Lights Out In Québec: The 1989 Geomagnetic Storm

I found myself staring up at the sky on the night of March 13, 1989, with my girlfriend and her parents in the backyard of their house. The sky was on fire, almost literally. Red and pink sheets of plasma streamed out in a circle from directly overhead, with blue-white streaks like xenon flashes occasionally strobing across the sky. We could actually hear a sizzling, crackling sound around us. The four of us stood there, awestruck by the aurora borealis we were lucky enough to witness.

At the same time, lights were winking out a couple of hundred miles north in Québec province. The same solar storm that was mesmerizing me was causing fits for Hydro-Québec, the provincial power authority, tripping circuit breakers and wreaking havoc. This certainly wasn’t the first time the Sun threw a fit and broke systems on Earth, but it was pretty dramatic, and there are some lessons to be learned from it and other solar outbursts.

Flux Ropes

The event I witnessed actually started three days earlier as a coronal mass ejection, or CME. A CME is a stream of plasma and tangled lines of magnetic flux, launched from the corona of the sun. It’s not clear what causes CMEs, but they certainly seem to be associated with other solar phenomena like solar flares and sunspots, and generally occur more often during times of high solar activity. Current thinking is that a CME is a magnetic flux rope thrown off by the sun when stressed magnetic fields inside the sun interact, releasing tremendous amounts of stored magnetic energy. These ropes expand out from the sun like a growing donut, picking up billions of tons of plasma from the sun’s corona and propelling it ahead of itself at a million miles an hour.

In some cases, these expanding clouds of plasma are Earth-directed, like in March of 1989. After a few days of travel, the expanding donut reaches the Earth’s magnetosphere, which is the limit of influence of the Earth’s magnetic field. The leading edge of the CME forms a shock wave that flattens the magnetic field on the day-side of the globe. This allows charged particles to slip into the Earth’s atmosphere, causing the aurora I saw at around 42° North latitude – a little far south for regular display of the Northern Lights, but not so far that they were unheard of. The 1989 CME was so powerful that it caused aurorae clear down to Florida, and even Cuba witnessed the display.

The Path of Least Resistance

As amazing as the aurorae in the 1989 CME were, they weren’t what caused so many headaches for Hydro-Québec. The action there was caused on the backside of Earth, away from the Sun. While it was compressing the day-side magnetosphere, the CME shock wave was also stretching out the night-side magnetosphere into a long tail of magnetic flux. Just like the collapsing magnetic fields inside the sun that started the CME in the first place, eventually the magnetic lines of force in the tail reconnected, releasing terawatts of stored energy back toward the Earth.

This is where the trouble started for Hydro-Québec. All that electrical energy needed to go someplace, and as is always the case, it took the path of least resistance. Most of the province of Québec sits atop a massive insulating sheet of igneous rock called the Canadian Shield, and the thin layer of soil stretched over it was soon conducting massive amounts of current. The ground connections of Hydro-Québec’s transmission system of high-tension lines and transformers eventually started conducting some of these earth currents, and at 2:43 AM, protective circuit breakers tripped at the Chibougamau substation in central Québec. This caused an imbalance on a 750 kV transmission line, which tripped breakers 150 km away.

Damage to $10 million transformer in NJ as a result of 1989 storm. Source: J.G. Kappenamn

Within one minute, the cascading failures had tripped automatic systems all over Québec, shutting down 21 gigawatts of supply and plunging the province into darkness for over nine hours. The cascade of failures wasn’t limited to Canada; thanks to interconnects between the US and Canadian grids, over 200 grid faults occurred within the first few minutes in the US. Operators were able to shunt around issues and avoid any major blackouts, though.

To their credit, Hydro-Québec didn’t just chalk this up to an act of God and continue with business as usual once the dust settled. They learned the lessons the sun had to teach them that day, and put in systems to prevent a recurrence. The simple expedient of decreasing the sensitivity of the protective relays that first caused the problem has avoided a repeat in similar storms; there was another CME in August of 1989 that scrambled computers at the Toronto Bourse but failed to result in any major blackouts. Hydro-Québec also set up an early warning system to keep an eye on space weather, and changed policies with regard to switching operations and power levels when storms are predicted.

