Blasting and boring through a warren of tunnels in the abandoned Homestake Gold Mine in South Dakota, engineers are preparing for the installation of the Deep Underground Neutrino Experiment (DUNE), the U.S.’s latest, greatest major particle physics project. But things are not proceeding as planned: construction-related setbacks have delayed DUNE’s full-scale debut from sometime later this decade to, at best, the mid-2030s. And, as DUNE’s schedule has slipped, so too has its competitiveness with other neutrino experiments, leading the U.S. Department of Energy (DOE) to announce last fall its controversial decision to split the megaproject into two phases.

That two-phase approach, particle physicists in the U.S. hope, will help this flagging flagship project to keep pace with rivals while also creating breathing room to reimagine the later stages of its design. “True success might not be exactly on the same path as we were thinking 10 years ago,” says Kate Scholberg, an experimentalist at Duke University.

Together with the Long-Baseline Neutrino Facility (LBNF), DUNE seeks to interrogate the most elusive particles in the Standard Model of particle physics, which many suspect are a portal to whatever theory comes next. But last year, the megaproject’s price tag was reevaluated to more than $3 billion for the first phase alone—roughly double the original estimate for the entire endeavor.

Excavating 800,000 short tons of rock to make room for four hefty detection modules has proven more complicated than anticipated. “Incorrect assumptions or premature estimates” about the condition of the mine meant that infrastructure had to be refurbished before excavation could begin, said a spokesperson for DOE. The cost of installing detectors was also underestimated. As a result, switching on LBNF/DUNE has been set back by several years, and the project must be reaffirmed by Congress given the size of the budgetary overrun.

Now as particle physicists gather for Snowmass, a process with periodic meetings for thrashing out the coming decade’s research priorities in the U.S., some have raised the specter that LBNF/DUNE could come second, or even third, in scoring some of its motivating science objectives. “How do we change the way that we’re thinking about DUNE to do those goals and more, or do those even better?” asks Jonathan Asaadi, an experimentalist at the University of Texas Arlington. “The whole point of the Snowmass process is to have these very hard discussions and then build consensus.”

From Neutrinos to New Physics

Neutrinos are, without question, the weirdest denizens of the vast and varied particle-physics bestiary: They come in three types, but somehow oscillate between these different forms as they travel. This is surprising, as neutrinos can only oscillate if they have mass—a property that conflicts with predictions from the otherwise highly successful Standard Model. Ever since the discovery of this shape-shifting behavior in 1998, physicists have struggled to pin down exactly how neutrinos oscillate, and three missing measurements of crucial parameters remain.

The first and most well-known parameter, charge-parity (CP) violation, dictates whether neutrinos and their antiparticle counterparts oscillate in the same way, and could help explain why there is more matter than antimatter in the universe. The second determines which are the heaviest and which are the lightest types of neutrino. And the third is related to how likely it is for one type of neutrino to turn into another type. Physicists dreamed up LBNF/DUNE a decade ago, during an earlier Snowmass jamboree, as a way to measure all of this and more.

The motivations were much the same for Japan’s competing neutrino experiment Hyper-K— an even bigger underground chamber that is filled with water instead of DUNE’s liquid argon. Hyper-K aims to rigorously measure CP violation in the late 2020s, before LBNF/DUNE’s now-delayed neutrino beam would even turn on. Meanwhile, combinations of several smaller neutrino experiments—namely IceCube, JUNO and KM3NeT—are expected to weigh in on these and other longstanding neutrino puzzles over the coming decade. “DUNE runs a very serious risk of not measuring these parameters first,” Asaadi says. “Historically, big projects in Japan have been able to stay to their schedule with much more fidelity than many big U.S. projects.”

Yet Asaadi and many others emphasize that these experiments are collaborators as much as competitors. “We’re all kind of rooting for each other and rooting against each other at the same time,” says Mark Messier, an experimentalist who worked on Super-K (Hyper-K’s predecessor) as a graduate student and now works on DUNE and other neutrino detectors.

For one, any discovery must be confirmed by another independent experiment to be taken seriously. “For the sake of the science, we don’t really care about who gets the answer,” says André Luiz De Gouvêa of Northwestern University. “But you don’t want to have a big time difference between the projects.… One project can steal the thunder of another.” And although high-minded physicists may accept the importance of validation instead of being first to a discovery, the concept of coming in “second place” is a harder sell for politicians who hold the purse strings.

Then again, the new phasing of LBNF/DUNE may be what ultimately keeps the project healthy and competitive. “You make sure that you’re getting results, rather than allowing everything to delay while you wait for the full scale,” says Gabriel Orebi Gann of the University of California, Berkeley. “The only way that you guarantee losing the race is not even running it.”

Yet this approach comes with a new risk: funding for the second phase is not assured. “People are being invited to think about phase two and what it’s bringing to the table a little bit more carefully than before,” de Gouvêa says. “You want to make sure you have a good story to tell, to convince yourself and everybody else that phase two is a worthwhile investment.”

Sunk-Cost Fallacy or Golden Opportunity?

