information. Only spent fuel management sites most vulnerable to sea-level rise were evaluated, and we selected the sites for evaluation using the following process. To start, we used the U.S. Department of Energy’s data on spent fuel storage locations (GC 859) as the full population of sites, which were last updated in 2013. The data specified each storage facility, including in cases where more than one was located at the same site. We then screened the data to include only SFM facilities located on the coasts. We categorized them according to the nearest water body (i. e. seacoast, inland near lake, inland near river, etc.) using the Interna tional Atomic Energy Association’s (IAEA) Power Reactor Information System (PRIS) database. For those SFM facilities still located at decommissioned NPP sites, the IAEA did not provide information. We used Google Maps to georeference these sites, and categorized them as “seacoast” cooling source designation as appropriate. Two of these sites (Crystal River and Maine Yankee) are technically located on rivers, but were included under the seacoast designation because their rivers are essentially deltas, and are anticipated to be directly impacted by sea-level rise due to their proximity to the ocean. Table 1 details the remaining NPPs after screening the full list to include only those most vulnerable to sea-level
� Low risk sites are those where the projected sea-level rise will not mean an encroachment of water on the plant; the measured distance between sea-level and site outline is greater than 0. � Medium risk sites are those where sea-level will be bordering or adjacent to the site’s borders, but not encroaching further; the measured distance between sea-level and site outline is 0. � High risk sites are those that will be largely surrounded or even submerged by water with sea-level rise; like medium risk sites, the measured distance between sea-level and site outline is 0. 4
L.M. Jenkins et al.
Energy Policy xxx (xxxx) xxx
penetrated on almost all sides, leaving them particularly vulnerable to flooding or other natural disasters. According to this assessment, six plants of lower risk remain (labeled low-risk in Table 2). For these, projected sea-level rise will not mean an encroachment of water on the outline of the plant, though several have only a meter or less in between the projected sea-level and that outline. Their risk is “low” in relative terms, but their spent fuel storage risks still warrant close attention.
Of the thirteen coastal sites in the US, we found seven sites that are of particular concern (labeled medium or high risk in Table 2): Calvert Cliffs (Maryland), Humboldt Bay (California), Millstone (Connecticut), Oyster Creek (New Jersey), St. Lucie (Florida), Turkey Point (Florida), and Crystal River (Florida). The distance measured between the plant border and the projected border of the sea-level rise was zero for all these sites, meaning that the water will be encroaching on the plant with regularity. While it was not possible to determine the precise location of the spent fuel pools or dry casks at each of these sites, we assume the risk of their exposure is heightened where the indicated water levels are in close proximity (low risk for those greater than 0 m of distance to the sea-level, medium and high for those with 0 distance to the sea and differentiated according to the criteria defined in the methods section). When we conducted a spatial analysis in QGIS, seven plants had a distance of zero meters between the sea-level rise border and the facility border.,A look at the resulting maps illustrates that not all sea-level encroachment presents similar risk; we draw a distinction between the Crystal River plant in Florida (for which parts of the site will be sub merged), and the Calvert Cliffs plant in Maryland, for which the sealevel will only change slightly, despite the water’s encroachment on the site’s outline (see Fig. 1). We found four sites for which the sea-level rise will be bordering or adjacent to the site’s borders, but not encroaching further (labeled medium-risk in Table 2), and three sites that will be partially or completely submerged by water (labeled highrisk in Table 2). While we have not analyzed the geography of each site, we speculate that the four plants that will avoid submersion are built higher above sea-level than the three that will not. The particu larities of the tidal patterns at each site could also be a distinguishing factor. The outline of each plant is indicated in the semi-opaque pink. Other cases that will see partial submersion – like the Crystal River plan – include Humboldt Bay (California) and Turkey Point (Florida) (labeled high-risk in Table 2). While all coastal nuclear plants and storage sites merit attention due to the heightened risk exposure from sea-level rise, these high-risk three should be prioritized due to their circumstances: six feet of sea-level rise will mean the sites’ borders are
