How Top Secret Nuke Sensor Data Confirmed the First Interstellar Object on Earth

In 2014, a small meteor from another star system crashed into the Earth. It was the first known interstellar object to hit our solar system. The scientists who figured that out couldn’t definitively prove it until, one day in 2022, the Deputy Commander of the U.S. Space Command said it was true with “99.999% confidence” on Twitter. The details around how classified military data became part of a science paper are opaque and, thanks to new documents obtained by Motherboard via the Freedom of Information Act, we know at least some of the story.

In 2019, Harvard astrophysicists Amir Siraj and Avi Loeb attempted to publish a paper on the discovery of the meteor. But acceptance of the paper was delayed because of government secrets and labyrinthine bureaucracy. Some of the relevant data had been collected using sensors the Pentagon uses to track nuclear explosions, which made it highly classified.

So, government scientists were in the weird position of sitting on classified data that would help Siraj and Loeb prove they were studying the world’s first known interstellar object, predating the famous ‘Oumuamua by several years. Motherboard filed FOIA requests in an attempt to figure out what was going on two years ago. After numerous delays, we finally got some answers.

A series of emails between members of the Los Alamos National Lab from the spring of 2020 paint a picture of friendly government employees who wanted to help scientists prove something significant, but had their hands tied by bureaucracy and secrecy. Loeb’s name is unredacted in the emails, but everyone else’s has been censored.

The process began in April when Loeb emailed a contact at Los Alamos notifying them that his and Siraj’s paper had been rejected for publication because a reviewer could not be found. “Avi, I shot off an email to a potential resource,” the contact replied. “I hope to speak to [redacted] as well. Sorry for this circumstance, which I hope will be just a hiccup.”

It turned out to be much more than a mere hiccup. Emails flew back and forth for weeks, Zoom meetings were held, and discussions took place. According to the emails, COVID-19 played a part in the months-long bureaucratic merry-go-round.

“I’m sorry about this hiccup. I still hope we can work this w/ [redacted] new contact,” said an April 1, 2020 email. “The COVID-19 situation means that many clearance holders who would normally have classified phone/ computer access are not ‘mission essential’ and are now working from home, so can’t work this review!”

The powers that be hemmed and hawed. “As you may know, the unclassified CENOS database—derived from multiple government sources—has velocity and energy but no uncertainties since they would give away too much,” another April 1 email said. “Consequently the referees rejected the paper although the authors have made very reasonable assumptions to estimate uncertainties.” 

The matter came up again on April 14. “The topical area for this paper would be, generically, the nuclear detonation detection system and specifically the portion of the federal sensors looking down,” an email said. “The reviewer would be familiar with classified aspects behind the public-facing bolides data base…and preferably have access to the classified velocity uncertainties.”

On April 24, Los Alamos apparently threw in the towel. “I have exhausted my network. I spoke to a number of people who found this a fascinating challenge, but in the end we were unable to satisfy both the expertise and access, and willingness to comment in the open. I apologize,” said an email from an unidentified person at the lab.

A May 7 email picked up the thread, however, renewing hope of a solution. “An issue arose several months ago in the Board of Physics and Astronomy regarding an unclassified manuscript for review that requires access to classified information. Perhaps [Intelligence Science and Technology Experts Group] could help,” the email said.

“As you may know, the unclassified [Center for Near-Earth Object Studies] database—derived from multiple government sources—has velocity and energy data about meteoric objects in the atmosphere but contains no velocity uncertainties since they would give away classified information,” it continued. “Consequently the referees rejected the paper due to missing velocity uncertainties although the authors made reasonable assumptions to estimate them. The uncertainties are necessary to determine whether the meteor is of interstellar origin.”

The email went on to explain that they’d had trouble finding a person to help. “While this circumstance is unusual, there may be a general need for reviewers with clearances to review unclassified manuscripts. Perhaps more generally useful would be allowing a pathway for granting a single-event exception from the controlling agency,” it said.

Weeks later, on May 26, another unidentified person replied. “In principle we might be able to help. Some practicalities might get in the way. First, [The National Academies of Sciences, Engineering, and Medicine] can’t generate a classified question for the ISTEG. That request would have to come from [the Office of the Director of National Intelligence.] We could probably get support for this, but would have to go through a process to do so,” it said.

These emails are only part of the puzzle. Somehow, Siraj and Loeb’s plight made it all the way to the Deputy Commander of the U.S. Space Command, who decided to issue a public statement saying they were right about their research. How, exactly, the discussion made the leap from Los Alamos to Space Command is still unknown.

That doesn’t mean it’s the end of IM1’s story. Siraj and Loeb are currently engaged in a seafaring project to recover traces of IM1 in the Pacific Ocean. During a 2023 expedition, the team recovered tiny “spherules” that they concluded were likely extrasolar in origin, and the pair plan on making another excursion in 2024. 

FOIA requests are a powerful tool for shedding light on government secrets. Washington, the DoD, and its associated departments generate vast amounts of data that people can see if they just ask. That’s the pitch, anyway. In practice, the FOIA process can take years to complete as departments drag their heels. Everyone agrees this sucks, but no one has come up with a good way to fix it.

With files from Tim Marchman.