For years before North Korea fired its first intercontinental ballistic missile this week, the Pentagon and intelligence experts had sounded a warning: Not only was the North making progress quickly, spy satellite coverage was so spotty that the United States might not see a missile being prepared for launch.
That set off an urgent but quiet search for ways to improve America’s early-warning ability — and the capability to strike missiles while they are on the launchpad. The most intriguing solutions have come from Silicon Valley, where the Obama administration began investing in tiny, inexpensive civilian satellites developed to count cars in Target parking lots and monitor the growth of crops.
Some in the Pentagon accustomed to relying on highly classified, multibillion-dollar satellites, which take years to develop, resisted the move. But as North Korea’s missile program progressed, American officials laid out an ambitious schedule for the first of the small satellites to go up at the end of this year, or the beginning of next.
Launched in clusters, some staying in orbit just a year or two, the satellites would provide coverage necessary to execute a new military contingency plan called “Kill Chain.” It is the first step in a new strategy to use satellite imagery to identify North Korean launch sites, nuclear facilities and manufacturing capability and destroy them pre-emptively if a conflict seems imminent.
Even a few extra minutes of warning might save the lives of tens of thousands of Americans — and millions of South Koreans and Japanese who already live within range of the North’s missiles.
“Kim Jong-un is racing — literally racing — to deploy a missile capability,” Robert Cardillo, the director of the National Geospatial-Intelligence Agency, which coordinates satellite-based mapping for the government, said in an interview days before North Korea’s latest launch. “His acceleration has caused us to accelerate.”
The timeline for getting the satellites in orbit, which defense officials have never discussed publicly, reflects the urgency of the problem. The missile launch by North Korea on Tuesday was initiated from a new site, a mobile launcher at the Pang Hyon Aircraft Factory. Capt. Jeff Davis, a Pentagon spokesman, said the missile “is not one we have seen before.”
That mobility is the problem that the new satellites, with wide coverage using radar sensors that work at night and during storms, are designed to address. Less than one-third of North Korea is under spy satellite coverage at a given moment.
American intelligence analysts detected indications of an impending launch in the days before the missile firing, according to a spokesman for the Defense Intelligence Agency, Cmdr. William Marks. But even after the launch, the Pentagon misjudged what it was looking at. Minutes after its 37-minute flight ended, the United States Pacific Command described the missile as an intermediate-range model, often seen.
Hours later, Secretary of State Rex W. Tillerson issued a very different conclusion: that the North had tested its first intercontinental ballistic missile, able to reach Alaska.
The commercial radar push is one of several new ways the administration is seeking to counter the North Korean threat. President Trump inherited a secret effort to sabotage the North’s missile launches. But its success has been spotty at best, especially of late.
And joint American-South Korean missile tests, conducted hours after the ICBM test, appeared to be part of the new strategy that includes Kill Chain — the missiles were designed to reach Pyongyang, where the country’s leadership lives.
Kill Chain was also mentioned in a joint statement issued last week by the United States and South Korea, a notable shift for the South’s new president, Moon Jae-in. He has rejected public discussion of pre-emptive military action, arguing it plays into the North Korean paranoia that the United States and its allies are plotting to end the Kim government.
Mr. Moon has spoken of reviving direct talks — a so-called sunshine policy, which he advocated as chief of staff to an earlier South Korean president.
But Mr. Trump has tried to build pressure, using warships, sanctions and missile defenses. He was recently presented with new options, including military ones, for responding to a sixth nuclear detonation by the North or a test of a missile that could reach the United States.
“The threat is much more immediate,” H.R. McMaster, Mr. Trump’s national security adviser, told a conference last week at the Center for a New American Security in Washington. “So it’s clear that we can’t repeat the same approach — failed approach — of the past.”
The new satellite initiative builds on technology created more for Wall Street than the Pentagon. From an office in an old Defense Department building within view of the Google campus in California, Raj Shah, the director of the Defense Innovation Unit Experimental, or DIUx, is already investing in companies that exploit tiny civilian radar satellites, able to pierce darkness or storms, in hopes that the Pentagon can use them by the end of the year, or early in 2018.