5. Discussion For this first analysis, we use proximity to sea-level as a proxy for risk. Sea-level rise, however, is only one impact with implications for spent fuel and nuclear power generation with the shifting landscape caused by climate change; others might include the increased frequency of extreme weather events, and of other natural disasters like fire or drought. A longer-term risk based approach is warranted, along with a comprehensive risk assessment of potential climate threats, despite various formulations of uncertainty (Granger Morgan et al., 2009). We propose that the nuclear sector and related decision-makers work to wards an integrated and iterative strategy towards incorporating risk (Fig. 2). This decision model can help policymakers take into account longterm risks of climate change, while encouraging reevaluation and analysis throughout the decision-making process. The model splits after the risk assessment stage, noting that further research and policymaking should happen simultaneously in order to avoid a situation in which nothing is done until research resolves all uncertainties. Doing both at once will allow the U.S. to optimize and adapt the process of reassessing regulations, policies and standards. While the model osten sibly begins with identifying potential climate threats (such as sea-level rise), the final step of identifying gaps in need of new policy action and research prompts a re-identification of potential threats, as they will likely have changed as policies change. And thus the cycle repeats itself. We recommend policy-makers use an iterative decision-making strategy (Fig. 2) when considering the results of our preliminary assessment; any examination of particular risk factors should instigate both policy consideration and further research. While we use proximity to water (with sea-level rise) as a proxy for overall risk in our analysis, ultimately a comprehensive assessment should be undertaken to more robustly capture the ensuing risks. A more comprehensive risk assessment could factor both site-specific risk factors, and more universal risk factors posed by climate change. Using the basic model proposed in Fig. 1 to map any of the following factors would be a fruitful way to add specificity to the risk assessment of spent fuel and climate change, as well as potential long-term strategies.
Table 2 Proximity of screened SFM facilities to water at 0- and 6-feet sea-level rise. Site name
Distance (m) to 0 feet SLR
Distance (m) to 6 feet SLR
Diablo Canyon Humboldt Bay San Onofre
Pacific Gas and Electric Company Pacific Gas and Electric Company Southern California Edison Company Dominion Nuclear Connecticut Progress Energy Florida Florida Power & Light Company Florida Power & Light Company Maine Yankee Atomic Power Company Constellation Energy
Entergy Nuclear Generation Company NextEra Energy, Inc. AmerGen Energy Company, LLC Dominion Virginia Power
Millstone Crystal River St. Lucie Turkey Point Maine Yankee Calvert Cliffs Pilgrim Seabrook Oyster Creek Surry
� Universal climate change risk factors are related to global weather patterns and geography, and include the frequency of the extreme weather events such as hurricanes and other storms, both historical and projected; the frequency of fire and drought; the frequency of flooding and whether or not the plant is located in a flood plain; and the frequency of earthquakes and the location of fault lines (Power Engineering, 2014; Eaton, 2012). � Site-specific risk factors include whether a site stores its fuel in spent fuel pools or dry casks; location of spent fuel within the layout of the overall site; how much fuel is stored; whether a plant is decom missioned, active, or has another status; and the population of the surrounding area. Policy-makers would benefit from understanding the total spent fuel stockpile within the US, particularly if that understanding factored in the total liabilities, including potential environmental, economic, and public health risks under different management regimes. Furthermore, this analysis only examines the SFM facilities in the continental US. While countries beyond the U.S. differ in their approach 5
L.M. Jenkins et al.
Energy Policy xxx (xxxx) xxx
Fig. 1. Satellite images show three different SFM facilities, with light blue indicating the current sea-level, and the dark blue indicating the level with six feet of sealevel rise. These show the distinctions between the low (a), medium (b), and high (c) levels of risk we assessed. Diablo Canyon (a) is an example of a low risk site because the water will not encroach on the outline of the facility (in pink) at all. Calvert Cliffs (b) is an example of a medium risk site because the water will encroach upon the site, but only slightly; it will be bordering or adjacent to the border of the facility. Crystal River (c), however, is considered high risk because it will be surrounded by water – six feet of sea-level rise will penetrate the site’s borders on almost all sides.
