Forgot your password?
typodupeerror
Power United States

Where a Massive Solar Storm Could Take Down the US Power Grid for Millions (cnn.com) 51

A new study published Friday in AGU Advances "depicts the most advanced picture yet of how a rare but massive solar storm could affect some US industries and electric systems more than others," reports CNN.

"During a 250-year solar storm, the East Coast would have widespread potential for grid failure." Researchers modeled what a Carrington Event — a 1-in-150-year geomagnetic storm or even rarer — could mean for the modern day. They estimate the US economy would lose about $1.5 billion to $2 billion per day from direct and indirect costs, as millions of people face power outages. The findings show the most vulnerable areas are the Northeast and the Northern Plains, especially locations at higher latitudes.

No one knows when the next big storm will hit. While the most severe geomagnetic storms are labeled as once-in-a-century or rarer, these events don't follow a fixed schedule. "That timing only averages out on very long timescales," said Anna Kelbert, a geophysicist at the Harvard-Smithsonian Center for Astrophysics, who was not involved in the study but previously studied solar storm effects on power grids. "The event can occur at any time, as soon as a week from now, or centuries later...." [Shawn Dahl, service coordinator at the Space Weather Prediction Center operated by the National Oceanic and Atmospheric Administration in Boulder, Colorado], said the sun is now coming down from a period of heightened activity, which is also when some of the biggest storms to hit Earth have occurred in the past. "We still need a lot of work to be ready...."

"We haven't really seen events like this size in the modern era, where we've had all of this high technology and we've had the power grid of this size," said Ed Oughton, researcher at George Mason University and lead author of the new study... Because the US grid's actual structure is proprietary for security reasons, the team used engineering models to create various network and transformer setups including the 10,464 substations and 16,256 transmission lines found across the country... The grid structure also contributes to higher risk in some areas. For instance, the eastern US is linked together in a single grid interconnection. A voltage issue in Maine, if not isolated early, could cascade through the grid to Washington, DC. The West works similarly, whereas the Texas interconnection operates mostly within the state...

Power is restored over time depending on any alternative electrical paths that could bypass the substation or availabilities of spare transformers, for instance. "This nationwide analysis shows that we are currently ill-prepared to face a major magnetic storm, and that the societal impacts would be catastrophic," Kelbert said.

"The study estimates power disruptions for 5.1 million people and 135,000 businesses," according to the article, and "The outages could last from hours to days to weeks depending on the damage and resources available to get back online, said Oughton."

Where a Massive Solar Storm Could Take Down the US Power Grid for Millions

Comments Filter:
  • dummies (Score:1, Funny)

    by Anonymous Coward
    SOLAR POWER IS A SCAM

    you put a giant semiconductor array on your roof and then act surprised when THE SUN notices

    solar flare hits the grid

    panel starts glowing

    inverter begins speaking ancient Japanese

    your smart meter grows ears

    and then BOOM

    CAT EVENT

    do you understand the physics here???

    photons electrons solar flare electromagnetic catastrophe cat ears

    this is why I generate my electricity by screaming into a hand-crank dynamo

    WAKE UP

    THE SUN IS TRYING TO TURN YOUR ROOF INTO AN
  • But one happens in the next 2.5 years, it will be Joe Biden's fault. ;-)

  • Not limited to the US though. See Japan with Fukushima and the ignorance of the designers how high tsunamis could get, despite historical markers being available. Be incompetent and sooner or later you have a catastrophe on your hands.

    • by jabuzz ( 182671 )

      That is a bad example. The height of the wall was sufficient to stop the site being flooded. What they didn't figure in was the land dropping over one metre in height, so all of a sudden the wall was no longer high enough. The bigger issue was having the backup generators in the basement, which is a dumb idea were ever you happen to be when their failure in a disaster situation has such dire consequences.

  • by CustomBuild ( 2891601 ) on Sunday September 06, 2026 @08:32PM (#66326510)

    The part that bothers me here isn't that a Carrington-scale storm could knock out the grid. We've known that for a long time. It's that we seem to be counting on "restoration" as though replacing a damaged transformer is roughly equivalent to replacing a server. It isn't. Some of these things are enormous, expensive, custom-built pieces of equipment with long lead times. If enough of them are damaged at once, there may not be a lot of spare capacity sitting around waiting to be plugged in.