As large-scale projects like LBNF/DUNE have ramped up over the last five years, Congress has increased the DOE’s overall budget for high-energy physics by nearly 30 percent. Funding for core research at universities as well as for R&D on new accelerator and detector technologies, however, has declined. The new phasing of LBNF/DUNE aims to rebalance the budget, said a spokesperson for the DOE. “The community really has to be serious about what are the reasonable timelines and goals,” Asaadi says, or the megaproject could “suck all the air out of the room.”

All of this is occurring during a turbulent period for the LBNF/DUNE’s host, the Fermi National Accelerator Laboratory (Fermilab). Last fall, Fermilab’s director Nigel Lockyer stood down for undisclosed reasons. Shortly thereafter Fermilab also received a rare “C” from the DOE in its Report Card for science and technology project management.

Despite the uncertainty, sticking to the original scope of LBNF/DUNE is to many still a no-brainer. “It’s hard to imagine a single neutrino oscillation experiment that is better,” says Peter Denton, a neutrino theorist at Brookhaven National Laboratory. Unlike other experiments, the neutrino beam would probe a wide range of energies, thus taking a fuller picture of how neutrinos are oscillating. Detectors filled with liquid argon can also track the paths of particles far more precisely than those containing water. “Liquid argon technology is difficult, but we get something for that difficulty,” Denton says. The hope is that this novel and ambitious setup will not only map out neutrino oscillations in high resolution, but that signs of new particles and forces could show up, and physicists may at last grasp the baffling origin of neutrino mass.

For these reasons, in 2014, the particle physics community rallied behind novel liquid argon detectors over tried and test water detectors—and most neutrino physicists see no reason for that to change. “There is very strong support within the community for [LBNF/DUNE] to happen,” says Orebi Gann. Yet in internal documents seen by Scientific American, a current co-spokesperson for DUNE successfully ran for election earlier this year on the basis that “LBNF/DUNE is currently experiencing a poor acceptance in the [high-energy physics] community … seriously challenging the future of DUNE.”

Most everyone agrees that with billions of dollars already allocated to the megaproject—by the U.S. and international partners—there is no turning back. “The ‘sunk cost’ fallacy is always present when you’re this far down the road,” Asaadi says. Luminaries of the particle physics community are haunted by the cancellation of the Superconducting Super Collider, a multibillion-dollar particle accelerator, in the early 1990s. Congress pulled funding after the budget ballooned and dubious spending on costly parties and catered lunches was revealed. As a result, “particle physics moved to Europe,” says Francis Halzen, principal investigator of the IceCube neutrino experiment. “Hopefully everybody has learned that by killing a project, the money doesn’t return to you, or even to science.”

Then again, unquestionably supporting a major project whose ‘world’s first’ aspirations may no longer be achievable carries risks, too. “We are in a catch-22. Cancellation of DUNE would be a black-eye to the credibility of high-energy physics,” an anonymous source and member of the DUNE collaboration told Scientific American. “We need to find a way out of this, and the way out isn’t obvious.”

All Together Now

At Snowmass meetings, a balancing act is now underway to bring particle physicists resolutely together again around LBNF/DUNE, with some portraying the newly phased design as an opportunity to turn lemons into lemonade. As well as upgrading the neutrino beam and detector size, the second phase also leaves two of DUNE’s four modules undefined —and fresh ideas are welcome. “There’s a lot of excitement about that, and a lot of creative ideas for the new detectors,” says Scholberg, who co-convenes the neutrino physics priority group for the latest Snowmass, which continues through this summer.

Theoretical insights in recent years have opened the door to new kinds of dark matter searches at LBNF/DUNE, while the to-be-determined detectors could also house a “neutrinoless double beta decay” experiment to look for evidence that neutrinos are their own antiparticle. Enhanced supernovae detectors and the pursuit of mysterious sterile neutrinos are also on the table for filling the two open detector slots. “If you enlarge that collaboration [to other areas of physics], that’s bringing in more people that can make this happen,” says Jocelyn Monroe, a dark-matter experimentalist at Royal Holloway, University of London.

Others prefer to double down on DUNE’s world-best ambitions, giving priority to advances in detector designs and analysis techniques that would not normally be considered in a race to come first. “A lot of really good ideas tend to get pushed away because DUNE is ‘on rails’: it has to work, and it has to be done this way,” Asaadi says. One novel proposal, called THEIA, combines a water-based detector, like Hyper-K, alongside a “scintillation” detector that is more akin to DUNE—to reap the benefits of both approaches.

Among physicists, there seems to be universal agreement on one thing: The stakes on turning DUNE’s sunk-cost fallacy into an opportunity are high. “What we’re really working on very hard right now is trying to establish all these connections with the rest of the Snowmass community,” Scholberg says. Discussions will continue through the summer and then feed into decisions of the Particle Physics Project Prioritization Panel, a U.S. federal advisory group, which next year will decide whether DUNE/LBNF remains a flagship U.S. project. “We need to make sure that everybody is on board,” de Gouvêa says. “Uncertainty is bad for a very long-term project because people have to invest a large fraction of their time on it…. You don’t want the effort to be in vain.”