Fig. 2. A longer-term adaptable approach to incorporating climate risks to nuclear power and spent fuel management and decision strategies, adapted from an approach articulated by M. Granger Morgan and his collaborators in their 2009 report addressed to Congress. The uncertain nature of the problem can be addressed with an iterative and adaptive approach, with no specific start or end. This reflects our understanding of the problem and the process as evolving.
to spent fuel, and therefore warrant different policy-making approaches, considering the spent fuel stored coastally on an international scale could multiply the overall risks reflected in this model. Such factors should inform more comprehensive risks assessments for nuclear power generation under climate change that include a more complete list of experts and stakeholders (e.g. the IPCC, IAEA, NRC, the U.S. DOE, and representatives of other countries).
Ultimately, we argue that the three sites of highest risk (Humboldt Bay (California), Turkey Point (Florida), and Crystal River (Florida), as mentioned above) warrant particular policy attention because they are the most likely to be damaged as a function of climate change. The crisis at Fukushima illustrated the potential gravity of that damage; the radioactive material contained within spent fuel pools (used as an interim storage measure at both Fukushima and at many sites in the U. 6
L.M. Jenkins et al.
Energy Policy xxx (xxxx) xxx
and regulators in the last decade, 66 percent of fuel remains in pools; it is unclear how much of that is overdue. So, with some legislative recon figuring, it would be possible to direct the readily available money to secure all spent fuel that is over five years old in dry casks. Doing so would provide a measure of short-term security for the spent fuel supply while a long-term solution is prepared. On that note, it remains crucial that the U.S. find a long-term storage solution, and soon. The majority of this NRC fund should be reserved for that purpose. As discussed above, the Yucca Mountain proposal was rife with controversy because it was imposed upon the Nevadan locals without their consideration. The Blue Ribbon Commis sion on America’s Nuclear Future, tasked with finding an alternative, proposes divorcing politics from the situation entirely. They suggested that a federally chartered corporation might do a better job than the DOE, which “must balance multiple agendas or policy priorities” (Wald, 2012). Perhaps they are right; politics have thrown a wrench into so lutions for both nuclear waste and for climate change in the past. However, momentum on this issue will support nuclear power genera tion in the U.S. Short-term options may not address long-term risks. Finally, the U.S. should use its clout to encourage international treaties and standards to take the climate change threat into ac count. The U.S. is not alone in ignoring the reality of climate risks to its nuclear power fleet or spent fuel storage process. Of four international and U.S. standards on nuclear energy and nuclear weapons, only one (the IAEA’s “Safety of Nuclear Fuel Cycle Facilities” report) mentions climate change as a factor necessitating specific regulatory action (IAEA, 2012). The others (from the International Standards Organization, the World Nuclear Association, and the U.S. NRC) barely mention the phe nomenon; if they do, it is only via reference to the vague “extreme weather events” necessitating preparation. These treaties and standards are crucial for international management of nuclear energy, and for them to be stuck in a climate change-ignorant delusion is dangerous. While implementing and extending these standards both domestically and internationally may seem purely symbolic, only through these in stitutions can climate change be normalized as a threat. Once that happens, the threat can be controlled for via regulation. The reality of the combined threats of spent fuel and climate change cannot be ignored any longer. Weathering the perfect storm of coastal spent fuel pools and rising sea levels will take a concerted, coordinated effort—one that will be made more difficult through partisan gridlock in the current Congress. Spent fuel is too big a liability for policy-makers to continue pushing it under the rug. While attaching it to the equally important and contro versial issue of climate change might make both problems even more intractable, it is crucial to understand risks in advance of environmental events so they can be adequately managed. Fukushima did not result in a complete management of all environmental risks to spent fuel man agement. In its aftermath, a more permanent, long-term strategy has yet to be confirmed. To date, long-term solutions have depended on the assumption that present plants will continue to operate as they have and that spent fuel will be transferred from short-term coastal storage to a Yucca-like alternative. Moving forward, a comprehensive nuclear en ergy plan cannot take only the generation-side risks into account. If the United States continues use nuclear power – which already contributes a substantial portion of the grid’s clean, low-to-no carbon electricity – then adequate long-term spent fuel management is unavoidable, as is the reality of rising seas.
