Wouldn’t such an event likewise knock out most automobiles, appliances, electrical systems, etc almost immediately leading to the collapse of urban food security and so on? Loss of internet seems like a minor inconvenience if you’re trapped in a city with a couple million hungry people and there’s no way to flee but foot or bicycle.
No, there is no fast (short wavelength) component to the variation of the geomagnetic field associated with coronal mass ejections caused by solar storms. Generally CME distortions of the Earth's magnetic field are large scale and slowish and only couple to and induce currents in very long conductors like the power grid or pipelines. Nothing human scale and disconnected from the grid will get damaged (like cars or phones). And most stuff connected to the grid will be fine, it's just the transformers infrastructure that will get damaged (not that that is just a "just" it's nasty with 2 month production wait times).
There are solar storm events which can induce faster timescale geomagnetic distances. Primarily the x-ray and EUV light from the solar flares themselves ionizing the upper atmosphere and the currents resulting from that. But to have a bright enough flare to do that in a damaging way would require an exceptionally large active region. So large you'd be seeing it with the naked eye on Earth before things went down.
Another fast timescale solar caused geomagnetic disturbance can happen from relativistic "sudden energetic particles" sometimes directed at Earth. But generally the amount of energy in these particles is minor compared to the component in the flare light (~2/3 on average) versus the CME if one happens (~1/3).
So, no. It wouldn't immediately knock out society. Especially with solar panels everywhere. But it would knock out the power grid for at least 2 months before new transformers could be built and installed. That might take down society.
Transformers themselves are pretty robust and it's believed that even high levels of GIC would not generate enough heat to cause widespread, permanent physical damage to modern equipment[1]
A more realistic scenario is that GIC would result in grid instability by causing protective relays to "trip" offline. Enough grid infrastructure tripping in succession could produce a cascading failure and widespread blackouts on a large grid. As seen in the 2003 Northeast blackout[2], it can take days to recover from such an event.
The more I have looked into this over the years, the less concern I have in these events being a major issue.
There is two things that puts my mind at ease.
1. Our electrical infrastructure, while flimsy in some ways is a lot more resilient than we take for granted. It is said the power grid fails in theory but works in practice. But given forward warning, it is possible to shut off or limit the energy flow prior to the event - it will mean voluntary blackouts for a few hours but it is better than a forced blackout for a few months.
2. The forward warning is actually really good now even being up to 2-3 days in advance, just look to any Aurora prediction site, we are all over tracking this stuff. For all the fear of a repeat Carrington event, we would know it is coming and deal with it.
We know when CME are coming but we lack any reliable ways to predict how geoeffective they will be. That's mostly a function of the magnetic handedness of the active region it comes from and if it will arrive at earth with their magnetic field in the right orientation (downwards out of the ecliptic plane, -Bz) to efficiently reconnect with the Earth's magnetic field.
There's been many a time when some large sequence of strong CME come and everyone expects intense geomagnetic activity but there's no -Bz so it just buffets things for a bit and causes auroral oval effects. Whereas other times some small lone barely CME will arrive with 7 hours of -20nT Bz and causes 45N aurora.
What I'm saying is that the people in charge of chosing to shut down parts of the grid are not ever going to get a confident assertion that damage will happen if they don't. There will be enough uncertainty that this kind of pre-emptive action days in advance will not happen.
At best we'll get a few hours of time to mitigate when the events arrive at L1 and the solar monitoring spacecraft there like ACE give actual warning.
> But given forward warning, it is possible to shut off or limit the energy flow prior to the event - it will mean voluntary blackouts for a few hours but it is better than a forced blackout for a few months.
The last few years have shown that leadership in certain areas has Cluster B personality traits, and will do the opposite of good advice just because it feels good to be contrarian.
The 2 month wait period for transformers both assumes that 1) there isn't a sudden huge spike in demand and 2) there isn't anything unusual getting in the way of building them, such as the grid being down literally everywhere.
Given how reliant we are on electricity and the internet I don't think it's actually a stretch to say it would be a society ending event.
