The debate over implementing comprehensive electric-competition policies throughout the U.S. economy still rages to this day. Pat Wood III, as the federal regulator, had to fight many tough,...
Geomagnetic storms and the limits of human experience.
maximum—expected at the start of 2013. (See Figure 1.) Yet the biggest storms don’t always coincide with the solar max, as Murtagh noted:
“The biggest storm on record since 1930 occurred four years after the solar maximum. We see that quite often.
“We [also] talk about those ‘Halloween’ storms … That was October 2003, also four years after the sunspot maximum.”
As the record shows, the U.S. experienced large solar storms in 1859 and 1921, plus smaller and more recent storms in 1989, 2000, and 2003. The first two events were the most severe, but they occurred before the construction of today’s extensively interconnected interstate grid, and during less-active-than-usual sunspot cycles. Nevertheless they saw telegraph service disrupted across large areas of North America and Europe. The 1989 storm brought down the Hydro-Quebec transmission system—an asynchronous grid that isn’t part of North America’s Eastern Interconnection, with which it has DC ties. That storm provided the first convincing evidence of how GMD and GIC could impose widespread effects on a geographically extensive transmission network. (See, NERC GMD Interim Report, pp. 3-6, see also, Comments of Foundation for Resilient Societies, FERC Docket No. AD12-13, filed April 26, 2012.)
Great uncertainty remains, however. As DHS Pugh noted, “we don’t know whether the limited space weather data collected over the past few centuries accurately represents the full range of possibilities.” In other words, while the 1859 Carrington Event was the most powerful storm in recorded human history, we have no basis for listing it as one-in-a-hundred.
Ben McConnell, the retired research scientist formerly at Oak Ridge, told FERC that according to some estimates, a 1,000-year storm would likely run 46 percent larger than in 1859. A 10,000-year storm would be twice as large.
“What magnitude of severity of geomagnetic storm,” McConnell mused, “could be a civilization-killer?”
Modes of Failure
Overall, the NERC report’s weakest link might not be its tone-deafness, but miscalculation about the possible causes of transformer failure. In short, the evidence heard at the FERC conference suggested that a number of low-intensity, long-duration GMD events might carry significant risk of damage for EHV transformers—and not just the rare high-intensity storm.
Consultant Kappenman raised the question:
“We have really not gotten much in the way of answers … from the transformer experts in this area. We know that … for a brief period of time you can stand a 30-percent over-excitation. But when you get out to longer duration, they talk about [how] you cannot tolerate even 10-percent over-excitation, even when the transformer is unloaded.”
As Kappenman added, “we’ve had some incidents of episodic failures of transformers, all associated with long duration, low-level storm activity.
“We think that is what happened in the case of South Africa.”
And FERC’s McClelland seemed to sign on that that possibility.
“One of the interesting aspects of the Oak Ridge report,” McClelland noted, “I guess it was you, John, that found that there was a sort of cluster failure associated with GMD events.
“Now if you’ve been in the industry any period of time, you know that those types of