By Eric Hand

The US Department of Energy (DOE) is planning to streamline the way it oversees the safety and security of its national laboratories. The move comes as a relief to lab directors who think that the existing oversight system is too onerous, but it is raising the hackles of groups that worry about breaches of nuclear security.

Although other federal agencies are regulated by the Occupational Safety and Health Administration, the DOE's 17 national laboratories--a few of which act as stewards for US nuclear weapons--are often held to higher standards by the department's Office of Health, Safety and Security (HSS).

A 16 March memo from Daniel Poneman, the deputy energy secretary, made public last week, outlines the plans for the HSS. It states that the department is aiming for "near-term relief from specific low-value burdensome requirements," and wants at least a 50 percent reduction in safety and security directives from the oversight office.

The Project On Government Oversight (POGO), a non-profit organization in Washington D.C., which obtained the memo, says that high levels of oversight should be strengthened, not streamlined. Peter Stockton, a POGO investigator, cited recent security failures such as a mock terrorist attack staged in 2008, which was able to penetrate defenses at Lawrence Livermore National Laboratory in California, a nuclear weapons lab. Stockton says the oversight changes will mean fewer and less-independent assessments. "We think this is all wrong," he says.

But Robert Rosner, former director of Argonne National Laboratory in Illinois, welcomes the changes. He says that the national labs--government-owned but independently operated--have suffered in an environment that tolerates mountains of red tape but accepts no risk. Laboratories have devoted increasing amounts of money and manpower to comply with the bureaucracy of DOE oversight. Rosner says it is much easier to conduct an experiment at the University of Chicago, which operates Argonne National Laboratory, than it is to do the paperwork needed to start an experiment at Argonne. Rosner says the changes are "a great step forward." He adds that the non-nuclear weapons labs, such as Argonne, shouldn't be judged by the same standards as the weapons labs.

Safety failings

The record of the science labs is not spotless. For instance, in 2004 at SLAC National Accelerator Laboratory in Stanford, Calif., a contractor was severely burned when a 480-volt electric arc ignited his clothes. The lab's accelerators were shut down for months, and an investigation blamed management culture.

Some at the lab feel that the response to the accident went too far. In the aftermath, all staff had to take hours of electrical safety training, regardless of whether their jobs put them near a socket. They also had to file annual assessments of the hazards of their job--even if that job was sitting down at a desk to work on theoretical physics.

Michael Lubell, director of public affairs for the American Physical Society in Washington D.C., says that eliminating oversight bureaucracy would be a way of increasing scientific productivity. "Are you happy if you have 98 percent compliance, or do you have to have 99.9 percent compliance?" he asks. "Each time you push it up one more notch, the efficiency is going to be affected and the costs are going to go up."

The labs will be given more authority to design their safety programs independently. Site visits by the federal office, say laboratory employees, are already switching from a culture focused on assessment, to one that aims to provide help in maintaining safety standards.

The extra independence does come with a price for the scientists, says Rosner. "They have to understand that now that DOE has done this, it's an additional level of responsibility on them to behave well."