Aug 11 2014

Another Carrington Event Inevitable

On September 1, 1859, a massive solar flare struck the Earth, resulting in beautiful auroras but also inducing currents in telegraph wires causing them to spark and start fires. Hours earlier amateur astronomer Richard Carrington was observing the sun and noticed large sunspots giving off a brief bright flare. In 1859 the telegraph was about the only electric infrastructure we had. What if a Carrington-type event struck today?

Solar flares result from the complex magnetic fields of the sun. Gas in the sun is so hot the electrons are stripped from the hydrogen, resulting in a plasma. Since plasma is therefore made from ions, it carries an electric charge, and when electric charges move they generate a magnetic field. Magnetic fields further induce electric current.

Sometimes the magnetic fields near the surface of the sun interact in such a way that they give off an explosion of energy, called a solar flare. There is also something called a coronal mass ejection, in which a bubble of hot gas erupts from the sun’s corona in a fashion similar to a solar flare. CMEs and solar flares often occur together, but not always, and their causal relationship is not clear.

Solar flares and CMEs cause a pulse of energy in the solar winds. Charged particles racing from the sun, if they were aimed in our direction, would strike the Earth about 17 hours after the visible solar flare, so we would have some warning. The charged particles would interact with the Earth’s magnetic field, causing aurorae.

More importantly, the magnetic fields induced by the ions striking the earth would induce powerful currents in any conductive material. Any electronic device that was not shielded would be fried, its components damaged or even melted.

Think about what this would mean for our civilization. Every computer would be gone. Our satellites would be destroyed also. The power grid would go down. Communications would be gone.

Essentially we would be plunged into a pre-electrical age. I was recently asked what would happen to jets in the sky. Their electronics would be fried also, knocking them out of the air. Of course, there would be sufficient warning to ground all aircraft.

Modern cars with their electronics would also not function. We would all be trapped, without communication, transport, or electricity. The damage would be so extensive it could take many months to restore basic services. Damage would be in the trillions of dollars. It’s hard to imagine there wouldn’t be massive loss of life and suffering as a result.

What are the chances of a Carrington event hitting. One published estimate is that there is a 12% chance of a similar event occurring in a 10 year span (at least between 2012 and 2022, this varies with the solar cycle). We’re not talking about the possibility of an asteroid striking something in the next 100,000 years – there is very likely to be such as event in the next 100 years.

In fact, a CME of the size of the Carrington event occurred just in 2012. Luckily, it missed the Earth.

So what should we do about it? First we need to have a reliable early warning system. Sun activity monitors are necessary to guarantee we see the solar flare, and then immediately distribute the warning throughout the world.

But of course we need to have some way to respond to the warning. Grounding aircraft and stopping any activity that would be dangerous if equipment stopped functioning is a start. What we really need to do, however, is shield critical electronic infrastructure against strong external magnetic fields.

This is possible. The military also does this, as nuclear explosions release an electromagnetic pulse with similar effects. We would need to shield our satellites against solar flares, and they need the ability to turn away from the sun and quickly fold their solar panels (if they have them) and close up any exposed electronics. We would also need to shield the power grid and build in safety shut offs.

Consumer electronics are another matter. I’m not sure how much I would pay for a shielded computer, but I might buy a shielded hard drive to back up my data. Would you pay a few hundred dollars extra to have your car’s electronics shielded from an EMP or Carrington size solar flare? That would be an interesting upgrade.

The cost of protecting our electronic infrastructure from a Carrington event would be billions of dollars, but the cost of a Carrington event would be orders of magnitude greater.

I suspect that any attempt to predict what would happen should such an event occur is an underestimate. It’s difficult to trace the sequence of progressive collapse that would happen to civilization. Further, it’s actually likely that this might occur.

This is definitely an investment in prevention that seems worth the probable expense.

41 responses so far

41 thoughts on “Another Carrington Event Inevitable”

  1. scpecoraroIII says:

    How would a carrington event effect people with pacemakers?

  2. scpecoraroIII says:

    Quite differently, I’m sure. Horribly worded question. What I meant to ask- how would a carrington event affect people’s pacemakers?

  3. BillyJoe7 says:


    (Disclaimer: I’m no expert, so you can skip this if you wish)

    I thought your original question was more interesting.