As memorable as the event was for those of us lucky enough to have seen the pyrotechnics, it was probably far more memorable for the six million Canadians who woke up to cold, dark houses, crippled traffic systems, and closed airports. Watching that display of nature’s quiet fury, and finding out the extent of the damage the next day, only served to underscore how puny we are, and how vulnerable our systems are no matter how hard we try to pretend otherwise.

56 thoughts on “Lights Out In Québec: The 1989 Geomagnetic Storm

  1. “Most of the province of Québec sits atop a massive insulating sheet of igneous rock called the Canadian Shield, and the thin layer of soil stretched over it was soon conducting massive amounts of current. ”

    Shame we can’t harness that.

    1. Yah, lets see, you’ve got molten iron at one plate, and mostly mineralised water at the other plate, 100km thick and 8 million kilometers square, so….. how many terafarads or how many exajoules can we jack that biatch with?

      Though yah, could find that when first numbskulls manage to drill through to magma, instant yottavolt discharge, because whole earth a supermegahypercap kablooie, vaporise everything in a 50 mile radius, huge freaking sprite into space…. gigaton nuke type effects…

      1. Well apparently ancient man “saw some shit”…
        And there’s ancient legends in many cultures about world cleansing thunderbolt and lightning events.

        And there be a lot of speculation about “instant fossilization” events, that are not naturally dying critters getting gradually buried in silt etc. That some high energy force or cataclysm did it.

        Now the Nubiru and planet X crowd wanna link this to interplanetary lightning or some shit, but…

        What if Earth has had supercap flashovers before, when the crust gets thin someplace, the charge is too, high, it’s triggered by solar storm, continental plates pull a little too far apart?? ….

          1. Yes, I assumed this was common knowledge to avoid posting 20 page essays starting from kindergarten level… which is why I mentioned the outliers where this has not apparently occurred as exceptional, there’s that small handful of “WTF”??? fossils that seem not to have been formed in the normal manner. There has been field observation of rapid, fossilization like, petrification of trees when high voltage transmission lines fell on them.

          2. It’s not that it ‘doesn’t normally happen’ it’s that electricity cannot fossilize anything.
            I was serious about wanting to hear about these other ‘WTF fossils’ but I suspect these are not fossils, just things that got coated in cement or dropped in cementicious puddles in coal mines.

    2. Oh, sure we can harness it! Just haven’t tried. A nice fat metal cable would be the path of least resistance (which is why Quebec lost power). I’m just not sure we can make a cable fat enough. It would require an array of ground rods at each end. For practical purposes we think of ground as the same potential everywhere, but it’s not!

      It was all clouds here… and the aurora was bright enough that the clouds glowed, but could not see it.

  2. I always thought that the magnetic fields induced a DC current in the power lines in CA, causing the transformers to saturate, overheat and blow up. I guess I’ll have to study some more to make my mental model fit better with reality.

    1. My understanding is the same… and I think this article may be wrong. It may be nice to start including citations, people!

      Also in Wikipedia–the non-conductive rock base of the area made the power lines the “path of least resistance”, and the induced currents caused the damage.

      In this article, it emphasizes that currents were induced in both the magnetosphere and the *power lines*.

      This NASA article mentions the induction of ground currents over much of North America for this event, but does not state explicitly that’s what happened in Quebec.

      Don’t get me wrong–ground currents are a real problem, and the cause of the satellite failures for this event. But the problem in Quebec seemed to be induced currents in the large loops created by their power lines. Local power generation (wind, solar, nuclear) would do a great deal to mitigate the damage caused by a solar storm that can happen at even milder storms than the 1989 one.

    2. It is an actual electron flow, guided in by the planet’s magnetic fields, causing the various gasses in the atmosphere to glow, flowing into the planet and through it and out the other magnetic pole, back up through the atmosphere again making it glow, and back out into space. It is literally a flow of electron current through the planet and it does love to follow the path of least resistance such as long distance cabling.

