High Impact, Low Frequency?

The BP oil spill has got us all thinking about low frequency, high impact events. But some people have been thinking about these things for quite a while, including the team at NERC - the North American Electric Reliability Corporation. Who knew such an institution even existed? Their job apparently is to worry about unlikely events that would be really, really bad if they ever came about.

The latest NERC report includes a detailed section on geomagnetic storms:

The analysis indicates that the [geomagnetically induced current] in over 350 transformers will exceed levels where the transformer is at risk of irreparable damage...  Such large scale damage could lead to prolonged restoration and long-term chronic shortages of electricity supply capability to the impacted regions, arguably for multiple years.

In other words, a really big storm could roast enough components (70-80% of transformers in some states) to put the lights out for years. A version of this event happened in March 1989 and knocked out parts of the Canadian grid for most of a day. But since then, the grid has been developed in ways that make it more vulnerable than before: essentially, higher voltage transmission means geomagnetically induced currents can travel further and do more damage.

From a different angle, we could perhaps get a similar effect from a high altitude nuclear explosion. Here's a photo of the last time we had a go at one of those. This one was 400km in the air but the electromagnetic pulse was enough to blow out the streetlights in Hawaii, 1500km away.

(This is the 1962 Starfish Prime explosion.)