“It’s a very challenging target,” said Mr. Shah, a former F-16 pilot in Iraq whose extensive experience in Silicon Valley appealed to Defense Secretary Ashton B. Carter, who set up the unit during Mr. Obama’s second term and recruited Mr. Shah.
“The key is using technologies that are already available, and making the modifications we need for a specific military purpose,” Mr. Shah said.
His unit made an investment to jump-start the development efforts of Capella Space, a Silicon Valley start-up named after a bright star. It plans to loft its first radar satellite late this year. The company says its radar fleet, if successfully deployed, will be able to monitor important targets hourly.
“The entire spacecraft is the size of a backpack,” said Payam Banazadeh, a founder of the company. Born in Iran, he learned satellite design at the University of Texas and NASA’s Jet Propulsion Laboratory, specializing in miniaturization.
Once in orbit, the payload, he added, would unfurl its antenna and solar panels.
“Everything is getting smaller,” Mr. Banazadeh said of the craft’s parts. “Even the next version of the satellite is getting smaller.”
Seeing the early fruits of the Pentagon experiment, the National Geospatial-Intelligence Agency is opening its doors to companies that can supply it with satellite radar data in addition to traditional images. Its outpost, set up this year, is in San Jose, the heart of Silicon Valley.
Federal officials rarely, if ever, acknowledge the poor reconnaissance coverage of the North from traditional military satellites. But William J. Perry, the former secretary of defense, recently said in Washington that if the North rolled out a missile to hit the United States or its allies, “there’s a good chance we’d never see it.”
The threat grew worse last year as North Korea began using solid fuels after decades of relying on liquid propellants to power its big rockets and missiles. While liquid-fueled missiles can take hours or even days of preparation, solid-fueled missiles can be fired with little or no warning.
Mr. Kim has made the effort a personal project, posing next to a large solid-fueled motor after a successful ground test last year. The North followed that firing with four successful flight tests, twice last year and twice this year.
The advances, said Young-Keun Chang, director of the Global Surveillance Research Center at the Korea Aerospace University in Seoul, moved the North significantly closer to a mobile intercontinental missile that could eventually pose “a serious potential threat to the United States.”
The key to detecting launch preparations is the near-constant presence of satellites that can see through clouds, rain, snow, foliage and camouflage and can detect the movement of military gear, including missiles. That requires space-based radars, which over the years have been highly expensive, with their big antennas and tendency to use large amounts of power. Like any radar, they fire radio waves at targets and gather faint echoes.
Space-based radars can also detect changes in ground elevation that signal hidden tunnels, bunkers and even radioactive cavities left by nuclear blasts, experts say, because such hollows cause the surface above them to subside ever so slightly.
But building the radars has historically been expensive for the government.
In 2007, the Congressional Budget Office estimated that a constellation of 21 radar satellites would cost the nation up to $94 billion — or more than $4 billion each. The report, published shortly after the North’s first nuclear detonation, zeroed in on whether the satellites could track Korean missiles on mobile launchers. It called the goal “highly challenging,” and said 35 to 50 spacecraft would be needed to make such detections rapidly.
The new generation of tiny, cheap satellites has made that outcome more achievable. Capella plans to loft its first radar satellite late this year and build up to 36 orbital radars, within the range the congressional report recommended.
In addition to Capella, private companies rushing to make and exploit new generations of small radar satellites include Ursa Space Systems in Ithaca, N.Y.; UrtheCast in Vancouver, Canada; and Iceye in Espoo, Finland. Like many new companies seeking to make small satellites, most have strong ties to Silicon Valley.
The National Geospatial-Intelligence Agency’s initiative, known as the Commercial Geoint Activity, builds on programs in which the agency bought radar-satellite data from Canada, Italy and Germany as part of its evaluation of the new civilian technologies.
Mr. Cardillo said the new partnerships could help the United States close the gaps in tracking Mr. Kim’s rapidly expanding arsenal of threatening missiles.
“If any of these companies, new or old, can help fill those gaps,” he said, “then I’m interested.”
