Nuclear-Waste Arks Are a Bold Experiment in Protecting Future Generations
Designing nuclear-waste repositories is part engineering, part anthropology—and part mythmaking
This article is part of a package in collaboration with Forbes on time capsules, preserving information and communicating with the future. Read more from the report.IGNACE, Ontario, C.E. 51,500—Feloo, a hunter, chews a strip of roasted caribou flank, washing it down with water from a nearby lake. Her boots press into thin soil that, each summer, thaws into a sodden marsh above frozen ground. Caribou herds drift across the tundra, nibbling lichen and calving on the open flats. Hooves sink into moss beds; antlers scrape dwarf shrubs. Overhead, migratory birds wheel and squawk before winging south. Two lakes remain liquid year-round, held open by hidden taliks—oases of water in a frozen land. Beneath it all lies the Canadian Shield: a billion-year-old granite craton, a basement of rock, scarred by ice, that has endured glaciation after glaciation. In 10 or 15 millennia, Feloo’s world will vanish beneath three kilometers of advancing ice.Feloo is unaware that 500 meters below her feet rests an ancestral deposit of copper, steel, clay and radioactive debris. Long ago, this land was called Canada. Here a group known as the Nuclear Waste Management Organization (NWMO) built a deep geological repository to contain spent nuclear fuel—the byproducts of reactors that once powered Ontario, Quebec and New Brunswick. The vault was engineered to isolate long-lived radionuclides such as uranium 235, which has a half-life that exceeds 700 million years—sealing them away from war, disaster, neglect, sabotage and curiosity for as long as human foresight could reach.On supporting science journalismIf you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.NWMO issued reports with titles such as Postclosure Safety Assessment of a Used Fuel Repository in Crystalline Rock. These studies modeled future boreal forests and tundra ecosystems, simulating the waxing and waning of vast glacial ice sheets across successive ice ages. They envisioned the lifeways of self-sufficient hunters, fishers and farmers who might one day inhabit the region—and even the remote possibility of a far-future drill crew inadvertently breaching the buried canisters.Feloo was born into a world that has remembered none of this. Records of the repository were lost in the global drone wars of C.E. 2323. All that endured were the stories of Mishipeshu, the horned water panther said to dwell beneath the lakes—and to punish those who dig too deep. Some of Feloo’s companions dismiss the legend; others whisper that the earth below still burns with poison. Yet every step she takes is haunted by choices made tens of millennia before—when Canada undertook the Promethean task of safeguarding a future it could scarcely imagine.In 2024 NWMO announced that Canada’s deep geological repository for spent nuclear fuel would be built in the granite formations of northwestern Ontario, near the Township of Ignace and the Wabigoon Lake Ojibway Nation. The decision capped off a 14-year siting effort that solicited volunteer host communities and guaranteed them the right to withdraw at any stage of the process. NWMO is now preparing for a comprehensive regulatory review, which will include a licensing process conducted by the Canadian Nuclear Safety Commission. This means the development of impact assessments that will be specific to the Ignace site. NWMO has also pledged an Indigenous-led regulatory process alongside federal oversight, with the Wabigoon Lake Ojibway Nation conducting its own assessments to ensure the project reflects Anishinaabe principles of ecological stewardship. If approvals proceed, construction could begin in the 2030s, and the repository could go into operation in the 2040s.A deep-time repository, like a deep-space probe, must endure without maintenance or intervention, independently carrying human intent into the far future.A deep geological repository can be seen as a reverse ark: a vessel designed not to carry valuables forward in time but to seal dangerous legacies away from historical memory. Or it can be understood as a reverse mine: an effort returning hazardous remnants to the Earth rather than extracting resources from it. Either way it is more than just a feat of engineering. Repository projects weave together scientific reasoning, intergenerational ethics and community preferences in decisions that are meant to endure longer than empires. As messages to future versions of ourselves, they compel their designers to ask: What symbols, stories or institutions might bridge epochs? And what does it mean that we are trying to protect future humans who may exist only in our imaginations?I am a cultural anthropologist. From 2012 to 2014 I spent 32 months living in Finland, conducting fieldwork among the safety assessment teams for Onkalo—an underground complex that is likely to become the world’s first operational deep geological repository for spent nuclear fuel. The teams’ work involved modeling far-future glaciations, earthquakes, floods, erosion, permafrost and even hypothetical human and animal populations tens of millennia ahead. That research became the basis for Deep Time Reckoning, a book exploring how nuclear-waste experts’ long-range planning practices can be retooled as blueprints for safeguarding future worlds in other domains, from climate adaptation to biodiversity preservation.During the Biden administration, I joined the U.S. Department of Energy’s Office of Spent Fuel and High-Level Waste Disposition, where I helped advance participatory siting processes modeled on approaches that had proven successful in Finland and Canada. I served as federal manager of the DOE’s Consent-Based