    I can’t find the answer to your second question but, assuming the pacemaker would fail, the effect on the person would depend on the reason for their pacemaker and their state of health. Pacemakers are used for two basic reasons: complete heart block and slow heart beat. In both cases, unless their state of health had deteriorated significantly since they had their pacemaker inserted, they would survive the failure of their device just as they survived when they first developed the condition that led to them needing the device. However they might lose consciousness if their heart beat was too slow to adequatedly perfuse their brains. The heart has a back up pacemaker which fires at about 35 beats per minute, so even with complete heart block, the person is likely to survive – though he might lose consciousness if he had a poor level of fitness.

    (Personal note: When I ran marathons, my natural pulse rate dropped to about 35 beats per minute, and when I was physically and mentally relaxed, it would often drop below that and I could feel the back up pacemaker take over – it caused the heart valves to shut with a little more force because of the way the heart contracted. I treated this by training a little harder which would raise my heart beat just enough to keep the back up pacemaker silent.)

  4. carbonUnit says:

    Are you sure equating a Carrington event with an Electro-Magnetic Pulse (EMP) event as could occur by setting off a nuke above a region isn’t an exaggeration? My impression with these has been that the primary badness of these things is that they’d induce massive currents in power (and communication) lines, which would take them down by destroying the attached equipment. Large power transformers (like at substations) would be the main problem, because they are hard to replace, especially if we had to replace nearly ALL of them. Then you get into the problem of providing power to the factories which make them and power the the infrastructure and materials sources needed by those transformer factories. A real bootstrap problem, but not everything is destroyed. My impression is that electronic devices and circuitry on the scale of building wiring would not be damaged by such an even (assuming they aren’t taken out by power surges coupled in from the long lines of the grid.) I don’t have time to check now but will. Of course, if electric and electronic equipment survives the event, they will be rendered inert by the lack of power. Portable & backup generators will be good for a while, but then fuel will become unavailable because the refineries are offline. Rooftop solar power looks real good in this scenario because it should be less vulnerable to the induced currents of the long transmission lines of the grid. It could provide some power to keep things from going totally dark until the grid is restored.

    I guess some thought to encouraging stocking of more spare transformers and equipment needed to restore the grid should also occur, but the magnitude of the replacement effort would seem to make that infeasible. Seems like the main mitigation strategy is and always will be warning. We must have the solar observation capabilities in place to know that we are about to be hit. Power companies must be ready, on a dozen hours notice, to shut down the power grid and protect critical facilities by disconnection and shunting to ground. Key facilities (government, hospitals…) must be ready to go to local backup power, which is hardened enough to survive. Home owners can disconnect at their breakers and isolate electrical and electronic equipment from power and communication lines. Better to spend a day in the dark than years!

  5. carbonUnit says:

    Steve, where is your source for damage to “any equipment not shielded”? I really doubt consumer electronics not exposed to electrical surges from the grid or communication lines would be harmed. I doubt this is any risk to a pacemaker. I’m pretty sure smartphones, personal computers and auto electronics would be fine. Of course, without power and communications, they’d be of very limited value. (I wonder how hard it is to protect a cell tower or radio/tv transmitter? Is isolating the transmitter from the grid and antenna enough?) All I’ve ever seen in articles is concern about the long lines of the power grid taking it down. For example:

  6. jasontimothyjones says:

    the way I understand it, the problem may be due to Geomagnetically Induced Current, essentially massive magnetic vibrations all around the earth as the CME hits the earths geomagnetic field. Im only guessing what the results would be, but vibrating magnetic force around a 510,072,000 km² magnet probably stop you mid angry birds.

  7. carbonUnit says:

    Two articles from NASA:

    Good writeup and video on the 2012 event. Without the STEREO spacecraft, we would not be aware of the danger that just missed us. (STEREO-A took a direct hit and survived.)

    Shows interconnectedness of economy to power, likely blackout areas:

  8. Pugg Fuggly says:

    Steve, it’s rare that I take issue with your assertions, but your characterization of the effects of a Carrington-type event seem a bit out of proportion. As carbonUnit suggests, I don’t see much support for all electrical equipment being fried. Looking at for example, seems to indicate that the power grid is most at risk, but that it could eventually be restored (possibly within hours if enough preparation is made).

    Unfortunately, I’m no expert, so I’m relying on other people’s work, but there seems to be a marked difference between the more fearmongering-style posts and the more analytical ones, and I’m wondering if perhaps your post falls in the earlier category.

  9. Pugg Fuggly says:

    It appears carbonUnit beat me to it. Perhaps I need to chill out a bit when I have my “someone is wrong on the internet!” moments.

