A future Space Weather catastrophe : a disturbing possibility

Shortly after midnight on September 2, 1859, campers in the Rocky Mountains were awakened by an "auroral light, so bright that one could easily read common print. Some of the party insisted that it was daylight and began the preparation of breakfast", according to the Rocky Mountain News. Magnetic observatories world-wide recorded disturbances in Earth's field so extreme that magnetometer traces were driven off scale, and telegraph networks experienced major disruptions and outages. The electricity which attended this beautiful phenomenon took possession of the magnetic wires throughout the country, the Philadelphia Evening Bulletin reported, and there were numerous side displays in the telegraph offices where fantastical and unreadable messages came through the instruments, and where the atmospheric fireworks assumed shape and substance in brilliant sparks. In several locations, operators disconnected their systems from the batteries and sent messages using only the current induced by the aurora. In Havana, Cuba, the sky that night appeared "stained with blood and in a state of general conflagration" and auroras were observed as far south as Hawaii and northern Venezuela (Figure 1). A British amateur astronomer, Richard Carrington, observed an outburst of "two patches of intensely bright and white light" from a large and complex group of sunspots the the center of the Sun's disk the previous day, and so the solar storm of 1859 has been dubbed "the Carrington event". It remains the most severe solar storm to affect the Earth in recorded history.


Figure 1. Locations of reported auroral observations during the first ~1.5 hours of the September 2, 1859, magnetic storm (orange dots). Courtesy J.L. Green, NASA.

What would happen if another solar outburst with the magnitude of the Carrington event were to hit Earth today? With society so much more dependent on electricity, the effects could be tremendously expensive, causing serious disruption to the economies of nations in the northernmost and southernmost portions of the globe. An example of this vulnerability occurred during the March 13, 1989 geomagnetic "Superstorm" when a severe geomagnetic storm triggered strong direct currents in the long wires of the Canadian electric power grid. The grid was not designed to handle this, and the result was a power outage that knocked out power to the entire province of Quebec, Canada--six million people--for nine hours. The 1989 event very nearly brought down the electrical power grid over a large portion of the U.S., as well, and we were very lucky that the storm was not stronger. The 1859 Carrington event was three times more powerful than the 1989 "Superstorm", and would likely cause the failure of a huge portion of the U.S. power grid were it to hit today.

Top geomagnetic storm events of recorded history
The intensity of a geomagnetic storm can be measured by counting the number of solar charged particles that enter the Earth's magnetic field near the Equator. This number is called the Disturbance storm time, or Dst. Reliable Dst measurements go back to the 1950s. Bruce Tsurutani of NASA used magnetic field measurements taken on the ground in Bombay, India to estimate the Dst for the Carrington event. Based on Dst, the strongest geomagnetic storms in history were in 1921 and 1859. I also show on this list the strongest storms since 1960:

1) Dst = -1600, Carrington event, September 2, 1859
2) Dst = -900, May 14-15, 1921
3) Dst = -589, March 13, 1989 Superstorm
4) Dst = -472, November 20, 2003
5) Dst = -401, October 30, 2003

The 1921 event wiped out telegraph service east of the Mississippi. The currents induced in some telegraph wires were so strong that numerous fires were caused and several operators were injured by exploding consoles. Radio reception was completely lost in New Zealand, but was strengthened in Europe. Auroras were seen as far south as Puerto Rico.

Another measure of geomagnetic storm intensity is the change in amplitude of the magnetic field over time, dB/dt. Using this measure, the 1859 Carrington event was ten times stronger than the 1989 Superstorm.

Reduced vulnerability to geomagnetic storms?
One argument against a major disaster due to a repeat of the 1859 Carrington event is that increased awareness by system operators and improved forecasts since the 1989 event have made electric grids safer. Operators of the North American power grid constantly review and analyze the potential risks associated with space weather events, consulting space weather forecasts such as those produced by NOAA's Space Weather Prediction Center. They also monitor voltages and ground currents in real time and have emergency procedures in place to follow should a major geomagnetic storm hit. In October 2003, when a significant solar flare and coronal mass ejection (CME) triggered a geomagnetic storm about 75% as intense as the 1989 storm, NOAA's Space Weather Prediction Center issued a series of alerts, warnings, and predictions, giving power grid operators advance warning that severe space weather conditions were imminent that would put the power grid at risk. Despite severe geomagnetically-induced currents (GICs), power transmission equipment was protected and the grid maintained continuous operation.


