Hacker News new | past | comments | ask | show | jobs | submit login
Solar Storm of 1859 (wikipedia.org)
148 points by oli5679 on May 4, 2020 | hide | past | favorite | 84 comments

Please read my 2016 testimony to the Federal Energy Regulatory Commission on the threat of solar storms:


Market these ideas to defense firms that could act as a supplier. A billion dollar program has to be enticing to someone.

I like how you think

OMG. Thanks for this info. I doubt us in the UK are being any smarter. I might question the probabilites as they've been updated since, see my comment https://news.ycombinator.com/item?id=23067785 but that gives zero excuse to do nothing.

This should be the top-voted comment, if anyone would care to read it. I's short and pointed.

Edit: this reminds me of Feynman's presentation on the Challenger disaster, only beforehand.

Is there anything an average person can do to help? An organization I can donate to, or something I can bring up to my representatives?

That is a terrifying report...

Thanks for assembling and sharing that report.

Did FERC do anything with it, or was it ignored?

FERC directed the North American Electric Reliability Corporation (NERC) to set a reliability standard for solar storms. Under that standard, not a single high voltage transformer in the U.S. electric grid is required under the standard to have hardware protection. Instead, paper studies exempt utilities from taking tangible action. As of May 2020, only one device to block harmful currents from solar storms has been installed in all of the U.S. electric grid. Fairly stated, FERC ignored my testimony and that of other experts.

Thank you for sharing this and I hope someone in government takes action to protect our grid. Based on your findings have you made any personal preparations for an event like this? I lean towards preparedness but know I can be a little over reacting when hearing about these possibilities.

This 19th century miner from rural area is more literate than most of today's people

>> I was gold-digging at Rokewood, about four miles from Rokewood township (Victoria). Myself and two mates looking out of the tent saw a great reflection in the southern heavens at about 7 o'clock p.m., and in about half an hour, a scene of almost unspeakable beauty presented itself, lights of every imaginable color were issuing from the southern heavens, one color fading away only to give place to another if possible more beautiful than the last, the streams mounting to the zenith, but always becoming a rich purple when reaching there, and always curling round, leaving a clear strip of sky, which may be described as four fingers held at arm's length. The northern side from the zenith was also illuminated with beautiful colors, always curling round at the zenith, but were considered to be merely a reproduction of the southern display, as all colors south and north always corresponded. It was a sight never to be forgotten, and was considered at the time to be the greatest aurora recorded... The rationalist and pantheist saw nature in her most exquisite robes, recognising, the divine immanence, immutable law, cause, and effect. The superstitious and the fanatical had dire forebodings, and thought it a foreshadowing of Armageddon and final dissolution

> miner from rural area is more literate

Or... The Perth newspaper editor took the liberty of rewriting his story - in an early example of sensational click-bait.

Or maybe GPT2 strikes again!

It should be noted that many modern systems are quite resistant to solar flares. Not by design, but mostly by chance.

Specifically, solar flares induce currents in long untwisted pair cables. If the cables have high impedance loads on the ends, high voltages will result, frying equipment.

The telegraph was an example of a high impedance long wire.

In modern equipment, long length high impedance lines are rare. Telephones are one, at around 900 ohms, although nearly all cables are twisted so barely affected. Power lines have an impedance below 1 ohm. Most Comms goes via fiber. Cable TV networks have a grounded shield so unaffected. Small household devices are small, and therefore don't have long lines to get big currents.

I'm sure a solar flare would break some stuff, but the vast majority of devices are more resistant than people expect.

The most vulnerable thing in today's world is probably actually disused phone lines, which are currently disconnected from anything else, so infinite DC impedance, yet very long, and could cause sparks and fires since they typically have paper insulation which is only good for 1000v or so.

I really do not buy this. Everything I've heard and read contradicts at least the "we'll be ok" view. Some very long untwisted thick wires are power cables.

On what evidential basis are you making this claim?


"An enormous solar storm could short out telecom satellites, radio communications, and power grids, leading to trillions of dollars in damages, experts say"

I asked an electrical engineer how well we're prepared for a large solar storm like the carrington event. She said 2 words: "we're not". What do you know that an expert working in this area doesn't?

More: https://science.nasa.gov/science-news/science-at-nasa/2014/2...


A similar storm [to the carrington event] could have a catastrophic effect. According to a study by the National Academy of Sciences, the total economic impact could exceed $2 trillion or 20 times greater than the costs of a Hurricane Katrina. Multi-ton transformers damaged by such a storm might take years to repair.