  10. BBBlue says:

    Wow, Dr. Novella sure has influence. Must have caused a run on tin foil hats as I can’t find a single one in local markets.

  11. The more powerful the event, the smaller the electronic device that would be affected. In most, it is probable that small electronic devices would either not be affected or would glitch but not be permanently damaged. Devices that are plugged in, however, could suffer from a surge, so there is still a huge risk. A warning would give people time to unplug everything.

    EMPs have the potential to be worse. It depends on the power, distance, and number of EMPs.

  12. There are several sources that I found that discuss this directly:

    I am still looking for a source, however, that offers more than an opinion, like an actual calculation or even something empirical.

  13. mumadadd says:

    “A warning would give people time to unplug everything.”

    Maybe I’m not a very nice person, because my primary concern when reading the OP was that my hi-fi, PS4, PC and smart TV aren’t insured. The potential chaos and loss of life from taking out the electrical and communications infrastructure didn’t cross my mind. As long as I’ve got time to get home and unplug everything, this won’t keep me up at night.

  14. Bruce says:


    I would say insurance companies would get out of paying for everyone’s appliances if that happened. They are sneaky like that… some kind of “act of god” clause or such.

  15. carbonUnit says:

    Even if all the electrically operated stuff is undamaged, a Carrington class event would still be a huge calamity, that point remains and it’s major. Yeah, planes would not fall from the sky, but there would be problems with radio communications during the event and without power, air traffic control would be gone. GPS satellites could be gone, not as much due to magnetic effects, but the particle storm. The planes would survive, but could not be flown safely without the ground infrastructure.

    Without power there would be no heating or cooling, no refrigeration, no water or sewage processing, no financial system. Food and fuel supplies would dry up. Communication would be very difficult. (No Internet!) We would be plunged back to pre-electrical life for years, even if our stuff survives intact.

  16. Pugg Fuggly says:

    I’m finding a lot of interesting articles on this topic, and it’s definitely more nuanced than I originally thought. I’m also glad I’m getting a chance to work out some of my skepitcal muscles, as I think I had a knee-jerk “it can’t be that bad” response at first and was working from an ideological bent.

    I also appreciate that Steve isn’t necessarily definitive about any of this and would like to see some hard data or calculations.

    That said, there does seem to be a consensus that a Carrington-size event could have pretty devastating effects on power grids. One lingering question is whether foreknowledge of the event could mitigate the worst of the damage.

  17. mindme says:

    Sadly if any government proposed a multi-billion dollar plan to harden our civilization, there would instantly spring up a denialist movement. I can imagine claims of solar astrofacists wanting to keep themselves smoking cuban cigars, driving european sports cars, and drinking fine wine by hoaxing the notion earth orbits a sun! Fah!

  18. I am digging deeper also. I am reading conflicting opinions, but all naked assertions without calculations.

    What I originally thought was that the fact that telegraph stations sparked indicates that electronic devices would be affected, but I’m not sure how good an analogy the two are, as the telegraphs involve miles of wire.

    A nuclear EMP is a good analogy for type of effect, but again I cannot find a comparison of magnitude.

    If anyone finds a good source, let me know.

  19. Pugg Fuggly says:

    Steve, I’ve come across a few interesting articles. One is at . Here I somewhat question the motivations. For example, a good chunk of the risk analysis is by John Kappenman, who, it turns out, owns a solar-storm risk consultancy. Perhaps there’s some incentive to exaggerate the potential for damage.

    Kappenman also plays prominently in this piece in Earth:

    This paper strikes me as a fairly well-balanced review of some of the literature:

  20. Pugg Fuggly says:

    That last one I linked has some pretty startling figures. They cite two different sources that say two-thirds of the US population would die within the first year of a Carrington-sized event. But then, elsewhere in the same article, most estimates of the financial impacts are in the $2 trillion range.

    Hard to know what to think.

  21. Interesting. One study from that last reference says that modern electronics are “one million times more vulnerable” to EMPs than electronics of the 1960’s, and getting more vulnerable, presumably because of delicate computer electronics.

  22. EvanHarper says:

    I agree with CarbonUnit that the depiction of cell phones and PCs being fried and airliners falling out of the sky sounds highly suspect. The current induced is proportional to the length of the conductor. The power grid and some other infrastructure like pipelines would be in great danger, but all you’d need to do for a piece of consumer electronics would be to unplug it.

  23. AmateurSkeptic says:

    Would this event last long enough to fry all of the electrical equipment on both sides of the planet or would half of the world still be OK?

    If it only would affect the side of the earth facing the sun, depending on how lucky or unlucky we were, we might all have to visit BJ7 or else he would have to Kon Tiki over to us.

  24. SimonW says:

    Having read some real science about the topic I’m now slightly less worried than I was before you mentioned it.

    The suggestion seems to be that if this was a major issue for personal electronics we’d see smaller scale examples as smaller more common event happen in places where the ground is conducive (sic) to it causing problems.

    Similarly the smaller issues we have seen, like Quebec 1989, were with particularly exceptional power grids lengths on particularly bad ground.

    Forewarned, I’d probably unplug sensitive electronics to be sure. Having seen local switchgear limp on after a lightning strike (they did service it a few days later causing a longer outage than the strike), I suspect we are better placed than in the 19th century.

  25. SquirrelElite says:

    Probably the closest we’ve come to simulating the large scale effects of such an event was the Starfish Prime test in 1962.

    Starfish Prime caused an electromagnetic pulse (EMP) which was far larger than expected, so much larger that it drove much of the instrumentation off scale, causing great difficulty in getting accurate measurements. The Starfish Prime electromagnetic pulse also made those effects known to the public by causing electrical damage in Hawaii, about 1,445 kilometres (898 mi) away from the detonation point, knocking out about 300 streetlights, setting off numerous burglar alarms and damaging a telephone company microwave link. The EMP damage to the microwave link shut down telephone calls from Kauai to the other Hawaiian islands.[5]

    The military’s critical communications infrastructure is radiation hardened, but the overwhelming majority of electronics used by the government are not, for simple cost reasons.

    Since the EMP is produced by the charged particles being accelerated (and their directions changed) by the Earth’s magnetic field, these effects will be spread over much more than half of the surface. Half the surface may be out of the line of sight, but only a small portion of the Van Allen belts will be.

  26. grabula says:

    I don’t know, I feel like my cell phone or computer going offline pales in comparison to what would happen to the power grids. I can’t seem to find anything about the ‘size’ of electronic infrastructure determining the effect, anyone else?

    Would these currents generate over wire not connected to power systems?

  27. RickK says:

    I too think that the effect of a Carrington event is different than nuclear EMP and that small electronics off the grid would be ok. If microelectronics were vulnerable to a flare, wouldn’t we have seen some such effect in 1989?

    The part that worries me is the potential effect on nuclear power plants that require uninterrupted power. It seems that many plants around the world would be seriously vulnerable to a major flare.

  28. Nitpicking says:

    (I work for an electric utility, but not as an engineer. Take comment with grain of salt.)

    All modern electric grids consist of a huge set of fuses with short stretches of wire connecting them. I’m not really exaggerating–here in New York, the longest run of the distribution system you’ll see without fusing is maybe a few hundred meters, unless you hit really rural areas. There is a fuse between the transformer that feeds your house (via the secondary low-voltage circuit) and the primary (high voltage) line, the primaries are fused every block or three, each primary terminates at a substation … by a fused connection, the transmission lines that feed the substations have both fuses and auto-grounding systems set up in the event of overvoltage, etc. I’m not saying there wouldn’t be problems, but they’d be more recoverable than this article implies.

  29. RickK says:

    Is the primary problem over-voltage? Or is the main danger that induced current of low period makes AC lines look like DC, saturates transformers and causes them to melt/fail? I think that is what happened in 1989.

  30. carbonUnit says:

    Very good article with a video that shows the footprint of the electrojets on the earth’s surface.

    Steve, I think the author, John Kappenman, would make a great guest for an SGU segment on the topic of geomagnetic storms. He seems to know his stuff, and could probably answer many of the questions raised here as to extent of damage, etc. Probably unnecessary, but maybe get Phil Plait in too, for the astronomical aspects.

  31. Nitpicking says:

    @RickK, these are by and large actual electrothermal fuses, or electromechanical ones that are triggered by magnetic flux. A current sufficient to blow a transformer would most certainly melt (actually vaporize) the fine-stranded wire in the fused cutouts long before it could damage more expensive equipment. That’s why we use fuses!

  32. RedMcWilliams says:

    Regarding the ~12% probability of another Carrington event. Does that mean a 12% chance of a solar eruption of the same magnitude, or an actual impact on the Earth? If the former, does the fact that we had a Carrington-type eruption in 2012 lower the statistical chance of actually getting hit in the next decade?