S.) is far more vulnerable to shocks than the more permanent dry casks. Because the United States has yet to implement a long-term storage solution for nuclear waste in the U.S., policy-makers have relied more and more on these spent fuel pools. Climate change and sea-level rise make that reliance riskier, especially on the coasts. 6. Conclusions and policy implications Nuclear power has the long-term capacity to contribute as a lowcarbon power source not only for the United States but also globally. In addition to economic challenges, however, it comes with complica tions for long-term spent fuel management that have been largely overlooked. Nuclear waste is going to remain on Earth for a very long time—far longer than the typical timeline of political decisions. While we have the capacity to predict its decay over the long-term, we generally lack the imagination to conceptualize the centuries or even millennia that our waste will outlast us. This problem persists across similar study of fossil fuels and carbon dioxide: the “long tail of the carbon timescale” has been neglected in calculations of global warm ing’s potential impact. Consequently, policy-makers have routinely underestimated the longevity of impacts related to today’s decisions (Archer et al., 2009). A variation on this same phenomenon is at work with nuclear waste. Our inability to conceptualize the inevitability of the interaction be tween spent fuel and climate change has compromised the political will to come up with a long-term strategy. Complicating the matter is the fact that there is no economic incentive to deal with the waste; in fact, decommissioning and nuclear waste cleanup is exorbitant. Electricity generation – like any sector – is a money-making game, whereas dealing with waste is costly.. It is true in other spheres of the energy sector: oil and gas companies frequently go bankrupt rather than dealing with cleaning up orphan wells or un claimed sites. However, the major difference between fossil fuels and nuclear fuel is the latter’s heightened risk to humans when left un managed. As we saw in Fukushima, the catalyst of a natural disaster can leave the area surrounding a nuclear waste site uninhabitable for the foreseeable future. Those stakes are high enough that managing the problem before disaster strikes is preferable to picking up the pieces in the aftermath. After Fukushima, the NRC considered two new policies related to 1) 2012 safety orders and 2) new decommissioning standards (Nunez, 2015). These were realized with the creation of the FLEX program, which expanded regulatory policy to include a “diverse and flexible coping strategy.” However, they mostly relate to the mechanisms for shutting down power generation in case of a crisis, rather than to spent fuel storage, which is arguably the greater threat. For the time being, there are three steps policy-makers should take in order to mitigate the dangers posed to spent fuel by climate change. From the broadest perspective, a new approach is required to incorporate climate risks into the management of the nuclear power fleet. While longer-term risks can be evaluated using such an approach, there are clearly short-term management decisions that can be made to reduce potential threats. We provide three such examples below. First, the task of moving any remaining overdue spent fuel from their cooling pools to dry casks should be a national priority in the short-term, especially where those cooling pools are located at the identified high risk sites. A 2010 Electric Power Research Institute (EPRI) study estimated that the cost of moving the fuel would be $3.6 billion (Alvarez, 2012). While that sum seems prohibitive, there are potentially funds already in place that could be redirected for this pur pose by the NRC. The NWPA established a user fee of 0.1 cent per kilowatt-hour, explicitly for the search for and establishment of a long-term waste solution. The law did not allow the funds to be used to secure existing storage facilities. As of fiscal year 2010, only $7.3 billion of the $25.4 billion accrued had been spent, leaving a pot of $18.1 billion (Alvarez, 2012). While strides have been made by both legislators
Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
L.M. Jenkins et al.