Siting Consortia—a nationwide coalition of 12 project teams from universities, nonprofits and the private sector that were tasked with fostering community engagement with nuclear waste management. Through it all, I came to see repository programs as civilizational experiments in long-term responsibility: collective efforts to extend the time horizons of governance and care so that shared futures may be protected far beyond the scale of any single lifetime or institution.An enduring question for all repository programs is whether—and, if so, how—to mark their sites and archive knowledge about them. There is no guarantee that the languages we speak today will remain intelligible even a few thousand years from now. Beowulf, written in an earlier form of English a millennium or so ago, already reads like a foreign tongue. The meanings of symbols drift just as unpredictably. A skull and crossbones, for instance, may denote poison, death, rebirth—or pirates—depending on culture and context. What, then, might a nuclear waste repository signify to people tens of millennia from now? How long can a warning sign, monument, or archive preserve the meanings we attach to it today? Or should we abandon the illusion of communicating with future humans like Feloo altogether—and instead build repositories that are meant to be forgotten?Nuclear organizations rely on familiar techniques to preserve institutional memory: documentation mandates, digital databases, mentoring pipelines, program redundancy, succession planning. Such mechanisms can sustain continuity for decades, even centuries—but their limits become clear when stretched across millennia. Archives can burn. Technologies can decay into obsolescence. Institutions can falter under political or economic upheaval. And today a new litany of planetary risks crowds the horizon: thermonuclear war, weaponized synthetic biology, climate-driven migrations, institutional collapse, even runaway artificial superintelligence.As NWMO prepares for construction in Ignace in the 2030s, the question of long-term communication must increasingly shift from theory to practice. Canada has participated in the Organization for Economic Co-operation and Development’s Nuclear Energy Agency’s Preservation of Records, Knowledge and Memory initiative, which has explored strategies ranging from warning markers to staged transfers of responsibility across generations. In a 2017 safety report, NWMO wisely conceded a limit: “repository records and markers (and passive societal memory) are assumed sufficient to ensure that inadvertent intrusion would not occur for at least 300 ... years.” Beyond that horizon, the premise changes. No monument, land-use restriction, monitoring system or archive can be trusted to endure indefinitely.Different countries have embraced different philosophies of how to safeguard nuclear waste repositories across centuries and millennia—and how and whether to try to send messages to those who, like Feloo, may one day live above them.The U.S. is home to the Waste Isolation Pilot Plant (WIPP), a deep geological repository carved into ancient salt beds in New Mexico. WIPP stores transuranic waste from the nation’s nuclear weapons programs. In the 1980s and 1990s, task forces convened scientists, artists, science-fiction writers and semioticians to design warning systems that were intended to deter drill crews or archaeologists living thousands of years in the future. Their proposals were dramatic: vast fields of concrete thorns bristling from the desert floor; monolithic slabs etched with multilingual warnings (“this place is not a place of honor ... nothing valued is here”); and signage depicting the anguished face of Edvard Munch’s The Scream. Some envisioned a quasi-religious “atomic priesthood” to preserve the warning through ritual. Others suggested bioengineered “ray cats” whose fur would fluoresce near radiation—accompanied by myths, songs and proverbs to ensure that unborn generations would know to flee.Finland’s Onkalo repository embodies a somewhat different a philosophy. Anticipating the future loss of institutional control and memory of the repository, Onkalo was designed to remain secure for millennia in the absence of monumental communication systems. As in Canada, the lack of exploitable resources in the granite bedrock is meant to deter future prospectors. Once its tunnels are packed with copper canisters and bentonite clay, Onkalo will be backfilled and sealed for perpetuity on a small, unassuming islet in the Baltic Sea sometime in the 2120s. The danger is to be buried so completely that there will be nothing left to remember: no attention-grabbing monoliths to tempt curiosity, no symbols to be misread. When I conducted anthropological fieldwork in Finland, some scientists likened the project to launching a probe into interstellar space: years of meticulous planning and testing culminating in a single, irrevocable release. After that, no repair or recall is possible. A deep-time repository, like a deep-space probe, must endure without maintenance or intervention, independently carrying human intent into the far future.Even the mightiest empires have cycled through collapse and renewal, through forgetting and rediscovery.France has charted a third path with its Cigéo repository, planned in the Callovo-Oxfordian clay of its northeastern departments of Meuse and Haute-Marne. A 2016 law requires Cigéo to remain reversible for at least a century after operations begin. In practice, reversibility means retrievability: the inbuilt capacity to recover waste packages from the underground deposition cells. Advocates see this as a balance between long-term containment and intergenerational agency: the idea that future citizens should retain the right to revisit, or even