Nuclear Waste and Energy Storage: What Each Position Is Protecting

Tiina has spent fifteen years as a geologist working on repository characterization for the Onkalo facility on the Olkiluoto peninsula in southwest Finland. In 2023, Finland's nuclear waste management company POSIVA received an operating license for Onkalo — making it the first permanent geological repository for high-level nuclear waste ever approved for operation in the world. The tunnels are drilled six hundred meters into Precambrian granite that has been geologically stable for 1.8 billion years. Spent fuel assemblies will be sealed in copper canisters, surrounded by compacted bentonite clay, and placed in deposition tunnels. When the tunnels are full, they will be sealed. The design premise is that the repository should function without institutional continuity — without anyone needing to remember it is there, without any society needing to maintain it. The geology is the containment.

Marcus spent twelve years working on the Yucca Mountain repository project in Nevada. He watched the United States spend thirty years and billions of dollars characterizing a site in the Nevada desert, earn a license recommendation from the Nuclear Regulatory Commission, and then watch the license application withdrawn by the Obama administration in 2010. The Government Accountability Office and the NRC's own career staff later concluded that the withdrawal was made for political reasons — opposition from Nevada's congressional delegation — not because the site failed safety criteria. The approximately ninety thousand metric tons of spent nuclear fuel that were supposed to go somewhere permanent still sit in temporary dry cask storage at more than seventy facilities across the country, some at reactors that shut down decades ago, with no permanent destination and no mechanism in place to find one.

The nuclear waste debate is often framed as a subsidiary of the nuclear energy debate: is spent fuel a manageable byproduct or an irresolvable liability? But the waste question is distinct from the generation question, and collapsing them prevents either from being engaged clearly. The ninety thousand metric tons already exist regardless of whether any new nuclear plants are built. The debate about what to do with them has four positions that disagree not just about the answer but about what kind of problem nuclear waste is.

What permanent repository advocates are protecting

The permanent repository position holds that deep geological disposal is the scientifically endorsed, morally appropriate solution to high-level nuclear waste — and that the failure to build a repository is a governance failure, not a technical or safety failure. The science is settled. The politics are not.

They are protecting the integrity of thirty years of scientific work. The international scientific consensus — including the positions of the Nuclear Regulatory Commission, the International Atomic Energy Agency, and the National Academy of Sciences — holds that deep geological disposal in stable rock is the safest long-term solution for high-level nuclear waste. This is not a disputed conclusion. Finland's Onkalo project demonstrates that it is technically achievable. Sweden, France, and Canada are building or advancing similar facilities. The United States spent thirty years studying Yucca Mountain specifically because it satisfies the geological criteria — the volcanic tuff is dry, the water table is deep, the geology is stable. Repository advocates are protecting the principle that a scientific consensus developed over decades, at significant public expense, should not be abandoned because of electoral politics in one state.

They are protecting a promise already made and paid for. From 1983 until 2014, nuclear utilities paid a fee of one tenth of a cent per kilowatt-hour of nuclear-generated electricity into a Nuclear Waste Fund — totaling over forty-five billion dollars — specifically to fund permanent waste disposal. In exchange, the federal government committed to begin accepting spent fuel by 1998. That deadline passed without a repository, utilities began suing the Department of Energy for breach of contract, and Congress halted the fee collection in 2014. Nuclear utility ratepayers funded the repository. Repository advocates are protecting the principle that a government that collects money under a specific promise and then abandons the promise for political convenience owes something to the parties that funded the work.

They are protecting the distinction between political opposition and safety failure. Yucca Mountain was not abandoned because it was unsafe. The NRC's license review was proceeding and had not produced a negative finding when the application was withdrawn. Repository advocates are protecting the principle that the appropriate mechanism for deciding whether a nuclear facility is safe is the technical licensing process, not a legislative veto driven by a senator's political interests. When the technical process is terminated before it produces a finding, it produces a different kind of precedent: any future repository site can be blocked by the same mechanism, which means no permanent repository will ever be built if political opposition from the host state is treated as sufficient cause for withdrawal. The Onkalo comparison is instructive. Finland's process took decades and involved extensive public engagement in the host community. The community ultimately supported it. The U.S. process imposed a site without consent and then abandoned the scientific work when consent could not be secured.

What interim consolidated storage advocates are protecting

The interim consolidated storage position does not contest that a permanent repository is ultimately needed. It argues that the thirty-year wait for a permanent solution has allowed the current at-reactor storage arrangement — more than seventy temporary sites, including sites at already-shut-down reactors — to become a de facto permanent arrangement by default, and that improving the interim situation does not require resolving the permanent question first.