      The flow of electrons through your local earth is measurable at home by sensing the changes to Earth’s magnetic field it causes at your locality. A wind protected laser pointer, balanced and hanging free, with a magnet attached to make a compass, will make a magnetometer sensitive enough to show this effect by showing small changes in the North/South orientation that you will observe as field variations in where the beam strikes a wall 15 or 20 feet away. Note that someone opening the fridge door, or a car parking outside, will also be detected so you have to be well away from such. Magnetometers are monitored for this as continually in relation to the sun. and you will note the daily variation in the field due to the rotation of the earth.

      The source of the electrons is the sun itself when it throws off a coronal mass ejection (CME) caused by a solar flare or coronal hole.

      It is an actual flow of electrons through the earth… and this means ground is not the same voltage everywhere on the earth that you thought it is… it’s ground for just your one locality, hence useful for safety.

    3. Steve is right, except the transformers didn’t blow up, they overheated and tripped. Since then Hydro-Quebec has installed dc-blocking capacitors on the lines and made some circuit configuration changes to minimize the chances of this happening again.

  3. A couple points of clarification–

    “Magnetic lines of force” do not exist. They are simply a geometric construct to help us visualize a magnetic field, which does exist, and exists everywhere. People get so caught up in visualizing a magnetic field with lines, that it can lead to some false assumptions about what’s happening. They turn out to be useful in relating to what is happening here, but the visualization is not quite the same as the true mechanism, a lot of which we still don’t understand.

    Visualizing lines lets us imagine them being “stretched”, adding an incredible amount of potential energy, so that when they “reconnect” that energy is released rapidly, like shooting a rubber band. But that’s only imagery we can relate to–understanding the tension in a magnetic field is harder to visualize, so it’s useful.

    Also, magnetic reconnection happens at both the bow shock and the tail of the magnetosphere, and that’s what pulls material from the solar wind into the atmosphere. Normally, the magnetic domains of the Sun and the Earth would be distinct and no coupling would happen. The movement of charged particles at this boundary can create a coupling between the two fields (we’re still arguing about exactly how this happens, though MMS is looking at the idea that electron diffusion is where it happens), which we visualize as magnetic lines breaking and reconnecting so that there are Sun-Earth and Earth-Sun lines in addition to the Sun-Sun and Earth-Earth lines that would normally occur. But that’s all imagery–there is only one magnetic field in the entire universe, and the Sun, Earth, and plasma motion are affecting that field in a way that we visualize with these reconnected field lines. It’s just easier to think about that way; but the idea of magnetic and electric field lines can lead us to incorrect assumptions.

    Just remember, magnetic (and electric) lines of force do not exist. (Interesting corollary–electrical energy is not actually transmitted through the wires in your circuit, but rather in the fields around them. Otherwise capacitors couldn’t possibly work. The conductors of your circuit constrain those fields such that energy is transmitted from the voltage source to the load. That’s even true for your breadboard.)

    1. All models are false representations of reality in order to simplify complex systems. Atoms aren’t literally what they are in the Bohr model, for example, but that model does permit the basic physical processes to be represented and applications derived from that knowledge. The “business-end” of reality. Sure, the models are incomplete – but that will ALWAYS be the case, otherwise it wouldn’t be a model, but rather the actual system under study.

      1. That’s true, but in this case field lines are not a model, but rather a geometric representation abstracting the model. As such, they can lead to notions that are false. For example, people usually simplify the field-line concept by saying “the density of field lines tells you the strength.” But I can double the density of field lines just by drawing twice as many. One I hear a lot in my field is that charged particles are “attached to” field lines. They’re not–they’re really not. If you think of field lines as real, most plasma drifts don’t make any sense. It’s simply useful to visualize, as it maps out the trajectory of those particles, baring drifts and collisions.

        In any case, field lines are *not* a model, but a visualization of an extremely abstract concept. They tend to imply behavior that is not consistent with actual models (Maxwell’s equations) or observations. They are a very useful tool to visualize what is happening, but please be careful with them. Field lines don’t actually exist.

