Hacker News new | more | comments | ask | show | jobs | submit login
1859 Carrington Event (hackaday.com)
101 points by titzer 29 days ago | hide | past | web | favorite | 66 comments

As I said over in the hackaday thread, many people both there are here are having trouble confusing the geomagnetic storms from flare ionizing light, geomagnetic storms from coronal mass ejections, and EMP from fast nuclear fission xrays interacting with free charges through compton scattering.

Geomagnetic storms from coronal mass ejections will not hurt small disconnected electronics or transistor junctions. The timescale of the rate of change in the Earth's magnetic field is not short. There are not high frequency components. Only long conductors like oil pipelines, telecommuncations wires, and power lines and the backbone transformers will be effected. The integrated circuits in your smart phone will not be.

Also, like I said in the hackaday thread, in most of the ways that matter (ie, geomagnetic effect at earth) the 1989 CME that blacked out part of Canada was significantly stronger than the Carrington event. We, as Earth, haven't really added that much more long distance conductors since then. I wouldn't worry.

That isn't to say there's no danger. Certainly trying to replace the backbone transformers when the electrical infrastructure is down would be very hard. Those things take months to make, ship, and install in a well operating society. A CME that knocks out the power grid everywhere would be devastating.

Good point about people confusing geomagnetic storms and flare ionizing light. To add to the last point about a CME that knocks out the power grid, I heard a presentation while I worked 2 summers ago at the frontier development lab, (https://frontierdevelopmentlab.org/fdl-2017/). I was on the solar-terrestrial interactions team.

One of the points in the presentation is that there are only one or two companies in the world that makes these transformers and there aren't a lot of these transformers just siting around waiting for a Carrington level event.

One of our goals for the summer was to use a LSTM neural network to forecast what would happen and possibly give warning when a CME slams into the earth. We could forecast a little bit into the future using geomag data. One of the subgoals of this work was to bring awareness and potentially help warn parties involved with the world's infrastructure.

A key point out of this is that there are dedicated researchers in both the public and private sectors that study sun. They are busily trying to understand how the sun works and help create systems that help warn satellite and power grid operators.

Two presentations from that year worth taking a look at might be of interest.

Solar-Storm Prediction: https://www.youtube.com/watch?v=2T_lHUwxTes

Solar-Terrestrial Interactions https://www.youtube.com/watch?v=n5QQnZLjSBE

Applications for this year's FDL summer program can be found at. https://frontierdevelopmentlab.org/apply/

> Only long conductors like oil pipelines, telecommuncations wires, and power lines and the backbone transformers will be effected

That is a huge relief. I was worried it was going to affect things that civilization has come to rely on heavily.

(Seriously though, your insight is very welcome and I thank you for it!)

>Some [telegraph] operators, in an attempt to spare their networks from further damage, disconnected their batteries from the lines, only to find that they could still send messages using only the current provided by the storm.

That's one of my favorite versions of "if life hands you lemons, make lemonade" stories.

Knowing what we know now about CME's, I wonder if it has ever been surmised that one option that might spare some expensive parts of the electric grid is a massive grid-wide disconnection prior to the CME arrival. This wouldn't save everything, but I think it would save a lot of blown megatransformers were they to be removed from transmission lines, reducing induced currents through them.

Everyone should read the Lloyd's analysis of a potential second event [1].

>The total U.S. population at risk of extended power outage from a Carrington-level storm is between 20-40 million, with durations of 16 days to 1-2 years. The duration of outages will depend largely on the availability of spare replacement transformers. If new transformers need to be ordered, the lead-time is likely to be a minimum of five months. The total economic cost for such a scenario is estimated at $0.6-2.6 trillion USD (see Appendix).

>While the probability of an extreme storm occurring is relatively low at any given time, it is almost inevitable that one will occur eventually. Historical auroral records suggest a return period of 50 years for Quebec-level storms and 150 years for very extreme storms, such as the Carrington Event that occurred 154 years ago.

Bear in mind, this report was authored six years ago.