"In my view the July 2012 storm was in all respects at least as strong as the 1859 Carrington event," says Baker. "The only difference is, it missed."


"An enormous solar storm could short out telecom satellites, radio communications, and power grids, leading to trillions of dollars in damages, experts say"

(1) Telecom sats are no longer essentially to much of anything. The vast majority of communcations happen over fiber lines, buried or laying deep under the ocean. GPS satalites may be affected, but the current crop are hardened and expected to survive.

(2) Radio communications certainly may be disrupted during the storm. Our cellphones might be offline for a week, so too local radio stations: a temporary inconveniance rather than end-of-days.

(3) Power grids may well be distrupted. Some transformers may blow. There are a handfull in the US that are of vital importance, but they could be replaced. A worst case senerio might see some large areas without GRID power for a period of weeks. But the power plants will still be there. Some sort of diminished local service would be restored.

Who will feel the effects: The military. Militaries still rely on telecomunications via satalite. They still rely on radio for long range communications.

We don't have all that many spare transformers sitting around, waiting to be used when one fails. And they aren't all that fast to build. If we lose a number of transformers, it may be a while before some of them are replaced.

You're handwaving a lot of serious effects of a powerful solar storm. The biggest issue by far is the power grid and depending on severity it can go from "we had a small spike that was easily absorbed by the network" to "the cables melted and set fire to everything around them".

The worst case scenario might be that your power grid is down and nobody can intervene because every intervention site is on fire. Big cities would be particularly affected since they don't do well for long without power.

> Our cellphones might be offline for a week, so too local radio stations: a temporary inconvenience rather than end-of-days.

All wireless communications, all maritime communication, being affected is much more than an inconvenience. The literal end of society and civilization? No. Potentially hundreds of thousands dead, billions of dollars in damages? Yes.

Your (2) ignores the dependency of civilian air traffic on working radios, AND GPS.

Your (3) ignores the TIME it takes to manufacture AND deliver/deploy grid-level/substation grade transformers.

There aren't many spares around, and it's not like you could buy them at the Home Depot/Lowes, etc.

Actually, if you're so inclined, you could gather that from various press releases of the involved parties, which i happened to come across because of idly watching various youtube videos of said heavy transports, which linked back to those. It's often a very complicated and slow multimodal operation, by ship, rail, and street, involving large detours to avoid bridges and other obstacles.

So experts are wrong and you're right. OK.

The reason I'm here is that there are often better informed experts here than in the cited articles. Maybe that's the case here or maybe not. Or maybe you're the best expert on the subject. I don't know, but I don't mind hearing the opposing view.

It's not like (I hope!) any government's public policy is being driven by HN comment threads.

> So experts are wrong and you're right. OK.

Yes, that position was already clear from their comment.

There are sometimes better informed experts here I suppose. But you didn't ask, you just made nonverified declarations which a little reading of the links posted here would have told you were wrong.

Didn't come across that you actually wanted to know the facts, but to assert that everything was fine, just fine.

That kind of thinking is what got us into the covid crisis we're slogging through at fantastic cost right now.

Of course that may not have been your intent but if not, what are you doing.

That isn’t actually true. Induced current will occur in nearly any conductor, and the biggest risk all over the world is in the loss of high voltage transformers.

This is absolutely, 100% a big problem. While these FOIA’d FEMA docs [0] are just one source, searching for “solar flare blackout” will net you all the sources you need. It’s the same as a nuclear EMP.

[0] (PDF): http://www.governmentattic.org/24docs/UnpubFEMAgeomagRpts_20...

The North American Electric Reliability Corporation apparently doesn't agree with that assessment:

"NERC recognizes that other studies have indicated a severe GMD event would result in the failure of a large number of EHV transformers. The work of the GMD Task Force documented in this report does not support this result for reasons detailed in Chapter 5 (Power Transformers), and Chapter 8 (Power System Analysis). Instead, voltage instability is the far more likely result of a severe GMD storm, although older transformers of a certain design and transformers near the end of operation life could experience damage."


Read this transcript [0] of remarks—given to the Federal Energy Regulatory Commission—at a 2016 hearing on US power transmission reliability. These remarks directly contradict the relevance and, indeed, the reliability of the NERC study quoted.

[0] (pdf): https://www.resilientsocieties.org/uploads/5/4/0/0/54008795/...

Well, I wasn't worried till I saw that. It reminds me of Trump saying there was no reason to worry about Coronavirus.

The stuff about blowing the transformers is exactly what I've heard too. If there are any EE who'd care to comment, I'd be curious to know if the level of power that starts flowing through HV cables could damage or destroy them. As I've not heard this I assume not (one bit of good news) but I'd like to know.

