It's the year 2040, and the Big One -- a civilization-smashing solar storm of a scale not seen since the 19th century -- is on a collision course with Earth.

Far out in space, where geostationary satellites orbit, a half-dozen school-bus-size satellites crack open and start dumping barium, lithium or sodium. Within minutes, sunlight transforms this material into an ionized gas shield that slows the oncoming massive blob of plasma.

Down on our planet's surface, a would-be global catastrophe -- potentially knocking out entire electrical grids -- is reduced to a nighttime display for anyone who cares to look up and see the Northern Lights.

Even its name, StormWall, sounds like science fiction, but authorities on space weather say it could work, mitigating an event that happens, they estimate, once a century. The trio of scientists who conceived of it say an international coalition could build such a system with existing or soon-to-arrive technology.

Back-of-the-envelope math suggests it could cost tens of billions of dollars. Yet with all of the electronics on Earth that increasingly govern our lives, and ever more infrastructure being put into orbit, from internet-delivering satellites to AI-training data centers, spending that much could be a no-brainer, says StormWall co-designer Brian Walsh, an associate professor of engineering at Boston University.

"A 100-year [solar] storm could cause huge power outages over entire continents," says Walsh, adding that such an event could damage space-based data centers and missile-defense systems. "If the cost is less than sending people to the moon, and people do the math, it will make sense in the very near future," he adds.

The solar wind

For decades, scientists have studied the interaction between the Earth's protective magnetosphere -- generated by its iron core -- and the otherwise-lethal "coronal mass ejections" from our sun. (Ever wonder why Mars is a dead planet? No magnetic field.)

Charged particles of solar wind constantly catapult toward Earth, but most get deflected by the magnetosphere, which acts like a real-life Star Trek shield. A portion of those particles are channeled to Earth's magnetic poles, and when the solar wind is particularly active, they become visible as the aurora borealis and aurora australis. Activity tends to rise and fall in cycles that average 11 years.

"This is a process that takes place every day, several times a day," says Allison Jaynes, a professor and physicist at the University of Iowa. "When it happens in a big way, that's when we get the bad effects."

The last really disruptive solar event was in March 1989, which knocked out Quebec's power grid for nine hours. In 2012, a solar "superstorm" more powerful than any seen in at least 150 years narrowly missed Earth. And in 2024, near the peak of the sun's current cycle, solar storm Gannon forced New Zealand power-grid operators to activate their mitigation strategy. The same storm caused a GPS outage during a critical planting time in the Dakotas and northern Minnesota that cost farmers an estimated $1 billion.

The idea of StormWall is to help boost the magnetosphere's defenses when big, bad stuff heads toward us.

Think of it as an air bag, says Daniel Welling, a professor at the University of Michigan and co-author of the StormWall proposal. It would only be deployed if and when all other protective measures have been ruled insufficient. And like an air bag, deployment is a one-time thing.

"Once you use it, the entire steering column has to be replaced, and the insurance company will declare your car totaled," Welling jokes.

"In theory, it should work," says Ian Cohen, head of solar and space physics at the Johns Hopkins University Applied Physics Laboratory, who isn't affiliated with the research.

The challenge, he says, is that experts can't predict space weather the way they can predict weather down here on Earth. We don't have as many sensors in space, and the processes that trigger solar storms and make them devastatingly potent are much more complicated.

To make the StormWall system work, scientists would need to detect early indications of a particularly nasty coronal ejection and track its progress through space, both from the ground and with a deep-space observation satellite, says Walsh. From there, an international panel would have to agree on a course of action and then press the button to deploy the system, he adds.

A really big rocket

There are many engineering challenges that would have to be surmounted to make such a system operational, says Cohen. A significant expense would be the launch itself: The system would require 838,000 pounds of ionizable material -- lithium, barium or sodium -- to be lofted to an altitude of 22,000 miles above Earth's surface. That's 68 times as high as the orbit of SpaceX's Starlink satellites.

That altitude is critical because that's the point in space at which the ionized material could follow what the researchers call "natural highways" in space, giving about six hours of protection before drifting away.

Currently, putting that much material to geosynchronous orbit is nearly impossible, since it would require many launches. The StormWall payload mass is many times the mass of a typical geo satellite, and those already require the world's most powerful rockets.

SpaceX is developing Starship to send large amounts of mass to deeper-space destinations using refueling operations it hasn't tested yet. The company projects Starship will begin launching to lower-Earth orbits in the second half of this year, but it isn't clear when the company could get its rockets to the orbits required by StormWall.

Then there is China's Long March 9, which is projected to have roughly equivalent launch capabilities, but its maiden launch isn't expected until the early 2030s.

When I ask Welling how soon StormWall could be launched, he says that even in the best-case scenario, just the research would require at least five more years. "You don't want an air bag that explodes, and you don't want an air bag that under-inflates," he says.

Even if it costs $100 billion, that price tag is only a 10th of the amount tech companies are projected to spend on building out artificial-intelligence infrastructure next year alone. And without a space air bag, any sufficiently large solar storm has the potential to turn all those data centers into inoperable piles of steel and silicon.

Write to Christopher Mims at christopher.mims@wsj.com

(END) Dow Jones Newswires

July 10, 2026 08:00 ET (12:00 GMT)

Copyright (c) 2026 Dow Jones & Company, Inc.