[1] http://www.lloyds.com/~/media/lloyds/reports/emerging%20risk...

Having been through a multi-day power outage in the 2017 Tubbs Fire (where many transmission lines + tformers had to be replaced over weeks): I can honestly say that I don't think we can fathom what would happen during a multi-week or month power outage.

Imagine: No phones. No internet. No cell or data. No fresh food. No running water. Can't get gasoline to go somewhere else... The tension is literally palpable.

It was by far the most mind-altering experience I've had in terms of making me realize how thin the veil of society really is.

I grew up in the South, so I've been without power (and once water) because of hurricanes a few times. My experience was that it was startling at first. Without A/C and lights, a house no longer feels like a home and instead feels like you're camping out in a building. It's hard to convey how different that fact feels. The whole concept of "being inside" gets fuzzy.

But then, you get used to it. If you have water, you shower in the dark. You grill a lot. It's not a big deal... for a couple of days.

Where it gets really scary is that the longer an area goes without power, the worse it gets. Stockpiles and safeguards assume a certain max duration and as those get exhausted, new problems appear. You run out of perishable food, gas in your car, and charcoal to grill with.

Eventually, hospitals run out of diesel for backup generators. When that happens, people start dying very quickly. I imagine at some point the sewer system fails and then disease starts spreading.

The Dark Ages are always there waiting to return as soon as the infrastructure is down long enough.

Hopefully you're able to translate experience to preparedness, if you weren't already, and given California propensity for earthquakes (where the Tubbs fire occurred), that you have enough food and water stockpiled for at least a few days, if not longer. These days an earthquake preparedness kit might also consider including a small solar panel and an old cell-phone.

> I can honestly say that I don't think we can fathom what would happen during a multi-week or month power outage.

Pretty much anyone living in a rural or semi-rural area where freezing rain sometimes happens in the winter or hurricanes sometimes happen in the fall has experience with a week or longer power outage. It wouldn't be "business as usual" but a lot of people would be fine. The intersection between those people and people posting on HN just happens to be low.

A week-long outage is annoying. A month-long outage would be disastrous, even for those areas. There are many things you can do without power, but e.g. drugs require refrigeration, and we have people whose life depends on drugs (think insulin, for starters).

But also, urban areas would be harder hit precisely because they rely on their infrastructure more. And those areas are also the ones with more people.

Shit can happen. Get prepared or be a victim. I'm lazy and unorganized still I'm readyish for 2 weeks of outage. (More than 2wk seems unlikely enough / posses harder problems to solve I don't worry bout it)

I have 3-4 weeks of water, more than that in food, crappy canned food, but food. Weeks of Coffee and TP for my bunghole. Trauma and first aide gear. Gun. I really should get a portable generator. I'd rather just get a couple of Tesla powerwalls and solar...

This is an oddly glib response to a thread that started out by saying some people would face outages of 1-2 years.

If I was going to become a doomsday prepper, this is about the only thing that could do it. The article ends by talking about two years of power outages in the US. That would basically end society as we know it. No communication, no cars, no clean water, no shipments of food, it'd be chaos. And since it'd be world-wide, we couldn't even just hold on until the Europeans shipped us new parts for our factories.

Likely only half the world, given how it arrives -- which probably wouldn't save us given the interconnectedness of all the supply chains. Could half the world plausibly build a cell phone?

If you wanted to buy a city-sized transformer today you'd be put on a list for delivery sometime in the next few years. Trying to replace a continent's worth is a non-starter.

Probably yes, just not the latest model.

The idea of the fall of civilsation following such an event is ludicrous.

Given emergencies, governments can setup the necessarily laws and contingencies to put focus on rebuilding the electrical infrastructure. The army has generators, which can be used to bootstrap building more and setting up makeshift transformers. Building transformers is not black magic. Nationalising eletricity-related patents and some companies to put resource on what matters and gearing up the rebuilding can be voted on rapidly.

Have you run any numbers to back up your beliefs?

For those who enjoy war-gaming how to survive a massive EMP, Jonathan Hollerman's Survival Theory is a good read.