They are protecting the difference between a good-enough interim solution and no solution. The Blue Ribbon Commission on America's Nuclear Future — the expert panel appointed by the Obama administration after Yucca Mountain was abandoned, co-chaired by Brent Scowcroft and Lee Hamilton — concluded in its 2012 report that the first priority should be establishing at least one consolidated interim storage facility. Current at-reactor dry cask storage is safe for decades — the NRC's Generic Environmental Impact Statement (NUREG-2157, 2014) assessed it safe for at least sixty years beyond reactor shutdown — but the current arrangement has structural vulnerabilities that a centralized facility would address. Reactors that shut down unexpectedly, as several have in the last decade, leave their spent fuel on-site without the security and maintenance infrastructure of an operating plant. Sites in coastal areas and flood plains face increasing climate exposure. And seventy-plus storage locations require seventy-plus separate security arrangements, monitoring systems, and regulatory oversight structures when one well-designed facility could handle all of it.

They are protecting consent-based siting as the mechanism that might actually work. The Yucca Mountain failure is partly a siting justice failure: the site was selected in 1987 through a political process that overrode Nevada's objections, and the resistance that followed was politically effective enough to eventually kill the project. The Blue Ribbon Commission recommended replacing top-down siting — in which the federal government designates a site — with consent-based siting, in which communities volunteer to host storage facilities and negotiate the terms of hosting, including economic development, regulatory participation rights, and contractual commitments about duration. Two consolidated interim storage projects — Holtec International's proposed facility in New Mexico and Interim Storage Partners' proposed facility in Texas — have received NRC licenses, partly on the basis of community interest. Both remain blocked by state-level opposition and a legal dispute about whether interim consolidated storage can proceed without a licensed permanent repository. Interim storage advocates are protecting the principle that a siting approach that actually produces a willing host is worth more than a legally correct process that produces a permanent opponent.

They are protecting communities bearing permanent risk under a temporary arrangement. The Zion Nuclear Power Station north of Chicago shut down in 1998. Its spent fuel sits today on the shore of Lake Michigan — the drinking water source for roughly thirty million people — in dry casks managed by a decommissioning company with no operating business. The people who agreed to host an operating nuclear plant in the 1970s did not agree to become a permanent nuclear waste storage site. The community has no legal mechanism to compel removal. Interim consolidated storage would give communities like Zion's a realistic pathway to transferring the waste somewhere purpose-built and purpose-governed, rather than leaving it in place indefinitely under an arrangement that was never designed to be permanent.

What advanced reactor and transmutation advocates are protecting

There is a position in the nuclear waste debate that declines to accept the framing of the question. The debate asks: where do we put the waste? This position asks: does the waste have to remain what it currently is?

High-level nuclear waste is not a uniform substance. Spent fuel assemblies contain uranium (most of what went in), fission products (the direct byproducts of fission, including cesium-137 and strontium-90), and transuranic actinides (plutonium, americium, neptunium, and curium, produced when uranium absorbs neutrons without splitting). The fission products have half-lives measured in decades — they decay to relatively benign levels within a few hundred years. The actinides have half-lives measured in tens of thousands of years. It is the actinides that require the ten-thousand-year regulatory horizon for Yucca Mountain and make permanent geological disposal so technically demanding. The repository has to remain intact and unbreached for longer than human civilization has existed.

They are protecting the recognition that the actinides are energy, not just waste. Fast neutron reactors — operating on high-energy neutrons rather than the slow neutrons of conventional light-water reactors — can fission the long-lived actinides, transmuting them into shorter-lived fission products. The Integral Fast Reactor program at Argonne National Laboratory, developed through the 1980s and early 1990s and cancelled by Congress in 1994, demonstrated this in principle. Charles Till and Yoon Il Chang, the scientists who led the program, documented in Plentiful Energy (2011) that spent fuel reprocessed through the IFR's electrochemical process produced fuel for the fast reactor and a waste stream with a hazard period of approximately three hundred years rather than three hundred thousand — a three-orders-of- magnitude reduction in the required regulatory horizon. A geological repository that needs to contain waste for three hundred years is a fundamentally different engineering and governance problem than one that needs to contain it for ten thousand.

