On the second day of a conference on cosmic threats to our planet, the proceedings were interrupted by an urgent message from Paul Chodas, the manager of NASA's Center for Near-Earth Object Studies.
New calculations suggested there was a 10 per cent chance that an asteroid named 2019 PDC would strike Earth in eight years, unleashing enough energy to level a whole city.
Scientists didn't know where it might hit, though New York, Denver and a wide swath of west and central Africa were all in the path of potential destruction. Chodas called the situation "uncertain" and "unprecedented."
So, he asked his audience, what did they want to do about it?
Before you start stocking up on canned goods know that this was a fictional exercise. The asteroid 2019 PDC does not exist. No city on Earth is thought to be imperiled by a catastrophic impact. Indeed, analyses of more than 20,000 known near-Earth objects suggest the chance of any hitting us in the next century is less than 1 in 10,000.
But the scenario that played out this week at the International Academy of Astronautics' Planetary Defense Conference in College Park illuminated the very real (if very slim) possibility that such an asteroid might be discovered one day - and revealed just how hard it would be for humanity to mount a response.
Congress first mandated that NASA track near-Earth objects, or NEOs - space rocks that circle around the sun and come within 30 million miles of Earth's orbit - in 1998, after people briefly panicked over a newly discovered asteroid that would pass by our planet.
Two decades later, scientists say they have identified 90 per cent of all NEOs 1km or larger - big enough to precipitate a global catastrophe. Research suggests these kinds of impacts happen once every 700,000 years.
The population of smaller objects such as the fictional 2019 PDC, which was estimated to be about 180m wide, is not so well defined - even though these objects could still demolish cities, states and even continents. This is the seventh tabletop exercise NASA has participated in to help game out what scientists and emergency managers would need to consider if one of these asteroids was headed our way.
But the agency's top concern this week seemed to be making sure everyone understood this was a drill. "If you tweet about this," communications officer JoAnna Wendell urged the audience, "please use the hashtag 'exercise only.' We don't want to get into a 'War of the Worlds' scenario."
Her words were comforting to keep in mind as NASA programmer Lorien Wheeler took the stage to explain just how catastrophic an asteroid such as 2019 PDC could be.
She had modeled millions of potential impact scenarios for the object, each featuring slightly different parameters. The asteroid might land in the Atlantic Ocean, triggering a catastrophic tsunami; or it could crash into New York, inflicting "unsurvivable" carnage on millions of people; or it could break up over a largely uninhabited area, causing relatively little damage. And it was still more likely than not that the rock would miss Earth altogether.
Scientists would need more than a year of observations before they could say exactly where the asteroid was headed. But it takes several years to build and launch a mission in response, and any deflection effort would have to happen before 2025 to be effective.
If humanity was going to try to stop the asteroid, we had to start considering our options now.
In reality, the discovery of any rock with a 10 per cent chance of hitting Earth would trigger an automatic response from United Nations' Space Mission Planning Advisory Group -- an international coalition of space agencies whose sole job is to coordinate the world's response to impending asteroid disasters.
Preventing an impact is possible - theoretically. Humans need only change the asteroid's velocity by a few centimetres per second; over the course of several orbits around the sun, that change adds up to push the rock fully in front of or behind the Earth. But the proposed methods for deflection are expensive and untested.
One, the "kinetic impactor" technique, involves crashing a spacecraft into the asteroid at high velocity to slow it down. (NASA last year gave the go-ahead to start design and assembly of a kinetic impactor test mission called DART, which will smash a spacecraft into a binary near-Earth asteroid called Didymos.)
Alternatively, scientists could detonate a nuclear bomb beside the asteroid, vaporizing part of its surface and causing the rock to recoil. This method is equally effective at increasing or decreasing an asteroid's speed.