It would probably end a lot of programming jobs in an instant. Wouldn't it? The needs would shift to the reconstruction of the infrastructure.

I guess make sure you get your exercise in so you can handle a shovel, eh?

General response info at [0]. Tech description of impact of space weather on electric grids at [1]

I know from work (don't have links, sorry) that some distribution and transmission network operators include solar disruption in their incident response planning.

[0] https://www.ready.gov/space-weather

[1] https://fas.org/irp/agency/dod/jason/spaceweather.pdf

Can someone put into rough terms the probability of encountering a Carrington event within the next 10 years? E.g., you're more likely to be killed by a swarm of killer bees in Maine than be hit by a Carrington event.

I know it's being taken sufficiently seriously that understanding/handling geomagnetically induced currents (GIC) in power transformers has become a focus of R&D activity.

From https://en.wikipedia.org/wiki/Coronal_mass_ejection#Future_r...:

> According to a report published in 2012 by physicist Pete Riley of Predictive Science Inc., the chance of Earth being hit by a Carrington-class storm between 2012 and 2022 is 12%.

Is it still 12% then?

Depends on whether that figure is based on a point sample of expectancy for an event of that scale, or if it’s based on the expectancy due to the time elapsed since the last event.

As it’s an independent probability (i.e. the time elapsed since the last event does not affect the probability of reoccurrence), it’s likely the former, and therefore the same expected value - 1.2% in any given year.

You're looking at something like the geometric distribution.

The chance that the event does not happen is 88%. And since years are independent and identically distributed as you say, 88% = q^10 so q~= 98.7%; therefore the chances that it does happen are ~=1.3%

(As it turns out, your approximation works because as the time-grid becomes finer and finer, the geometric process becomes a gamma process. It does fail dramatically for 2-7 years.)

Seems odd the probabilities are actually did because of longer term solar cycles

> the previous solar flares had cleared the space between Earth and the Sun to make the plasma cloud travel faster

Is the matter in space such a slowdown on the plasma cloud that once cleared it allows that cloud to travel much faster?

Think of it more in terms of the shape of the pulse than of the absolute speed. (And take it for granted that dumbing things down a bit is, unfortunately, normal for the popular press. It's usually just as easy to give a more accurate description in less-technical language, but for whatever reason, it's rarely done.)

Ok, the "dumbed down" theory sounds more legitimate. I was wondering why a cloud of ejected plasma would travel substantially faster once you "clear the space" in front of it. Especially since much of the distance traveled is before hitting the magnetopause and Earth's magnetic field. Not much to slow it down out there, certainly not something a previous flare could clear up.

The matter ejected by a major flare like that would be traveling much faster than the normal solar wind of particles emitted by the sun. It's those particles, captured by the earth's magnetic field, that we see in the northern lights as glowing curtains or streams. Our magnetic field does concentrate the flow of particles, but still it gives an indication of the particle flux.

My understanding is that these particles can get channeled into the upper atmosphere and when they lose energy they ionize atmospheric gasses, and those ionized gasses (such as oxygen and nitrogen) are the actual glow. So you don't see the actual particles themselves.

You can go the next level and say it's the photons emitted that you actually see ... then go the next level and say it's a constructed view that we really "see" in our brains not the raw results of photon impacts on our retinas.

Not sure what the next level is? Something about societal artistic interpretation of aurorae impacting the perception of the visual image created, perturbing our sense of reality?

It's all in your mind, man. :)

Hmm... I would have assumed that two plasma clouds would've covered substantially different path, as Earth itself would've significantly moved along its orbit in between :-/

If you knew it was coming and had a few minutes to prepare, would putting your HDDs and SSDs in a Faraday cage protect them from damage during such an event?

Wouldn't disconneting from the grid/any other longer cables be enough?

I don’t think you need to be on the grid. A sufficiently large CME would induce currents in any unshielded electrical device. Kind of like an EMP bomb would break most cars by destroying the starter motor.

Yes, but those voltages/currents may be negligible. You can find some numbers online from the past events, like 4V/km of a power grid. That would not kill anything I have at home.