overturn, choices made today. This logic resonates with those who view spent nuclear fuel as a future resource more than a liability. Jenifer Schafer, an associate director for technology at the DOE’s Advanced Research Projects Agency–Energy, has argued that “nuclear treasure” may be a more fitting term than “nuclear waste,” as the fissile materials inside it could someday power future innovations in nuclear reactor design. From this perspective, burying spent nuclear fuel too conclusively risks foreclosing possibilities that future generations might prefer to keep open.Taken together, these examples reveal how differently societies imagine their obligations to the far future. The American strategy reflected a lingering cold-war-era faith—tinged with hubris—in design ingenuity to frighten descendants away. The Finnish plan entrusted geology with the work of erasure, even if humans’ memory were to lapse as the landscape quietly reclaimed the site. The French framework preserved the right of future citizens to reject the decisions of today. Canada still has regulatory milestones and First Nations approvals to meet before NWMO can break ground at Ignace. In the decades ahead, however, it, too, will have to specify how it will stage its approach to intergenerational communication.What is certain, though, is that NWMO’s deep geological disposal efforts will unfold not only as a technical project but also as a cultural statement—a statement about care across generations, the limits of understanding across difference and the moral responsibilities of present-day Canadians to those not yet born. Like all repository efforts, NWMO’s work in Ignace will serve as a mirror: a message not only to the future but also to the present, reflecting what we choose to remember, what we choose to forget and how we hope to be remembered ourselves.As NWMO refines its approach to remembering, forgetting and communicating with societies of the future, it would do well to look beyond the nuclear industry for inspiration.Japan’s Kongō Gumi construction firm, founded in C.E. 578, operated independently for more than 1,400 years before it became part of the Takamatsu Construction Group in 2006. Adapting across vast social and political transformations, the Catholic Church, France’s Hôtel-Dieu hospital (C.E. 651) and Morocco’s University of al-Qarawiyyin (C.E. 859) have each endured for more than a millennium. Bali’s subak irrigation system, established in the ninth century, continues to flourish through a network of water temples that unite ecological engineering with Hindu philosophy and ritual. In New Mexico, three-century-old acequia canals still function under community governance, with elected mayordomos overseeing water sharing through collective labor. In Australia, the Brewarrina fish traps have been maintained across countless generations of Aboriginal peoples. What principles of intergenerational adaptation, renewal or continuity might NWMO glean from such long-lived systems?The Memory of Mankind (MoM) project in Austria could also be instructive. MoM’s mission is to preserve a snapshot of human civilization for the distant future, a cultural time capsule designed to outlast war, decay and digital obsolescence. Deep inside the Hallstatt salt mine, MoM stores ceramic tablets engraved with texts and images engineered to resist heat, radiation, chemicals and water. Its archive includes everything from scholarly works to recipes and personal stories. Led by ceramist Martin Kunze, MoM represents a philosophy of strategic redundancy. To guard against loss, Kunze distributes miniature tablets worldwide, each etched with maps pointing back to the Hallstatt archive—a physical embodiment of a principle articulated by the digital-preservation project LOCKSS: “Lots of Copies Keep Stuff Safe.” What might it mean for Canada to apply that same principle to the challenge of nuclear memory?Indigenous cultures offer another paradigm of long-term message endurance: storytelling as recordkeeping. Aboriginal Australian oral histories recount volcanic eruptions in western Victoria that align with geological evidence dating back nearly 37,000 years. Narratives describing islands drowned by rising seas have likewise been corroborated by climate science. Such traditions demonstrate that oral knowledge of environmental change can persist across timescales that far exceed those of our most advanced digital media, which often decay or become unreadable within decades. What might NWMO learn from cultural systems of memory grounded in ceremony, cosmology and story transmission?If built properly, NWMO’s deep geological repository will outlast governments, economies and the very languages that name it. It will join a global lineage of reverse arks: monuments to societies that dared to think beyond themselves. If the facility is someday uncovered by a far-future archaeologist, its depth, placement and engineered barriers could reveal what our civilization judged to be dangerous, how we calculated risk and how we imagined future humans would think, live and interpret signs. Yet scientific literacy cannot be assumed across deep time. Even the mightiest empires have cycled through collapse and renewal, through forgetting and rediscovery. To posterity, a nuclear waste repository might be read as a sacred monument, an extraterrestrial stronghold, a strange geological formation, a chamber of forgotten gods—or something beyond our present-day imagination altogether.In the end, Canada’s proposed Ignace repository will be an artifact of our own self-understanding: stone and metal fashioned into a signal meant to traverse vast orders of time. Its interpretation will belong solely to the future—to whatever beings, human or otherwise, may one day unearth what we once chose to hide.