    And the interconnection issue is worth thinking about. We have spent decades making the grid more interconnected because that's generally efficient and useful. But a system that is wonderfully efficient under normal conditions isn't necessarily resilient when the disturbance is spread across an entire region. If Maine and Washington DC are ultimately part of the same electrical machine, "just isolate the problem" sounds a lot easier on paper than it might be at 2 AM during a once-in-a-century event.

    What makes this particularly annoying is that this isn't some exotic science-fiction threat. We can watch the Sun. We know roughly what kind of event we're worried about. We know which parts of the grid are vulnerable. And we know that having some additional transformers, better monitoring, better procedures and a real emergency plan would cost money, but probably far less money than discovering after the fact that we didn't have any of those things.

    I'm not expecting the lights to go out tomorrow. But "it might happen next week or 200 years from now" is actually a pretty good argument for preparing now. You don't buy a fire extinguisher because you know the date of the fire.

    • by stabiesoft ( 733417 ) on Sunday September 06, 2026 @09:09PM (#66326534) Homepage
      To emphasize your point, because of AI, Lead times are 2.5 years. https://www.industrialsage.com... [industrialsage.com] We could be in the dark for a very long time. I'd hope they would redirect the AI transformers to people, but who am I kidding. AI is more important.
      • by jabuzz ( 182671 )

        Not really, the government would in such a scenario tell the AI firms they just got bumped to the back of the queue, because power for the general population is more important. The alternative is the population riots and that will not work out well for those in charge of the AI firms if they kick up a fuss. Especially in the US were the mob will be armed with guns.

        • You mean the government endorsing the flock ai cameras? That government? I could see a tiananmen square with even worse consequences. Just give the assignment to ICE and the tank would roll over everyone, not just one.
      • You raise a good point. Just about the time AI starts to take over the world and exterminate humans...solar storm to the rescue! The storm will take out AI and leave only us meat-sack brains behind.

    • Forget the sun, today it's Putin's drones destroying the power grid.
      Already happening in Germany.

      • Forget the sun, today it's Putin's drones destroying the power grid.
        Already happening in Germany.

        NATO did it in Serbia ~30 years ago. IIRC they used clouds of graphite ribbons tossed from the air into electrical substations.

    • by thegarbz ( 1787294 ) on Monday September 07, 2026 @04:30AM (#66326826)

      It's that we seem to be counting on "restoration" as though replacing a damaged transformer is roughly equivalent to replacing a server. It isn't.

      The reason we're counting on restoration is because the design of these systems protect critical equipment like transformers. We calculate fault levels, speed of rise levels and design systems to interrupt and protect critical infrastructure (at the expense of the customer's connected). You're not going to be replacing large transformers. You're going to be replacing small scale control electronics in the substation. You're going to be replacing countless single action fuses. You'll be dealing with a few switchboard fires, hopefully isolated to a single feeders at a time so that a spare one at the end of the switchboard could be put back in operation. Then the big issue will be a black start.

      The majority of the damage from the Carrington event was for telegraph equipment precisely because we already do a significant amount of electrical infrastructure protection, and we've gotten significantly better at this over the decades.

    • The only way to prevent it from taking down the grid is to put massive specialised active shunts in every link to block the very low frequency currents. Not going to happen, so they "just" turn it off, repair everything which broke from turning it off and do a massive blackstart. At least monitoring of transformer temperature is a lot better now than during the Carrington event.

    • by tlhIngan ( 30335 )

      The thing is, it's not hard to protect the grid. The Sun is monitored and if something like this happens, it can take anywhere from hours to days to hit. Which means we have plenty of time to start protecting things. They don't come out of the blue, we have plenty of warning.