Both options raised red flags among the Planetary Defense Conference attendees. The kinetic impactor strategy carries the threat of imparting so much force to the asteroid that it "disrupts" it, causing the rock to break apart into potentially even more dangerous pieces. But few people were excited about the notion of putting a nuclear device on top of a rocket. Which country would provide the weapon? And who would get to control it?
One man stood up to point out that there were no scientists at the conference from any of the African nations in the asteroid's path.
"In a real-life scenario, their interests would need to be represented," he said. "Especially if we're going to push the thing toward them."
Ultimately, the group decided to keep every option on the table. They would immediately launch a reconnaissance mission to fly past the rock and get a better understanding of its trajectory. A scientific spacecraft already in flight would be rerouted to 2019 PDC. Meanwhile, several space agencies would begin work on a fleet of kinetic impactors and an additional spacecraft capable of carrying a nuclear device.
In a speech, NASA administrator Jim Bridenstine acknowledged that these exercises could seem a bit outlandish - fodder for a Bruce Willis action flick, perhaps, but not serious scientific discussion.
But real life has at times come perilously close to imitating these fictions.
'DINOSAURS DIDN'T HAVE A SPACE PROGRAM, BUT WE DO'
Fifteen years ago, scientists detected a 300m-wide asteroid named Apophis, which early calculations suggested had a more than 2 per cent chance of colliding with Earth on Friday, April 13, 2029.
The discovery launched frantic efforts to uncover old pictures of where the asteroid had been, which would help researchers understand where it might be headed. Ultimately, scientists determined that Apophis will fly safely past Earth at a distance of about 31,000km - within the orbits of the moon and even some geosynchronous satellites.
If the skies are dark and weather are right, Earthlings should be able to see Apophis swish by with their naked eyes. The asteroid will resemble a shooting star as it streaks through space.
The geologic record reveals that Earth hasn't always been so lucky. The Chesapeake Bay was formed by a meteor that struck the East Coast 35 million years ago, scattering debris from Texas to New Jersey. Some 15 million years after that, a meteorite blasted the sand of the Libyan desert, creating a sea of foggy green glass.
And of course, there's the Chicxulub impactor 66 million years ago, which triggered the extinction of three quarters of all life on Earth, including most of the dinosaurs. That rock was thought to be dozens of miles in diameter and unleashed the energy of billions of atomic bombs.
"We know for a fact the dinosaurs did not have a space program," Bridenstine said, quoting television personality Bill Nye. "But we do, and we need to use it."
LOOKING INTO THE FUTURE
By Wednesday, the scenario fast-forwarded to December 2021. The reconnaissance spacecraft had finally given astronomers their first good look at 2019 PDC, revealing that the asteroid was actually a peanut-shaped "rubble pile" about 137m across and 260m long.
And, unless humanity took drastic action, just after 10pm local time on April 29, 2027, it would enter Earth's atmosphere at a speed of 69,200 km/h and smash right into Denver.
The resulting blast wave would obliterate the city and some of the surrounding suburbs - setting fire to buildings, vaporizing vegetation, melting cars. People as far away as Pueblo, Colorado, and Laramie, Wyoming, would feel the Earth tremble and see their windows break. The drought-parched trees of the Rockies would be set ablaze and billions of dollars worth of infrastructure would be lost.
Worst of all, there were some 2 million people living in the "unsurvivable" zone.
Horrific though the scenario sounded, physicist Mark Boslough, a global catastrophe expert, chose to look on the bright side.
"That asteroid was headed toward Denver even before it was discovered," he said. "So it seems like very bad news, but this is actually good news, because we still have five years to deflect this thing."
Between Wednesday night and Thursday afternoon, the clock spun forward three years. The diverted scientific spacecraft arrived at the asteroid and began to observe it. Six kinetic impactors were built, but one blew apart during launch and two others failed en route to their target. Widespread controversy over the use of a nuclear device meant that the second rendezvous spacecraft launched without any cargo. But that may have been for the best: It, too, suffered a total system failure before arrival.