The problem is not directly with the wires, its the addition of the quasi-DC induced currents in an AC transformer. Even the addition of a small DC current can lead to unwanted harmonics or in extreme cases spot heating (this is what you think of when people say 'knocking out a transformer').

If you want a very in-depth review: https://doi.org/10.1002/2016SW001501

Well over my head but printed it out and will read anyway. Thank you.

Electrical cables are untwisted, but they are also designed to handle a lot of current so the overwhelming majority would be fine. Unfortunately, ultra long distance power transmission cables are the exception.

The long term damage from that is significantly less than assumed. Yes, making ultra efficient transformers designed to last decades is extremely time consuming, but building something less efficient in under a week is also possible. There would likely be various electrical issues until they where replaced, but that’s vastly better than societal collapse.

I do not believe the international supply chain is sufficient to replace any at-risk high-voltage transformers that are destroyed by an electromagnetic pulse. There is a considerable body of research around this topic.

This is not a question of 1:1 replacement via the existing supply chain. This is using a 1920’s solution that gets us 80% of normal not the much better but more difficult 2020’s solution to the same problems.

Consider, a modern ship’s propeller is an marvel of modern engineering that would be extremely difficult to replace with something nearly as good. But, welding some plates to the drive shaft is vastly better than nothing in an emergency. Similarly, if the choice is a poor solution or no electrical grid, having some power makes a huge difference.

Look at this: https://en.wikipedia.org/wiki/Electrical_grid#/media/File:A_...

The purple lines there stand for the highest voltage level of the grid which usually runs at 380.000 to 400.000 Volts(with a few higher exceptions). These and their transformers are the most vulnerable to what we are talking about.

What do you hope to 'McGuyvering' there with maybe 110.000 Volts when the whole structure and anything connected to that is designed, specified, and built for at least three times the voltage?

This is what we are concerned with: https://en.wikipedia.org/wiki/Electrical_grid#/media/File:Un... Not all of those lines are equally at risk, but for the sake of argument assume everything over 200KV is gone.

400kV is not directly stepped down to 33kV there are generally intermediaries at 132kV and or 66kV. So if you can send 132kV or even 66kV much of the equipment can be used.

Getting enough power to operate refrigeration and stop lights is a big deal. If you can only send 20% of normal power across those lines then that’s what you’re stuck with. In places such as Hawaii, local generation is close enough that your it’s no big deal, in others you may end up with rolling blackouts or charging 1+$/kWh.

Seems like the trillions bailout money ought to be spent on hardening these systems to prevent a solar flares or EMP type of events.

I think you have it the wrong way around. Geomagnetically induced currents (GICs) are more prevalent in high-voltage wires because they have low DC resistance. The longer a line, the larger the induced voltage. High-voltage lines can extend up to 1,000s of km, as they are most efficient. Then its basically V=IR. So low-resistance wires and long distances = larger GICs.

The danger is not from the cables burning out (extreme GIC events can lead to a couple of hundred Amps, which shouldn't be a problem for the cables), but the addition of the DC current in an operational transformer will push it into core saturation leading to nasty effects. Twisted/untwisted cables don't come into it.

Source: postdoc who models Carrington-scale events, and did my PhD on GICs in a regional power network.

It is clear to me that geomagthrowaway has some real technical expertise on solar storms and GIC.

if we had 16h to prepare, could high voltage transformers be saved by linemen racing to disconnect them and wrap in tinfoil or some other way?

Disconnecting? Sure, reasonable idea. If the current doesn't get into the transformer, then it can't destroy the transformer.

But wrapping tinfoil? The kind of energies we're talking about would vaporize the tinfoil without breaking a sweat.

Disclaimer: I am not an EE.

Not an expert but re. your lovely image of wrapping them in foil, what I understand is that the long power lines pick up the huge surges and it's these that damage the transformers. So AFAIUI just disconnecting the transformers would protect them. But I repeat my disclaimer.

In fact, manual disconnects are included in the mitigation protocols proposed by researchers when providing guidance to the U.S. government on this topic.

I got the impression that the ground potential would change on such an event, that it's not only caused by induction on wires. Isn't it the case?

Differences in ground potential means that any cable that is grounded on more than one place now has a huge amount of current trying to run through it, on a large potential.

What about common mode induction - when the inducted current flows in the same direction in all strands of wire (regardless of twisting) with earth return path? This is what AFAIK causes the problems with power transformers and can happen on long metal pipes too.

Common mode induction is an issue in any twisted pair system.

