DOE looks hard at taking over deep underground site

After NSF was forced last December to throw in the towel on establishing an underground laboratory in the former Homestake gold mine in South Dakota, the US Department of Energy began analyzing the possibility of making something of the site. If it does, the facility would be smaller and would have a tighter scientific focus than NSF had planned. DOE will decide soon—in time for inclusion in the fiscal year 2013 budget—whether to drop Homestake, go full force, or fund it at an amount that keeps options open.

Since at least 2000, US scientists have been working to have a deep underground lab for experiments that need shielding from cosmic rays and background radioactivity. For DOE, those are the long baseline neutrino experiment (LBNE) and searches for neutrinoless double beta decay and dark matter. The multidisciplinary NSF plan for the Deep Underground Science and Engineering Laboratory (DUSEL) also included programs in nuclear astrophysics, geology, biology, and engineering, and efforts in public outreach and education (see PHYSICS TODAY, February 2011, page 21 ).

The LBNE aims to look for CP violation and to nail down which of three neutrino types is heaviest by studying oscillations among them; Fermilab has made the experiment its top priority. A beam of muon neutrinos from Fermilab would zip some 1300 km to a “far” detector at Homestake. Neutrino oscillations would be studied by comparing the observed neutrinos with the initial beam makeup determined by a “near” detector at Fermilab.

The LBNE collaboration has not yet decided whether to go with the mature water Cherenkov technology, which would require a 200-kiloton detector at Homestake, or with liquid argon, for which the detector would be about one-sixth the size and far more sensitive to particle interactions. But the liquid argon method needs engineering development and prototyping. “We’ll go with whichever gives us more physics for the money, once we know how much we have,” says LBNE project leader Jim Strait.

Cost, benefits, and US leadership

The cheapest option for DOE is to put the LBNE far detector at Homestake but put the double beta decay and dark-matter experiments in Canada’s SNOLAB. But that lab is not an option for the LBNE because it’s too close to Fermilab—765 km. Nor does it have a large enough cavern. The total cost for the LBNE, including infrastructure at both Fermilab and Homestake, would be $1.2 billion–$1.5 billion; the first additional experiment at Homestake would add about $100 million in infrastructure, and more would cost $15 million each. It would not be cost effective to put the dark-matter and double beta decay detectors in Homestake unless the LBNE is installed there. Those are among the conclusions reached by an independent assessment requested by DOE Office of Science director William Brinkman and released in late June. Carried out over three months, the assessment involved gathering and synthesizing information; no original cost estimates were made.

Despite the extra cost of collocating the three experiments at Homestake, Caltech’s Jay Marx, who cochaired the assessing committee, says, “We pointed out that programs will go on for several decades.” In a 23 June briefing to DOE’s High Energy Physics Advisory Panel, Marx noted the benefits of having a common facility serve as an intellectual center and said that “locating the facility in the US would promote US leadership in these fields for the foreseeable future.”

Adjusting to the switch

The House energy and water development appropriations bill cautions DOE against creating and managing a new national lab but encourages the department to maintain Homestake’s viability for science. DOE has requested $15 million for FY 2012 to keep the Homestake option afloat. Mike Headley, the site’s project director, who works for the South Dakota School of Mines and Technology, says his design team has shrunk to half its earlier size, but the $15 million would keep the “very knowledgeable” team intact and keep the mine from flooding. “After getting over the initial shock [of NSF pulling out], we certainly want to see the facility move forward,” he says. Some $70 million from philanthropist T. Denny Sanford plus about $50 million from the state of South Dakota have been invested in Homestake, and two physics experiments are getting under way. But an end to federal money would be the death knell for the site.

Headley’s team, together with the experiment collaborations, has begun modifying the lab design. A smaller Homestake facility would be built on a single level, 4850 feet (1480 meters) deep, instead of two or three levels as had originally been envisioned for DUSEL. It would still have a big cavern for the LBNE, but only one additional experimental hall, rather than two; that second hall could house up to three experiments.

The scaled-down option—which for political reasons is not referred to as a national laboratory—has been dubbed the Sanford Underground Research Facility, or SURF. The estimated infrastructure cost is $150 million, significantly less than the $575 million for NSF’s grander plan. The annual running costs for SURF are estimated at around $20 million, less than half the $50 million–$70 million for DUSEL.

On 12 July, the National Research Council released a report that will also figure into DOE’s decision on Homestake. The NRC evaluated the science proposed for DUSEL, the need for it in the context of similar science programs in other regions of the world, and its broader impact. The science has been reviewed several times over the past decade, and the NRC conclusions held no surprises: The three experiments that DOE is now considering are of “paramount and comparable scientific importance,” with each addressing “at least one crucial question upon which the tenets of our understanding of the universe depend.” The report also gives high marks to an accelerator-based study to measure low-energy nuclear cross sections. And it notes that a facility in the US would not only provide access and leadership advantages for the physics community, but could also benefit scientists in other fields.

If DOE does run with Homestake, it would be built incrementally. “DOE is more amenable to phasing things in, which allows us to come up with lower costs in preparing the facility,” says Kevin Lesko, who led the NSF-supported efforts for DUSEL and is continuing in that leadership role for SURF. He and others point out that DOE has more experience than NSF in running large projects. But, Lesko says, “We hope that NSF can be reengaged and that more experiments would eventually come to Homestake.”

Especially given cuts in federal spending, DOE’s decision is “complicated,” says Strait. “We don’t know whether to expect thumbs up, thumbs down, or thumbs sideways.”