Energy Policy xxx (xxxx) xxx
Appendix A. Supplementary data
Masson-Delmotte, V., et al., 2018. Summary for policymakers. Global warming of 1.5� C an IPCC special report. https://doi.org/10.1017/CBO9781107415324. McDonald, J., 2018. Work halted for 10 days when loose bolt is found in San Onofre radioactive waste canister. Los Angeles Times. https://www.latimes.com/business /la-fi-san-onofre-spent-fuel-bolt-20180325-story.html. Morgan, M.G., Abdulla, A., Ford, M.J., Rath, M., 2018. US nuclear power: the vanishing low-carbon wedge. Proc. Natl. Acad. Sci. 115 (28), 7184–7189. https://doi.org/ 10.1073/pnas.1804655115. National Geospatial-Intelligence Agency, 2018. Length of a degree of latitude and longitude [Online calculator tool]. Accessed September 2018. https://msi.nga.mil/m sisitecontent/staticfiles/calculators/degree.html. Nuclear, Radiation Studies Board, 2016. Lessons Learned from the Fukushima Accident for Improving Safety and Security of U.S. Nuclear Plants: Phase 2. National Academies Press, p. 51. https://www.nap.edu/catalog/21874/lessons-learned-fromthe-fukushima-nuclear-accident-for-improving-safety-and-security-of-us-nuclearplants. Nuclear Energy Institute, 2019. Used nuclear fuel. https://www.nei.org/advocacy/make -regulations-smarter/used-nuclear-fuel. Nunez, C., 2015. As sea levels rise, are coastal nuclear plants ready? Natl. Geogr. https ://news.nationalgeographic.com/energy/2015/12/151215-as-sea levels-rise-arecoastal-nuclear-plants-ready/. Panwhar, S.T., Pitt, R., Anderson, M.D., 2000. Development of a GIS-Based Hazardous Materials Transportation Management System. A publication of the University Transportation Center for Alabama, pp. 1–36. http://rpitt.eng.ua.edu/Publication s/TTaccidentsandEnv/route%20selection%20with%20GIS%20modeling.pdf. Wildfires prompt evacuation of san Onofre nuclear plant. Power Eng., 2014 https://www .power-eng.com/articles/2014/05/wildfires-prompt-evacuation-of-san-onofre-nucle ar-plant.html. Staedke, S.G., Nottingham, E.W., Cox, J., Kamya, M.R., Rosenthal, P.J., Dorsey, G., 2003. Short report: proximity to mosquito breeding sites as a risk factor for clinical malaria episodes in an urban cohort of Ugandan children. Am. J. Trop. Med. Hyg. 69 (3), 244–246. https://doi.org/10.4269/ajtmh.2003.69.244. Sweet, W.V., et al., 2017. NOAA technical report NOS CO-OPS 083, National Ocean service center for operational oceanographic products and services: global and regional sea level rise scenarios for the United States, center for operational oceanographic products and services national (January). https://tidesandcurrents. noaa.gov/publications/techrpt83_Global_and_Regional_SLR_Scenarios_for_the_ US_final.pdf. Online mapping tool and dataset U.S. National Oceanic and Atmospheric Administration, 2018. Sea level rise viewer, Accessed September 2018. https://coast.noaa.gov/slr/. U.S. Nuclear Regulatory Commission, 2010. Standard review plan for spent fuel dry storage systems at a general licensed facility: NUREG-5631, pp. 2–13. https://www. nrc.gov/reading-rm/doc-collections/nuregs/staff/sr1536/r1/sr1536r1.pdf. U.S. Nuclear Regulatory Commission, 2012. Spent fuel storage in pools and dry casks, key points and questions & answers. https://www.nrc.gov/docs/ML1209/ML1209 70230.pdf. U.S. Nuclear Regulatory Commission, 2013. Spent fuel criticality management database. https://www.nrc.gov/docs/ML1321/ML13212A064.pdf. U.S. Nuclear Regulatory Commission, 2015. Backgrounder: radioactive waste. htt ps://www.nrc.gov/reading-rm/doc-collections/fact-sheets/radwaste.pdf. U.S. Nuclear Regulatory Commission, 2015. Backgrounder on storage of spent nuclear fuel. https://www.nrc.gov/reading-rm/doc-collections/fact-sheets/storage-spentfuel.html. U.S. Nuclear Regulatory Commission, 2017. Spent fuel storage in pools and dry casks, key points and questions & answers. https://www.nrc.gov/waste/spent-fuel-s torage/faqs.html. Vinson, D., et al., 2011. Inventory and description of commercial reactor fuels within the United States: fuel cycle research & development. Dep. Energy. https://sti.srs. gov/fulltext/SRNL-STI-2011-00228.pdf. Wald, M., 2012. Revamped search urged for a nuclear waste site. N. Y. Times. https:// www.nytimes.com/2012/01/27/science/earth/nuclear-waste-panel-urges-consentbased-approach.html.