As long as the DC impedance to ground anywhere is <500 ohms, I wouldn't imagine enough voltage to build up to break down even a tiny signal transformer. In power systems, the connection between neutral and earth does that.

Long ethernet cables might be a concern, since the common mode DC impedance at the termination is normally infinite... Analogue audio equipment also frequently has infinite common mode impedance to prevent ground loops.

I really hope you take to heart the sources and links that people will respond with, and not just switch to defensive mode. This is a dangerous topic, and important to get right.

Thanks for this. I keep seeing FUD like such an event would wipe hard drives (which seems like a reach, but I'm not an electrical engineer).

What sorts of consumer electronics (if any) might be affected by a flare?

All of them, if there’s no power grid.

That doesn't answer my question AND isn't even accurate.

(Ex: I could charge my tablet with an Anker solar charger I have and as long as the electronics hold up have access to the videos/music indefinitely)

We’re talking no power grid for months or years. Most portable electronics would be technically fine, but unable to operate for any significant period of time because of the lack of power.

Your best bet might be to stash a solar panel some place shielded, and then use that to charge the batteries after the CME has passed.

As i understand it people mistake two different phenomena for the same. The geomagnetic storm induced by coronal mass ejection of our Sun is dangerous for the power grid because it could fry its transformers, because basically the long distance transmission lines work like an antenna, amplifying that. But it is a slow and weak thing, so anything disconnected at the time should be ok.

The fear about disconnected electronics becoming fried comes from the Nuclear Electromagnetic Pulse(NEMP). Which is a fast thing, inducing large spikes in a short amount of time, thereby frying transistors on-chip, if unshielded, even when disconnected.

If all of food/water/gas/electricity/fuel distribution is badly interrupted, there's the bigger picture.

You can grow or shoot food. I'm Appalachian, I know more than just how to code :)

> What sorts of consumer electronics (if any) might be affected by a flare?

* Old things (ie. pre 1990) - ie. before it was common to put protection diodes on every CMOS input.

* Analogue things designed to be very sensitive, particularly at longer wavelengths. FM/AM radio, ECG machines, microphone inputs...

* More modern things if a power cycle can't reset state.

This isn't true. Your electronics might be fried by a flare if they were in space, but the geomagnetic field changes due to a flare will not induce currents in any conductor shorter than several tens of kilometers. I think you are thinking of nuclear EMPs which could affect electronics?

Thanks for explaining this in such simple terms. I am grateful. I wish I learned enough about electricity long time ago; and somehow I never get the resolution to do it today - despite, having an computer science background and curiosity for all things science, I suspect it shouldn't be particularly hard.

Is this true for train overhead lines?

The government response to the Royal Academy of Engineering report on space weather shows remarkable complacency.

The RAE:

Since the last peak of the solar cycle, the Great Britain transmission system has developed to become more meshed and more heavily loaded. It now has a greater dependence on reactive compensation equipment such as static variable compensators and mechanically switched capacitors for ensuring robust voltage control. Thus there is increased probability of severe geomagnetic storms affecting transmission equipment critical to robust operation of the system. The greatest effects of GICs are normally experienced at the periphery of the transmission systems, as in Figure 9.

The UK government response: Some UK systems are more resilient and robust to space weather than counterparts abroad:  the GB power grid network is highly meshed and has a great deal of built in redundancy. This potentially makes it less susceptible to space weather effects than power grids in some other countries.

Just another existential threat to our civilisation that we all know about but assume someone else somewhere else is dealing with.

A toxic combination of normalcy bias, the availability heuristic, and hyperbolic discounting?

As seen in Y2K, spend money to fix a problem.

If you are successful, the problem doesn't occur, the population listen to those who say "it wasn't really a problem in the first place", and mainstream populists adopt that position to win votes (regardless of their own thoughts). You don't get to play that card again

If you aren't successful, the problem occurs anyway and you get accused of not caring

Without Democracy those in charge wouldn't be concerned by trying to win the population over

Without the ability for anyone to reach millions of people overnight people wouldn't be swayed en-mass by contrarians

Often you're damned if you do, damned if you don't. We can see the same thing unfolding in the UK and US over covid19.

I hope we don't look at this in a few years wondering what should have been done...

Is there any hope of avoiding huge problems due to solar flares happening in the near/medium-term? A silver lining from the Coronavirus pandemic is that people may think about these rare events more.

