Douglas Birch / The Center for Public Integrity

WASHINGTON — When geophysicist H. Jay Melosh attended a meeting of U.S. and ex-Soviet nuclear weapons designers in May 1995, he was surprised by how eager the Cold Warriors were to work together against an unlikely but dangerous extraterrestrial threat: asteroids on a collision course with Earth.

After Edward Teller, the father of the hydrogen bomb, urged others meeting at the Lawrence Livermore National Laboratory in California to consider building and orbiting huge, new nuclear weapons for planetary protection, some top Russian weapons experts lent their support.

“It was a really bizarre thing to see that these weapons designers were willing to work together — to build the biggest bombs ever,” said Melosh, an expert in space impacts who has an asteroid named after him.

Ever since, he has been pushing back against scientists who still support the nuclear option, arguing that a non-nuclear solution — diverting asteroids by hitting them with battering rams — is both possible and far less dangerous.

But Melosh’s campaign suffered a setback last month when Energy Secretary Ernest Moniz signed an agreement with Russia that could open the door to new collaboration between nuclear weapons scientists in everything from plutonium-fueled reactors to lasers and explosives research.

A Sept. 16 Department of Energy announcement cited “defense from asteroids” as one potential area of study.

President Barack Obama has committed the United States to seeking a world without nuclear weapons. But NASA is spending hundreds of thousands of dollars a year to study their use against asteroids, and the U.S. nuclear weapons labs appear to be itching to work with their Russian colleagues on the problem.

Moreover, weapons experts in both countries are citing the asteroid threat as a reason to hold onto — or to build — very large-yield nuclear explosives, which have declining terrestrial justification.

David Wright, co-director of the Union of Concerned Scientists’ Global Security Program, said he hoped any joint asteroid defense work would not become a “jobs program” for weapons scientists.

“When you’ve got the weapons labs sort of pushing for this in the various countries, it starts to make me feel a little uneasy,” he said. “Which doesn’t mean it’s not a legitimate thing to do, but you want to know it’s being done for legitimate reasons.”

Bong Wie, the director of Iowa State University’s Asteroid Deflection Research Center, said he has a three-year, $600,000 grant from NASA to design a “hypervelocity nuclear interceptor system,” basically an ICBM-borne warhead fitted with a battering ram.

The ram would separate from the bomb before impact, gouging a crater in the asteroid so the bomb could then blast it to bits.

Keith Holsapple, an engineering professor at the University of Washington, said NASA has given him a five-year, $1.25 million research grant to study how either an impact device or a nuclear explosion could deflect an Earth-bound asteroid from its path.

The leading supporter of the nuclear solution in the United States is probably David Dearborn, a research physicist and weapons designer at Lawrence Livermore who is presently helping refurbish the U.S. nuclear arsenal.

Wie called Dearborn a senior figure among scientists studying the nuclear option. “I am just following in his footsteps,” he said.

Dearborn said that he was offended years ago when other researchers told reporters that nuclear weapons simply couldn’t work against asteroids. “That’s just not true,” he said, calling these claims scientifically “indefensible.”

For years, Dearborn worked to refute the skeptics on his own time. Since 2012, he said, he and a colleague have had a grant from Lawrence Livermore worth several hundred thousand dollars to work on the project part time, he said. He estimated that about a dozen other scientists have studied aspects of the approach at U.S. weapons labs.

Nuclear weapons could be used in two ways, he said. When the collision is still a decade or longer away, a “standoff” nuclear blast could knock the asteroid off course. When the time to impact is short, he said, defenders would try to blow up the asteroid.

“You fragment it with enough force so that the pieces spread out,” and most miss the Earth, he said. Small bits of rock would burn up in the atmosphere.

There would be no need to build new weapons or test old ones, he said. But shattering a large asteroid close to hitting Earth probably would require a weapon with a yield of about a megaton, or 1 million tons of TNT, he said, which is roughly the power of the largest in the current U.S. arsenal.

Mark Boslough, a physicist at Sandia National Laboratories in New Mexico, another weapons lab, has worked part time in the past on modeling the effects of nuclear blasts on asteroids. He agrees that, on short notice, nuclear “really is the only option.”

Melosh disagrees. He was co-investigator on a 2005 NASA mission known as Deep Impact, which launched an 820-pound copper-covered battering ram that gouged a crater out of the comet Tempel 1 in 2005. He says that 90 percent of the biggest asteroids already have been found and ruled out as a near-term threat, demonstrating there is time for finding suitable, non-nuclear alternatives — such as hitting asteroids with rams, zapping them with lasers, tugging them off a kamikaze trajectory, or deflecting them with solar sails.

Even as a last-ditch effort, he said, nuclear weapons can’t work using existing warheads but only with new, even larger nuclear explosives than exist in any arsenal. “A lot more people have been recorded to have died from nuclear weapons than have been recorded to have died from asteroid impacts,” he warned.