They are protecting the closing of the nuclear fuel cycle as a solution to both waste and resource depletion. Conventional light-water reactors use approximately two to four percent of the potential energy in their uranium fuel; the rest ends up in spent fuel as a waste burden. Fast reactors in a closed fuel cycle can use close to all of the energy in uranium, including the enormous stockpiles of depleted uranium sitting at enrichment facilities as a byproduct of conventional fuel production. TerraPower, co-founded with Bill Gates's backing, is developing a natrium reactor design. Terrestrial Energy is developing a molten salt reactor. Several other advanced reactor companies are pursuing designs that address waste through fuel cycle closure. These advocates are protecting the possibility of a nuclear energy system that generates electricity, reduces the hazard period of existing waste, and uses its fuel far more completely than the current generation of reactors.

They are protecting option value that the other positions foreclose. Building a permanent repository for the current waste stream is not incompatible with transmutation — but a policy locked into geological disposal as the only solution may be less likely to invest in the research and demonstration needed to make transmutation work at scale. Transmutation advocates are protecting the principle that a problem whose solution is technically uncertain should not be treated as settled until the technical alternative has been genuinely tried. The Integral Fast Reactor was cancelled three years before its planned operational demonstration. That cancellation was not a finding that transmutation doesn't work; it was a funding decision. The difference matters for what the current policy should be.

What no-new-waste advocates are protecting

The fourth position holds that the debate about where to put nuclear waste is downstream of a more fundamental question: whether to keep generating it. The appropriate response to producing a long-lived hazard without a demonstrated solution is not to optimize the storage arrangement — it is to stop producing the hazard while working on the solution.

They are protecting intergenerational consent at its fullest extension. The anti-nuclear movement has long argued that generating nuclear waste imposes costs on future generations without their consent. The no-new-waste position makes this specific: the United States has been generating high-level nuclear waste since the 1950s. As of 2026, after sixty-plus years and billions of dollars of effort, there is still no operating permanent disposal facility. Every year of continued nuclear operation adds to an obligation that has not been met and whose meeting remains uncertain. The no-new-waste argument is not that geological disposal is impossible — Finland's progress suggests it is — but that it is irresponsible to continue generating waste whose permanent disposal is not yet arranged. The obligation to future people is not discharged by promising to build a repository. It is discharged by building one. Until the promise is kept, expanding the waste stream is expanding an obligation that has been pending for three generations.

They are protecting the communities that have historically borne the localized costs of nuclear siting. From the 1940s through the 1980s, uranium was mined on Navajo Nation land under conditions that exposed miners, their families, and surrounding communities to radiation at levels that produced elevated rates of lung cancer and kidney disease for decades after mining ended. The mines were inadequately remediated. The communities that bore these costs did not consent to them in any meaningful sense, and the benefits — electricity for distant cities — did not flow to them. Traci Brynne Voyles documents in Wastelanding (2015) how this pattern — energy extraction from Indigenous land, with the costs remaining after the extraction ends — is systematic rather than incidental. Proposed interim and permanent waste storage sites in New Mexico and Texas are in predominantly rural, low-income, and Hispanic communities. No-new-waste advocates are protecting the principle that the calculation of nuclear energy's costs must include the siting of waste facilities, not just the aggregate deaths-per-terawatt-hour statistics that anonymize the specific communities bearing concentrated localized risk.

They are protecting appropriate skepticism toward technical promises in a domain of broken promises. The Nuclear Waste Policy Act of 1982 promised a permanent repository by 1998. It is 2026. The Hanford Site in Washington State — the largest nuclear waste cleanup project in U.S. history — has been leaking from underground storage tanks into groundwater for decades despite a cleanup effort that has consumed billions of dollars and whose timeline and completeness remain uncertain. Transmutation advocates offer a compelling vision; the Integral Fast Reactor program was cancelled before commercial-scale demonstration. Consolidated interim storage has NRC licenses and no congressional authorization to use them. No-new-waste advocates are protecting the epistemic principle that in a domain where confident assurances have a long track record of not being kept, the burden of proof runs against new assurances rather than for them.

Where the real disagreements live

The four positions share the premise that the current situation — ninety thousand metric tons of spent fuel at temporary sites with no permanent destination — is unsatisfactory. The disputes run along three lines that rarely get named explicitly.

What kind of problem is this? The permanent repository position treats nuclear waste as an engineering and governance problem with a known solution: build the geological repository. The interim storage position treats it as a sequencing problem: improve the interim arrangement while working toward the permanent solution. The transmutation position treats it as a technical problem with a better solution than geological disposal: change what the waste is. The no-new-waste position treats it as a moral and political problem: the question of where to put waste is secondary to the question of whether to generate more. These are not competing answers to the same question. They are different diagnoses of what the question is — which means that a policy successfully addressing one diagnosis may leave the others completely intact. A geological repository, built and operating, would satisfy the permanent repository position and the interim storage position. It would not satisfy the no-new-waste position's concern about communities bearing new siting costs. It would not satisfy the transmutation position's case that the waste stream itself could be transformed.