Figure 2. Computer model study showing electrical systems that might be affected by a geomagnetic storm equivalent to the May 14-15, 1921 event. The regions outlined by the heavy black lines are susceptible to system collapse lasting months or years. A population in excess of 130 million might be affected, at a cost of $1-2 trillion in the first year after the event. The network of thin black lines shows the location of the nearly 80,000 miles long-distance heavy-hauling 345kV, 500kV and 765kV transmission lines in the U.S.--the main arteries of the U.S. electrical grid. The circles indicate magnitudes of geomagnetically-induced current (GIC) flow at each transformer in the network, and the color of the circle indicates the polarity of the current. Image credit: John Kappenman, Metatech Corp., The Future: Solutions or Vulnerabilities?, presentation to the space weather workshop, May 23, 2008.

Increased vulnerability to geomagnetic storms?
On the other hand, the evolution of open access on the electrical transmission system in recent years has resulted in the transport of large amounts of energy across the power system in order to maximize the economic benefit of delivering the lowest-cost energy to demand centers. The magnitude of power transfers has grown, and the increased level of transfers, coupled with multiple equipment failures, could aggravate the impacts of a storm. For example, the long distance between Hydro-Quebec's hydro-generation stations and load centers is one of the factors that is believed to have contributed to its crash during the 1989 Superstorm. In a remarkable 2008 National Academy of Sciences report, "Severe Space Weather Events--Understanding Societal and Economic Impacts Workshop Report", John Kappenman of Metatech Corporation theorizes that a future repeat of the Carrington event or the 1921 geomagnetic storm could result in catastrophic failure of large portions of the electrical grid that would last for years, costing 1-2 trillion dollars in the first year, and putting million of lives at risk. Full recovery from the event would take 4-10 years. The possible extent of a power system collapse from a repeat of the great magnetic storm of May 14-15, 1921--the second strongest geomagnetic storm in recorded history (Figure 2)--shows that a large region of the U.S. with a population of 130,000,000 might be affected.

The strong electric currents that would flow through the the electrical grid during a repeat of the Carrington event are likely to cause melting and burn-through of large-amperage copper windings and leads in electrical transformers. These multi-ton, multi-million dollar devices generally cannot be repaired in the field, and if damaged in this manner, they need to be replaced with new units. There are only a handful of spares in reserve, so most of the region affected by the collapse would remain without power until new transformers could be custom built. During the March 13, 1989 Superstorm, geomagnetic-induced currents (GICs) melted the internal windings of a 500kV transformer in the Salem Nuclear plant in southern New Jersey (Figure 3). The entire nuclear plant was unable to operate until this damaged transformer was replaced. Fortunately, a spare from a canceled nuclear plant in Washington State was available, and the Salem plant was able to reopen 40 days later. Had the spare not been available, a new custom-built transformer would have been required, potentially idling the power plant for years. The typical manufacture lead times for these transformers are 12 months or more. According to a January 2009 press release from Metatech, Inc., 300 Extra High Voltage (EHV) transformers in the U.S. would be at risk of permanent damage and require replacement in the event of a geomagnetic storm as intense as the 1921 or 1859 events. Here's where it gets really scary. According to the press release:

* Manufacturing capability in the world for EHV-class transformers continues to be limited relative to present market demand for these devices. Further, manufacturers would be unable to rapidly supply the large number of replacement transformers needed should the U.S. or other power grids suffer a major catastrophic loss of EHV Transformers.
.
* Manufacturers presently have a backlog of nearly 3 years for all EHV transformers (230 kV and above). The earliest delivery time presently quoted for a new order is early 2011.
.
* Only one plant exists in the U.S. capable of manufacturing a transformer up to 345 kV. No manufacturing capability exists in the U.S. at present for 500 kV and 765 kV transformers, which represent the largest group of At-Risk transformers in the U.S.

We have the very real possibility that a geomagnetic storms of an intensity that has happened before--and will happen again--could knock out the power to tens of millions of Americans for multiple years. The electrical grids in Europe and northern Asia have similar vulnerabilities, so a huge, years-long global emergency affecting hundreds of millions of people and costing many trillions of dollars might result from a repeat of the 1859 or 1921 geomagnetic storms.


Figure 3. The Salem nuclear plant transformer (exterior shot is one of three phases) and two images of internal heating damage to conductors and insulation from stray flux heating caused by geomagnetically-induced currents from the March 13, 1989 Superstorm. Image credit: John Kappenman, Metatech Corporation.

When might another Carrington event occur?
Geomagnetic storms on the scale of the 1859 or 1921 events are very rare, and no one knows when such an event may recur. Huge geomagnetic storms can occur at any portion of the 11-year sunspot cycle, but are most likely within a year of solar maximum. The 1921 event occurred three years after solar maximum, and the 1859 and 1989 events within a year of solar maximum. According to NASA's Bruce Tsurutani, a massive X22+ solar flare event on April 2, 2001, near the peak of the last solar cycle, was even larger than the flare that triggered the 1859 Carrington event. Fortunately, the 2001 flare was not pointed at the Earth, and we escaped a repeat of the Carrington event.


Figure 4. The largest solar flare ever recorded was observed on April 2, 2001. It was rated X-22 on a scale that only goes from one to twenty. The flare was more powerful than the flare that accompanied the worst geomagnetic storm in history, the 1859 Carrington event. Fortunately, the 2001 flare was not aimed at the Earth. Image credit: NASA.

We're currently at the deepest solar minimum since 1913, according to NASA. The sun has been remarkably free of sunspots this year, and it now looks like the next solar maximum will be at least 13 years removed from the previous peak in 2000. It could be much longer than that--unpublished research at the University of Michigan suggests that it might not be until 2030 that we'll see the next real solar maximum. A similar quiet solar period, called the Maunder Minimum, occurred in the late 1600s.

Given our history of two geomagnetic storms capable of causing large U.S. blackouts in the past 160 years, the odds of a potentially catastrophic space weather event are probably around 1-2% per year. A catastrophic failure of the electrical system from such an event is not a sure thing, but we should anticipate the possibility. Doing the research for this post has made me quite concerned about the possibility of long-term blackouts in the U.S., and I am planning on keeping a few more emergency supplies on hand as a result (this includes enough gasoline to drive to Michigan's Upper Peninsula, which would be less likely to get hit with power outages). Blackouts like the August 2003 blackout that affected 50 million people and lasted for 24-36 hours can also occur due to such mundane causes as trees interfering with power lines. Had that blackout occurred in winter, it could have lasted several days longer, since power plants take a much longer time to start in cold weather. Emergency generators typically only have fuel for 72 hours, so everything from hospital services to water pumping ability to natural gas delivery for heating will be threatened by large regional blackouts lasting more than 72 hours.

What can be done to reduce our vulnerability?
According to a January 2009 press release from Metatech, Inc., the installation of supplemental transformer neutral ground resistors to reduce GIC flows is relatively inexpensive, has low engineering trade-offs, and can produce 60-70% reductions of GIC levels for storms of all sizes. A Congressionally mandated "EMP Commission" has estimated the cost of this hardening in the existing U.S. power grid infrastructure to be on the order of $150 million. It would also be helpful to replace the ailing ACE satellite, which monitors solar storms and can provide advance warning of when a major geomagnetic storm is imminent. In any case, the future expansion of the electrical grid throughout the world needs to be designed with geomagnetic storms in mind. If large solar and wind power generation plants are developed, they will likely require an extensive new network of 765 kV transmission lines to deliver this energy. The higher voltage transformers needed for this expansion are the most vulnerable type of transformers to geomagnetic storms, and the new power system should be carefully designed to reduce this vulnerability.

For further reading
A May 2013 study by AER and Lloyd's (available here) concluded that the total US population at risk of extended power outage from a Carrington-level storm is between 20-40 million, with durations of 16 days to 1-2 years. The duration of outages will depend largely on the availability of spare replacement transformers. If new transformers need to be ordered, the lead-time is likely to be a minimum of five months. The total economic cost for such a scenario is estimated at $0.6-2.6 trillion USD.

January 2009 press release from Metatech, Inc., "An Overview of the National Academy of Sciences Report on Severe Space Weather and the Vulnerability of US Electric Power Grid".

August 2008 Scientific American article, "Bracing the Satellite Infrastructure for a Solar Superstorm".

2008 National Academy of Sciences study, , "Severe Space Weather Events--Understanding Societal and Economic Impacts Workshop Report"

March 2009 newscientist.com article on Space Weather threats.

Excellent 2007 lecture by John Kappenman of Metatech Corporation, "Electric Power Grid Vulnerability to Geomagnetic Storms" (50 minutes).

Jeff Masters