Interesting. Getting quantitative is a great way to put things in perspective. I would love to know that "unplug your stuff" is enough to save it from CME damage.

How extreme would a CME have to be for it to "fry" things that are not connected to any power source, like a disconnected, powered-off computer? Is it even possible?

It would be extreme enough that the earth would be sterilized of life due to radiation from the solar event, so your electronics won't be a big deal in comparison. The good news is that this won't ever happen.

Induction is going to be more or less proportional to the area enclosed by a conductor, which essentially means long transmission lines are the things to worry about. And this is length scales on the order of 10^5 or 10^6 meters, versus 10^-1 meters for your devices.

No. CME will only induce currents that are large enough to be noticed (much less damage anything) in very large conductors. You don't need to Faraday cage anything that you own, unless you own parts of the power grid.

Tangenital, but your comment reminded me of a Top Gear episode where a car acted as a Faraday cage. Perhaps putting any vulnerable electronics inside your car would be sufficient?

Or a biscuit tin or other metal box. Or a fridge. Or a freezer.

Cars are very leaky Faraday cages. If they weren't, you wouldn't be able to use your phone inside one. Because of the glass apertures, they can only provide protection from very, very long radio wavelengths - MF, LF, and VLF, all much longer than FM/VHF/UHF/microwave.

They'll also (probably) protect you from direct lightning hits, because the discharge passes through the conductive skin and won't usually take a detour through any occupants.

Flare events won't affect your phone anyway. Deliberate EMP might, but EMP would quite likely to be a precursor to nuclear war, so keeping your phone working might not be your absolute top priority.

If so, would a microwave have the appropriate shielding?

Freezer is probably better.

Thanks. I hope to never need this information, but I appreciate being able to think it through in advance.

I don't actually know this to be 100% true, I base this off the fact that my iphone (and android) still has plenty of reception in my microwave but is entirely cut off when in my freezer. This behavior could be inconsistent across microwaves and freezers and blocking radio communication may not be a suitable proxy for blocking a CME that would be many orders of magnitude more powerful.

These devices are way too small to be damaged by CME.

https://www.goodreads.com/book/show/851876.The_Sun_Kings Despite cover and book jacket hype IMO a terrific book on the history of human sun observation and the Carrington event

Backup to DVD. Get a wood stove.

What are you gonna use to restore the DVD backup after a modern Carrington-type event?

A laptop stored in a Faraday cage and a diesel/solar generator should do the trick. Although getting food and water will probably be a more immediate concern :)

If you store laptop in Faraday cage, why not include HDDs with the data too? Laptop also has HDD (and BIOS etc.), either it all works or all is fried

Hope your solar system doesn't have any control boards. ;)

Solar charge controllers are one of the least expensive parts in the system, so anyone serious about preparing for that kind of scenario should just have a spare. Whether the panels themselves would be affected I'm unsure of.

If you’ve ever stuck a dvd in the microwave, you’ll know this is a non-starter. Flux of the magnitude we’re talking about would cause arcing within the substrate, destroying it.

Back up to paper.

That's not true. Only long conductors are affected significantly by these events.

> Flux of the magnitude we’re talking about

How high exactly? At the surface of the earth? I mean, if it's microwave level energy, that would imply it was harmful to humans?

(Conversely, the non-harmfulness of the Carrington Event to life should be good evidence against the people worried about phone masts...)

To say such an event would be really bad if it happened today is an understatement.

how many nuclear reactors are hardened against this event?

Western reactors are designed to fail safe in the event of a complete failure like this, they'd just scram them.

Of course the same time as the electrical storm trashes the grid shutting down all the baseload nuke plants would just add to the chaos.

Bullets would explode. I wonder?

The powder in ammunition burns, not explodes. A necessary bit of pedantry because something has to ignite it, which is the primer. And a primer needs a sharp blow from a sharpish object (I’ve hammered primers practically flat without them going off).

In summary, ammunition will not go off.


Applications are open for YC Summer 2019

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