Depending on the magnitude of the storm, having enough advance notice would be enough. On a larger scale, turning off satellites, dialing down the amount of electricity distributed over the powerlines to allow headroom for that excess caused by the storm, having a lot of spares on standby, etc. On a smaller scale, individual readiness where possible (backup power generation, supplies, alternate telecommunication means, etc.). But there's a point beyond which there's no avoiding catastrophic issues - electricity and telecommunications down for extended periods of time, plus the myriad of side effects.

As far as I know we have no feasible way today to completely mitigate the effects of a storm similar in scale to the 1859 one. A strong enough solar storm can induce currents able to simply melt the transmission lines (especially low resistance ones - ultra high voltage, 1000+kV) possibly causing fires along the way, destroy transformers, and more. There will always be a storm that surpasses the GIC tolerance on most or all transformers. Pipelines and any other type of cable of significant length would also be seriously affected.

Despite the low cost, the USA is unprepared: https://news.ycombinator.com/item?id=23061684

Ditto the UK. See my other comment here.

On a personal level, maps batteries, generator and solar panels could come in handy.

I think it's possible to prepare the grid too.

For those of you who want to dig deep into the truth of solar storm (GMD) vulnerability of the North American electric grid, take a few minutes to read the first part of our comment to FERC during the docket process for NERC standard setting. And then say what you think on the thread. https://www.resilientsocieties.org/uploads/5/4/0/0/54008795/...

Can you imagine the sort of additional havoc that this would cause if it happened now? We've been somewhat mercifully spared of natural disasters but C19 and something like this would be horrific.

Llyod’s of London released a report in May 2013 on the potential effects of a significant Coronal Mass Ejection (CME). The projected costs were 0.6 - 2.6 Trillion USD.


As other comments have said, the biggest pain would be the availability of replacement small transformers.

If people are interested, I'll give my opinion on this paper: https://www.sciencedaily.com/releases/2019/03/190312103717.h... Anyone want me to comment?

Are there any models that estimate the chance of this re-occuring? What do they look like?

2020 seems like the sort of year that this would make sense happening, at least from a narrative point of view.

Indeed it seems like the person who calculated the 2012 made a typing error. At least if we get through this year, there will be some epic movies coming next summer...

Edit: ok then https://twitter.com/RobinWigg/status/1070319949510520832

This is probably why we're talking about it now.

Cue for an article about a census of asteroids that cross the Earth's orbit tomorrow.

Imagine being able to see the tubes they plow through the polar lights, maybe their plasma contrails would glow like a rainbow, or maybe you could see iridiscent vortices!

What a show!

Yes. My comment https://news.ycombinator.com/item?id=23067613 was partly based on a paper on the storm of 2012. Their estimate of probablility of a carrington-type thingy was 12% per decade (from memory but I believe it is correct).

Edit: this probability appears to have been amended recently. From https://www.sciencedaily.com/releases/2019/03/190312103717.h...


According to this new research, the probability of a similar solar storm occurring in the following decade ranges from 0.46% to 1.88%, far less than the percentage estimated before. "In 2012, the results reported in scientific literature estimated the probability to be around 12%, ten times more than our more pessimistic estimation,"


So much better, but one of the paper's authors says "A probability close to 2% which is what we have calculated for a highly intense storm should not be looked over if we take into account the consequences of such an event"

From what I’ve heard, they seem to occur about every a hundred and fifty years, so one could happen any time. However, we don’t know if it’s an event that becomes more likely over time or if it’s completely random chance.

To be fair we've been more rigorously observing the Sun for ~250 years (24 solar cycles). We've had 1 major storm, mentioned in this article. We can assume weak solar cycles like the one we're in now and the one expected to come after will not bring any surprises but in the end we cannot predict them more than we can predict earthquakes.

There have been more solar flares like the Carrington Event, they just didn't hit Earth. The last one was in 2012.


Ah, I read the "chance of it reoccurring" as "another massive storm hitting Earth". I'm sure they happen all the time, most just point in a different direction. We probably can't even detect the ones heading in the opposite direction from Earth, erupting from the other side of the Sun.

Actually we can detect them, thanks to the Stereo satellites. (And also indirectly through helioseismology.)

*Satellite... We've been down to one since 2014. Still does good sentry duty though.

Based on what we see of other stars, we’re overdue.

Perhaps the biggest long-term issue of this (once the immediate damage is repaired, which would perhaps take years) is the loss of decades of human knowledge - everything on hard drives, tapes, flash memory, and just about anything that isn’t an optical disc - with no real warning.

Well, small electronics such as hard drives and mobile phones are not in direct danger from these sorts of events; the damage is caused by induced current on miles-long cables. Our flash memories and tape backups will not be harmed unless they are burned up in power surges from unprotected power outlets.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact