By PETER BEHR of ClimateWire
Published: June 9, 2011
A massive
burst of solar wind that erupted from the sun Tuesday is expected to deliver only a "glancing blow" to the Earth's vulnerable magnetic field, NASA officials said yesterday. But it will preview what some experts call a potentially existential threat to the power grids of the United States and other nations, and the populations that depend on them.
Antti Pulkkinen, who leads NASA's "Solar Shield" satellite-based detection system at the Goddard Space Flight Center, said the cloud of ionized particles from Tuesday's violent "coronal mass ejection" will largely miss Earth, giving some North American residents a glimpse of the aurora borealis, or northern lights, this weekend. "It will not be a major event [for] the power grid," he said.
However, NASA spacecraft detected a much larger eruption last weekend on the backside of the sun headed away from Earth, generating a much faster-moving cloud.
"If this event was on a collision course with the U.S., we would have had a major space weather event," Pulkkinen said. "In this regard, we got lucky."
The next peak cycle of sunspot activity is predicted for 2012-2014, bringing with it a greater risk of large geomagnetic storms that can generate powerful rogue currents in transmission lines, potentially damaging or destroying the large transformers that manage power flow over high-voltage networks.
"Geomagnetically-induced currents on system infrastructure have the potential to result in widespread tripping of key transmission lines and irreversible physical damage to large transformers," a 2009 report (pdf) by the North American Electric Reliability Corp. (NERC) and the Energy Department says.
Agreement on the seriousness of the threat, but not the solution
In the worst-case scenario, the stockpile of spare transformers would fall far short of replacement needs. Urban centers across the continent would be without power for many months or even years, until new transformers could be manufactured and delivered from Asia. The transformers are not made in the United States.
"If the solar storm of 1921, which has been termed a one-in-100-year event, were to occur today, well over 300 extra-high-voltage transformers could be damaged or destroyed, thereby interrupting power to 130 million people for a period of years," Joseph McClelland, director of the Office of Electric Reliability at the Federal Energy Regulatory Commission, said at a May 31 House Energy subcommittee hearing on the issue.
"The U.S. society and economy are so critically dependent upon the availability of electricity that a significant collapse of the grid precipitated by a major natural or man-made EMP [electro-magnetic pulse] event could result in catastrophic civilian casualties," Rep. Trent Franks (R-Ariz.) said at the same hearing.
The U.S. grid currently relies for its defense on warnings from NASA that would alert U.S. utilities to take actions to protect their systems. Tuesday's storm did not require a response, NERC said.
But the alerts have the effect of advice and there are no mandatory, enforceable procedures or emergency actions, NERC officials say. No comprehensive plan exists to retrofit the transmission grid with protective devices, although the Electric Power Research Institute, the industry's primary research and development organization, is developing a range of technical responses.
The threat is a top priority for FERC and NERC, their officials say, but the two organizations have sparred over the reach of new federal authority that could be created to upgrade the grid's protective equipment and defensive plans.
While the House last year passed the "GRID Act," addressing vulnerabilities of the bulk power sector to natural threats and cyber attacks, action in the Senate is tied up by conflicting bids for jurisdiction by five different committees.
Learning from a 1989 solar storm
Pulkkinen said that NASA's satellite and computer systems provide a vital early warning capability. The solar eruptions can be detected several days before the space weather strikes the Earth, and more detailed threat analysis can be generated 24 hours before an event begins, he said.
The warnings weren't available during the last major solar storm that began on March 13, 1989.
That massive impulse blacked out the entire power grid in Quebec in 92 seconds, giving operators "no time to even assess what was happening to the power system, let alone provide any meaningful human intervention," said John Kappenman, author of a January 2010 report (pdf) by Metatech Corp. prepared for Oak Ridge National Laboratory.
More than 80 percent of Quebec's system was back up within 11 hours, Kappenman said. But he and other experts say the challenge to the U.S. grid is more serious for several reasons.
The Hydro-Québec power system draws almost entirely on hydroelectric generation, which permits a relatively simple and rapid restoration, he said. In contrast, the U.S. grid relies predominantly on steam electric generation, making restoration much more difficult, and a recovery as fast as Quebec's "is highly unlikely."
High-voltage lines act as antennae, attracting geomagnetic disturbances from the sun, and the larger and longer the lines, the greater the effect. The total length of U.S. high-voltage lines has increased nearly tenfold since the 1960s, Kappenman reported. The average length of the highest-capacity U.S. lines currently is four times greater than the smaller lines used more than a generation ago.
"Today's sprawling high-voltage power grids are more susceptible to space weather impacts than ever before," Kappenman said.
Designing defenses against space weather
Utilities have spent several billion dollars installing equipment to protect their transmission networks from lightning strikes, Kappenman added, but installations to defend against space weather are lagging. In part, that's because the research on the threat and the best countermeasures has not been completed. The transmission networks are so interconnected and interdependent that one company's investment in protective equipment could be nullified if a neighboring utility did nothing.
But some available defenses are also not being used, Kappenman said. He cited devices called "series capacitors" that can block the flow of geomagnetic currents on transmission lines. In the western section of the U.S. grid, many are in place. In the entire eastern grid, however, only two lines have this protection.
In the Quebec event, grid devices called compensators that were essential to counteract the rogue currents' effects tripped off to protect themselves. That "pulled the legs out" from under the province's grid, precipitating a rapid collapse of voltage, Kappenman said. This illustrates the need for complex modeling of the grid's vulnerabilities, he added.
The 1989 disturbance, which was felt by power systems deep into the United States and damaged a transformer at a nuclear plant, is far from the maximum. A report in the spring issue of the EPRI Journal notes that the 1921 storm was 10 times stronger (but it hit a far less developed and exposed grid network). The strongest solar storm on record hit in 1859, battering the U.S. telegraph system in places, EPRI said. That event may have been 50 percent more powerful than the 1921 storm, it reported.
Recent storms "do not represent the most severe storm events that are plausible," Kappenman said.
Richard Lordan, a senior technical executive at EPRI's Palo Alto, Calif., office, said his organization is researching a range of defensive strategies. The amount of long-distance power transfers can be reduced as much as possible, and generators close to population centers can be brought online.
"As a final resort, you can consider removing transformers from service," he said -- deliberately blacking out parts of the grid. EPRI has developed sensors that could detect high geomagnetic currents in transformers that would alert operators that the equipment had to be taken offline.
"If the storm is severe enough, operators have ways of islanding the system," he added, just as rolling blackouts have been used in extreme power emergencies like the Texas winter storm during this year's Super Bowl week. "It's preferable to take the system out and protect the equipment. It's a last-resort step.
"There are other [protective] devices under development," he said. "But the industry wants to be sure these devices don't impact the reliability of the system."
Critics call security process 'too slow'
The Department of Homeland Security has funded EPRI's design of a modular replacement transformer that is now being tested. It can be adapted to the range of substation configurations around the grid, and shipped in three pieces by truck to wherever it was needed. It will be installed for field testing in 2012.
But there are a host of unanswered policy questions before the replacement transformers could provide effective backup, he said, beginning with how they would be paid for.
"What would be the appropriate deployment strategy? How many are needed? Who owns them? Who maintains them? And who determines when an event is severe enough to warrant deployment?" he said. "These conversations are going on."
FERC and NERC agree that mandatory authority is needed to deal with solar weather emergencies that are days or hours away. NERC President Gerald Cauley says his organization "should be given authority under FERC oversight to address grid security vulnerabilities by enforceable means other than standards."
But they aren't in accord over whether the federal government can step in and direct a transformer replacement program. Cauley told the House hearing last month that NERC's current "bottoms-up" process for developing grid security standards that begins with its power company stakeholders is the right approach in this case.
The provisions in the House-passed GRID Act spelling out FERC's authority to order a transformer replacement program are not needed, Cauley said. "FERC already has the authority to order us to address these topics today," he said.
McClelland responded, "the commission has said on numerous occasions that when it comes to national security, the process -- the standards development process is too slow. It's too open, and it's too unpredictable."