  33. BillyJoe7 says:


    There is a 12% chance per decade. That is not affected by what happened in 2012.
    Throwing heads on one coin flip, does not affect the chance of throwing heads on the next coin flip. It’s 50/50 every time (assuming fair coin and fair toss)

  34. Pugg Fuggly says:

    To address the other half of RedMcWilliams’s question, the 12 percent figure is the researchers’ estimate for another “Carrington event.” So not just any solar event of that magnitude, but one that directly affects the Earth.

    Although, frustratingly, this is never explicitly stated that I can tell, just heavily implied.

  35. BJ – You are correct only if each event is independent. This is likely not true with solar events. There is a solar cycle for one thing. Also, not sure if a huge solar flare in any way dissipates pent up energy. Is it like an earthquake releasing built up friction, or a volcano releasing pressure?

    Pugg – yes, my reading is that this is the probability of the Earth getting hit with a solar flare the size of the Carrington event or greater.

    We don’t really know how common such flares are overall, because we have not had that kind of solar monitoring. That’s also why I take the estimate with a grain of salt. But it’s probably not off by an order of magnitude.

  36. BillyJoe7 says:


    “You are correct only if each event is independent”

    I was assuming that the “12% per decade” assessment is correct.
    If it is correct, then it would be correct to say that during the ten years after 2012 – in fact for any ten year period – there would be a 12% chance of another carrington event.
    However if, for example, the solar cycles were so regular that a carrington event occurs exactly every 84 years, then it would be incorrect to say that there is a 12% chance per decade of such an event occurring. You would simply say that the event occurs every 84 years. If the cycles were less predictable, you might have to broarden this to every 70-98 years. Or if even less predictable, every 50-118 years.

  37. Stormbringer says:

    Nitpicking I was thinking the same lines with the protection of the equipment.
    There might be a delay with the fuses getting replaced especially at the local level but most equipment should safe.
    Satellites might be much more jeopardy, while they can be shut down to minimize the flux could there be enough of a push from the particles to destabilize the orbits.

  38. carbonUnit says:

    I think the problem is unexpected huge DC currents through ground. From the Spectrum article linked above:

    “The three-phase transformers used in this scheme are directly grounded through their neutrals. The ground is assumed to be an infinite sink that can absorb any brief, large fault current and at the same time keep the voltage across the grid from spiking and damaging equipment. In retrospect, that design is flawed: During a geomagnetic storm, the sink becomes a source of GICs, flowing into the grid from the ground.” *

    Kappenman goes on to describe a device which utilizes capacitors to handle the normal small AC flow between neutral and ground and a special vacuum tube based circuit to bypass the capacitor in fault situations. This should protect the transformer. Steve, the company that’s developing this is Advanced Fusion Systems in Newton CT, practically your backyard!

    * how does one do nice quotes and other things in these comments????

  39. BillyJoe7 says:


    “how does one do nice quotes and other things in these comments????”

    Type quote in angle brackets at the beginning, and /quote in angle brackets at the end

  40. Potato says:

    Nitpicking points out that the modern grid has been set up to help deal with this. A solar flare in 1989 knocked out a large part of Quebec’s power grid (where there were huge stretches of wires), and many learned from that lesson. In the 2003 Halloween solar flare very little went down.

    Still, a large enough event could possibly knock out the power grid for a few days-weeks, which would be chaos a-plenty. Smaller devices would likely be fine; the danger from planes would be the ionizing radiation dose to the passengers rather than the electronics melting and the planes falling out of the sky.

    The paper seems to estimate that a Carrington-type event would only be ~2-3 times more intense than the 1989 event, so I don’t think melting small devices is in the cards.

  41. nlflowers says:

    Everyone seems to be caught up on whether or not another Carrington event would damage consumer electronics. If it was large enough to wipe out the grid, nationwide, for only two weeks, it would still threaten societal collapse. No power means no money. You can’t get cash out of the ATM. You can’t swipe your card at the grocery store. You can’t buy anything. For many of us, we cannot work without power and could not thus earn a living. But, it also means you would have no indoor plumbing. You would not be able to get potable water to drink let alone bathe. People would be relieving themselves in the streets, which is the least threatening thing they would be doing in the streets. You can imagine the National Guard rolling out food and water to a single city for a few weeks, but not to every city and town. They certainly couldn’t keep the peace in every city for weeks. It would be total societal collapse in days, not weeks. The severity of the threat cannot be overestimated, which only underlines Dr. Novella’s original point: we should shield our critical infrastructure, now.

Leave a Reply