Supplementary data to this article can be found online at https://doi. org/10.1016/j.enpol.2019.111106. References Alvarez, R., 2012. Improving spent-fuel storage at nuclear reactors. Issues Sci. Technol. 28 (2), 77–84. American Society of Mechanical Engineers, 2019. Shippingpoint nuclear power station. Accessed February 2019. https://www.asme.org/about-asme/who-we-are/engineer ing-history/landmarks/47-shippingport-nuclear-power-station. Archer, D., et al., 2009. Atmospheric lifetime of fossil fuel carbon dioxide. Annu. Rev. Earth Planet Sci. 37 (1), 117–134. https://doi.org/10.1146/annurev. earth.031208.100206. Chapple, S., 2018. The San Onofre nuclear plant is a ‘Fukushima waiting to happen’. Los Angeles Times. https://www.latimes.com/opinion/op-ed/la-oe-chapple-san-onofre20180815-story.html. Department of Energy, 2019. Used fuel disposition research & development [Office homepage]. Accessed April 2019. https://www.energy.gov/ne/fuel-cycle-technolo gies/used-fuel-disposition-research-development. Department of Energy, in cooperation with the US Nuclear Regulatory Commission, US Environmental Protection Agency, US Department of State, 2017. Sixth national report for the joint convention on the safety of spent fuel management and on the safety of radioactive waste management. https://www.energy.gov/sites/prod/files/ 2017/12/f46/10-20-17%206th_%20US_National_Report%20%28Final%29.pdf. Eaton, J., 2012. Record heat, drought pose problems for U.S. electric power. Natl. Geogr. https://news.nationalgeographic.com/news/energy/2012/08/120817-record-heat -drought-pose-problems-for-electric-power-grid/. Elnitsky, J., 2013. Crystal River Unit 3 – Post-shutdown Decommissioning Activities Report [Letter to the U.S. Nuclear Regulatory Commission]. Duke Energy Florida. https://www.nrc.gov/docs/ML1334/ML13340A009.pdf. Energy Information Administration, 2017. Operable electric generating plants in the United States by energy source: layer information for interactive state maps. htt ps://www.eia.gov/maps/layer_info-m.php. Energy Information Administration, 2018. What is U.S. electricity generation by energy source? https://www.eia.gov/tools/faqs/faq.php?id¼427&t¼3. Environmental Protection Agency, 1982. Summary of the Nuclear Waste Policy Act, 42 U.S.C. x10101 et seq. https://www.epa.gov/laws-regulations/summary-nuclear-w aste-policy-act. Galbraith, K., 2011. A new urgency to the problem of storing nuclear waste. N. Y. Times. http://www.nytimes.com/2011/11/28/business/energy-environment/a-new-urge ncy-to-the-problem-of-storing-nuclear-waste.html. Granger Morgan, M., et al., 2009. Best practice approaches for characterizing, communicating, and incorporating scientific uncertainty in decision making. US Climate Change Science Program: synthesis and Assessment Produce 5.2. http s://www.academia.edu/24925284/Best_Practice_Approaches_for_Characterizing_ Communicating_and_Incorporating_Scientific_Uncertainty_in_Decision_Making. Gray, E., 2015. Coast to coast: spent fuel dry storage problems and recommendations. In: NRC Reg Con. Powerpoint presentation. https://www.nrc.gov/public-involve/confe rence-symposia/dsfm/2015/dsfm-2015-erica-gray.pdf. Holt, M., 2018. Nuclear energy: overview of congressional issues. Congr. Res. Serv. https://fas.org/sgp/crs/misc/R42853.pdf. IAEA, 2012. IAEA safety standards: storage of spent nuclear fuel. Specific safety guide No. SSG-15. http://www-pub.iaea.org/MTCD/publications/PDF/Pub1503_web.pdf. Jordaan, S.M., Siddiqi, A., Kakenmaster, B., Hill, A., 2019. The climate vulnerabilities of global nuclear power. Glob. Environ. Politics 19 (4), 3–13. Kopytko, N., Perkins, J., 2011. Climate change, nuclear power, and the adaptationmitigation dilemma. Energy Policy 39 (1), 318–333. https://doi.org/10.1016/j. enpol.2010.09.046. Macfarlane, A., 2011. The overlooked back end of the nuclear fuel cycle. Science 333 (6047), 1225–1226. http://www.sciencemag.org/content/333/6047/1225.full.pdf.