      And the best way we have to protect the grid is to turn it off. This can be a bit inconvenient, but if you have say, 24 hours notice that the power will be shut off for a few hours, it's a lot easier to prepare for. Transformers are eas

    • The problem will not be the event, but that someone with Trumps intelligence (or less) and an urge for civilizational suicide (like Trump) will be in charge of preparation. Does not matter what the crisis will be. The next pandemic will not have a vaccine, by choice. No amount of climate change will alter policy. The continued decline in public education wonâ(TM)t help much either.
  • by sacrilicious ( 316896 ) <qbgfynfu.opt@recursor.net> on Sunday September 06, 2026 @08:39PM (#66326516) Homepage

    No one knows when the next big storm will hit. While the most severe geomagnetic storms are labeled as once-in-a-century or rarer, these events don't follow a fixed schedule.

    And now thanks to Intergalactic Warming, these storms will be bigger and even more frequent.

    • And now thanks to Intergalactic Warming, these storms will be bigger and even more frequent.

      That is awesome. You win the internet today sir!

  • That's a 1 in 167 year event. And counting...
  • by sarren1901 ( 5415506 ) on Sunday September 06, 2026 @10:47PM (#66326632)

    I feel like Carrington gets all the attention because he was English and it was the last major observable storm. Thing is, a Miyake event would be 10x worse and we've yet to deal with one in modern times.

    https://www.esa-spaceweather.n... [esa-spaceweather.net]

    Has a really good read up on both events, their differences and the potential damage. Maybe Miyake is ignored because we literally can't do fuck all about it because spending the money to actually have backup equipment is "inefficient".

    As it is, we are lucky we dodged a massive coronal mass ejection in 2012 by sheer luck. Had it happened a week earlier, Earth would of taken a direct hit.

    • by evanh ( 627108 )

      We just have to be prepared to turn everything off. Early warning should give enough time to shutdown power stations. Maybe not fission plants but that's just one more indictment on that tech.

      It's certainly not an impossible task. The most important action being to trip all automated disconnects in the distribution network. The bulk of which would be at substations.

  • by PPH ( 736903 ) on Sunday September 06, 2026 @11:30PM (#66326668)

    ... a decent system of solar observatories. So we will know if such a storm is coming. And that the massive outage is not just a result of my local power companies typical operations.

    • The way it hits depends on so many factors, there is no good model for how it affects the different links (no standard measuring of geomagnetic induced currents to help build it) and most transformers have no specs for how sensitive they are to the induced currents.

      Unless the ejection it's truly apocalyptic I think they'll just wait and see (for something the size which triggered the Canadian blackout for instance).

      • by PPH ( 736903 ) on Monday September 07, 2026 @01:19PM (#66327280)

        and most transformers have no specs for how sensitive they are to the induced currents.

        Not entirely true. It's not difficult to model the saturation effects of a d.c. current superimposed on the incoming lines and its subsequent effects on transformers. All that needs to be done is to sense it and operate protective breaker trips to isolate them. As far as predicting the generated circulating currents produced by a storm, who cares? You sense the currents and de-energize the equipment before it is damaged.

        Yes, you get large outages. But if you save the hardware, the system can be restored relatively quickly after the storm.

  • Here's a fun fact... They were having a drill... and a real one hit!

    https://www.rnz.co.nz/news/sci... [rnz.co.nz]

  • This can has been kicked down the road since Richard Carrington and Richard Hodgson's event back in 1859.
  • " A voltage issue in Maine, if not isolated early, could cascade through the grid to Washington, DC." Why is that even possible? What do the engineers say?

    Of course, I know the answer, ROI isn't everything, it's the only thing, who cares about the actual service you're providing.

  • "...the sun is now coming down from a period of heightened activity, which is also when some of the biggest storms to hit Earth have occurred in the past."

    As opposed to the big events that happened in the future? Has somebody been time traveling and not telling us?

  • Happened in 1859, took down a lot of the TELEGRAPH lines, some even caught on fire. That was before electricity was around! Just imagine what that would do to the power grid, smartphones, internet and pretty much everything with a chip! The Amish would be about the only ones that could survive! ;)

Today's scientific question is: What in the world is electricity? And where does it go after it leaves the toaster? -- Dave Barry, "What is Electricity?"

Working...