        1. Pffft, I suppose you’re going to tell me they are like isobars and I CLEARLY saw an isobar pass overhead when this wet weather blew in…. ;-)

          But fer serious, something came up in last year that was a brain blower in reassessing the actual physicality of field lines. New research/results. Wish I could remember what it was.

  4. What’s going to happen when our cars need GPS to work and the GPS satellites get wiped out because of a solar event? The more we become dependent on technology the more devastating the consequences. This is going to happen, it’s only a matter of time.

      1. It will be more of an issue for self-driving cars. As long as they’re designed to be able to be driven manually if needed, they should still work. Though whether people will know how to drive said cars… that’s a different problem.

        1. Self-driving cars still have an impressive suite of cameras an range finders, you’re not going to get in a wreck if the space weather isn’t accurately predicted. At worst you’ll just miss your turn.
          So far most solar storms have been pretty mild but given the military past of GPS they’re probably hardened against solar storms. So far just turning off the satellite and facing sensors away from the sun seems to work.

        2. I would like to have a self driving car, but I would not buy one without a manual option. Sometimes driving is fun also. I want to be able to decide myself, if I drive or get driven.

      1. A bicycle is a sports device, not a useful means of transportation except for a very small niche of distances which are a little too long to walk and not so far, that a car is necessary.

  5. Here in central NH we were lucky enough to see the light show (one of only a handful of times I’ve seen aurora) without losing our power (there were probably some short outages, don’t quite remember). It was quite a sight! It’s always amazed me just how powerful nature can be.

  6. “…served to underscore how puny we are, and how vulnerable our systems are no matter how hard we try to pretend otherwise.”

    What’s the total amount of solar energy intercepted by Earth, across the entire spectrum of radiation, heat, light, radio, the sun emits?

    Compare that to how much energy humans manipulate in heat, light, and radio, then compare that to something like a single category 5 hurricane like Katrina.

    How many Katrinas per second worth of solar energy hits Earth?

  7. It’s all fun and games until your eavestroughs (gutters) start exploding… (3:45)

    What seems to have happened here is a 37kV distribution line has arced over a transformer onto 13kV residential distribution, and the air/insulator gaps designed to cut/break 13kV aren’t doing a real good job at stopping Mr Honey Badger 37kV…

    Anyhoo, gets to a point where it seems that the 37kV starts inducing voltage in a metal eaves trough…

    So yah, we will have had a “real” E-M storm when your eavestroughs explode.

  8. Farther back in history there was the “Carrington Event” for those that want to read further and scare themselves. Telegraph wires melted and the drippings started buildings on fire. Some telegraph operators experiencing strange problems disconnected the batteries powering the telegraph, yet the telegraph continued to work. There were no electric distribution lines back in those days, but they are everywhere now. In places the Aurora were so bright that people thought it was morning and started preparing breakfast.

  9. 2017 “hyperflare” Z class event sometime in July, would probably make some interesting reading.
    Apparently we narrowly avoided a Carrington class event around August 2012, would have made the 1989 event look like a hiccup next to a tactical nuke.
    Next time we might not be so lucky, someone did the calculations that if the power was off longer than 2 days we would be looking at a mass extinction due to about 48 simultaneous total meltdowns at nuclear plants spanning Europe and America.
    A week on from that the high level waste tanks at Sellafield, La Hague and others would fail catastrophically assuming nothing was done, even the diesel generators would have run out of fuel.

    1. Can you cite “someone” here? Nuclear plants usually have a week to a month worth of fuel for their generators, and can go into cold shutdown in a week, necessitating no further cooling to remain safe.

      Even assuming your worst case scenario and we had 48 simultaneous Chernobyls, well, there are still people living in the exclusion zone around that plant and the wildlife in the area seems to be doing well enough, so I think mass extinction is unlikely.

      That’s not to say the failure of our infrastructure and the resulting civil unrest, famine, epidemics, and possible wars wouldn’t do us in…

  10. From Montreal we could see red (!) borealis aurora illuminating the sky. It was still the cold war at the time,; I tought US was under nuclear attack and we were getting EM peaks from there… Fortunately I had a small transistor radio, Ithen I could get the real story.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.