Now it was September 3, 2024, and the world found itself facing a whole new kind of disaster: The three surviving kinetic impactors succeeded in deflecting most of the asteroid. But in the process they'd broken off a 50m-to 80m fragment that was still headed our way.
The resulting debris cloud destroyed the observing spacecraft, and the asteroid was behind the sun, blocking our view of it. Yet early calculations suggest that the fragment would strike somewhere along a path that extended from Omaha to New York and into the Atlantic. The object was now too small to pose a significant tsunami risk, Wheeler said, but if it hit land -- particularly a densely-populated place such as New York -- as many as 11.5 million people would be affected. And it was too late to attempt another deflection.
The prospect of losing a global financial center was shocking, even as make-believe. "You may freeze the economy," one audience member worried. "You might actually be looking at total economic collapse."
Aerospace engineer Brent Barbee agreed. That's why his team was considering a "last-ditch effort" to prevent an impact. The world might "scramble," he said, to design and deploy a 300-kiloton nuclear device that would blast the asteroid into tiny pieces two to four months before it was due to hit Earth.
The proposal immediately raised concerns.
"If you couldn't launch a nuclear device before, what makes you think you could launch it now?" one researcher wanted to know. ("Great question," was Barbee's reply.)
"Can we do a small kinetic impactor that puts it safely in the ocean?" asked another. ("Too many uncertainties," Chodas answered.)
"It worries me when you say that we are going to scramble to put together a 300-kiloton nuclear device," said a third.
She had a point, Barbee acknowledged. That so much of the initial fleet had failed shows how difficult it is to quickly build a successful spacecraft. "We probably should have been planning the backup option from the beginning," he said.
Breaking the fourth wall, some of the researchers complained that the scenario was starting to seem unrealistically worst-case.
"It is not likely at all," Chodas agreed, "and we're not saying it is likely. But we will learn by studying these what-ifs."
Things were about to get even more sensational. On the last day of the exercise, with 10 days to go before impact, Chodas announced that the disruption mission never got off the ground, and the fragment - now known to be 60 meters - was on a path to obliterate much of New York City.
The skyscrapers of central Manhattan would be vaporized. Fire tornadoes would rip through the region. The "mother of all gusts" would blow from Yonkers to Staten Island and New Jersey to Queens. Ten million people would need to escape the disaster zone -- more Americans than have ever been evacuated before.
A chorus of "oohs" rippled through the ballroom. "Tough," one person muttered.
Suddenly, the tone of the conversation shifted. They had left behind the comfortable certainty of physics. Now they had to plan for the unpredictability of people.
"We're very concerned," said Carol Lewis, who was playing the role of an "affected citizen." "We've actually called ourselves 'mad as hell.' And we have questions."
What would happen to people in nursing homes? What would happen to pets? Who would prevent looting of abandoned buildings? Would they be compensated for their losses? Where would the survivors go? And would they ever be allowed to return?
"Don't think we haven't been doing anything before we're at T minus 10 days," said Leviticus Lewis, a real-life operations manager for FEMA (and Carol Lewis's husband).
"I can tell you there already is a New York evacuation plan," he assured people. "I can't tell you what it is."
Then Joseph Nuth, a NASA asteroid researcher, stood up and stated the obvious: "It's our fault."
"How are we going to compensate those people, get them their lives back?" he asked. "This is an enormous liability that we took on when we screwed this up."
That led to a question of whether the U.N.'s "Convention on International Liability for Damage Caused by Space Objects" could apply to citizens suing their own government.
"The questions now are so deep and so far-reaching," said Victoria Andrews, NASA's deputy planetary defense officer.
And, she noted, they're not quite so esoteric as they might seem. Most natural disasters cannot be predicted as far in advance as an asteroid impact, but as a consequence of climate change, many are becoming more and more likely. Deadly wildfires and disastrous hurricanes that once were rare now seem inevitable - and communities have to prepare.
"This exercise gets us thinking about how to do that," Andrews said.