What time horizon is the relevant one? The nuclear waste debate is necessarily a dispute about time, but the four positions use radically different timescales. The NRC's finding that dry cask storage is safe for sixty years beyond reactor shutdown is technically precise and tells you nothing about what happens over centuries or millennia. The regulatory horizon for Yucca Mountain is ten thousand years — a period longer than recorded human history. The hazard period of long-lived actinides without transmutation is three hundred thousand years — a period longer than anatomically modern humans have existed. The transmutation position argues that reducing the hazard period to three hundred years makes the governance problem institutionally tractable, since three hundred years is within the range of human institutional experience. Which time horizon is the right one for making current policy is not a technical question. It is a question about what kind of obligations the present generation has to the future — and how far into the future that obligation extends before it becomes unenforceable by any mechanism we can design.

Who has standing to make siting decisions? The Yucca Mountain failure turned largely on a siting justice question: Nevada was selected partly because it had a small congressional delegation and less political power to resist. The consent-based siting alternative raises its own questions: does a community genuinely consent when it volunteers to host a nuclear waste facility in exchange for economic benefits, or does limited economic alternative make consent a misleading description of the process? And beyond the current community, there is the question of communities that will exist near the storage site in fifty or five hundred years. Those communities have the highest stake in a decision with a multi-millennium horizon, and they have no standing in any current decision-making process. This is the structural absence problem at its most extreme: the people who will live longest with the consequences of current decisions are not absent from the process for procedural reasons — they are absent because they do not yet exist.

What sensemaking surfaces

The nuclear waste debate has a distinctive structure that is worth naming separately from other debates in this library: the fundamental asymmetry between the time horizons of energy generation and the time horizons of waste management. Nuclear plants operate for forty to eighty years. Their waste must be managed for hundreds of thousands of years. Any institution that currently exists — regulatory agency, corporation, democratic government, nation-state — will be unrecognizable within the operational lifespan of the waste, let alone its full hazard period. The Finnish Onkalo project addresses this directly by designing a repository that doesn't require institutional continuity: the geology is the containment, not the oversight. This is why the geological disposal consensus is not merely preference — it is a response to the recognition that human institutions are not a reliable long-term containment mechanism.

The transmutation position is compelling in principle and genuinely uncertain in delivery. If fast reactor transmutation works at commercial scale, it transforms the waste problem from a ten-thousand-year engineering challenge to a three-hundred-year one — and three hundred years is within the range where human institutions have actually maintained continuity. The question is whether that technical transformation will be demonstrated and deployed on the timeline the waste problem requires. The Integral Fast Reactor was cancelled before commercial-scale demonstration; the currently operating advanced reactor designs are still in development or early construction. A strategy that depends on transmutation being available within a generation is not the same as a strategy backed by demonstrated technology.

The interim consolidated storage position is the most immediately actionable of the four — and it faces the most immediate institutional obstacle. Two NRC-licensed consolidated storage facilities cannot operate because federal law, as some courts have interpreted it, requires a licensed permanent repository before interim consolidated storage can proceed. This is the institutional default in a specific form: the legal framework makes it impossible to improve the interim situation without first resolving the permanent situation. That is exactly the situation the country has been in for forty years. The Blue Ribbon Commission's recommendation — fix the interim situation while working on the permanent one — has not been implemented not because it was rejected as bad policy but because no administration has been willing to spend the political capital to move the enabling legislation through Congress.

What sensemaking reveals is that this debate has four distinct layers of difficulty operating simultaneously, and the four positions each primarily engage one or two of them. There is a technical layer: can the geology contain the waste? Can fast reactors transmute it? There is a governance layer: what institutions can make credible commitments across the required time horizons? There is a justice layer: which communities bear the siting costs, and did they consent? And there is a moral layer: what does the present generation owe to people who will be alive in a thousand or ten thousand years, who will live with decisions made before they were born, about a substance they had no part in generating?

The most serious version of the no-new-waste argument is not that nuclear energy is dangerous — it is that a society generating a long-lived hazard without a demonstrated permanent solution is, in effect, transferring an unresolved obligation to the future. That obligation is not discharged by promising to solve it. The structural absence of future people — not merely from a planning commission meeting, but from any process we can design, across timescales that exceed anything human governance has sustained — is the deepest problem in this debate. None of the four positions has a fully satisfying answer to it. The geological repository position comes closest, because it is designed to work without institutional continuity. But building it requires the political will that has been absent for forty years.