Geomagnetic storms present a unique risk to power grids, power grid customers and insurers. The modern world, and particularly the East Coast of the U.S., is increasingly vulnerable to widespread, prolonged power disruptions and outages caused by geomagnetic storms.
While strong storms have occurred in recent history, for several decades, Earth has been spared from extreme space weather events like those of 1859 and 1921. Yet, even weaker geomagnetic storms are causing attritional losses to utility infrastructure, and utility operators are regularly compelled to take protective measures due to space weather. Power and utility insurers, as well as traditional first-party insurers, should consider the potential impact of geomagnetic storms on their portfolios.
Extreme geomagnetic storms are a "low-frequency, high-impact" event. But a high-impact event could result in widespread, prolonged power outages. Consequently, the issue is under review at the highest levels of government in the U.S., the United Kingdom and elsewhere.
A severe solar storm could severely impact the global economy. Key reading on the subject should begin with the seminal 2008 National Academies report and the January 2010 Metatech report and continue through the October 2012 Federal Energy Regulatory Commission notice of proposed rule-making that may lead to directives that could require the U.S. power industry to provide reliability standards for geomagnetic disturbances.
The authority for space weather warnings in the U.S. is the National Oceanic and Atmospheric Administration's Space Weather Prediction Center (SWPC), in Boulder, Colo. The SWPC works closely with colleagues in the U.K. and around the world on providing space weather products to the large and growing community in need of such services.
The Science of Space Weather
Space weather science is fascinating. "Space weather," in the context of this article, focuses on coronal mass ejections from the sun. Coronal mass ejections (CMEs) are the electromagnetically charged detritus that regularly explode in the vicinity of sunspot clusters on the surface of the sun.
It is widely reported that 2013 will be the height of the 11-year solar spot cycle. There is, however, little causal link between the solar cycle and the frequency or intensity of geomagnetic storms. Further, not all CMEs are directed at Earth and some that are directed at Earth have little or no impact.
CMEs take 16 to 90 hours to travel from the sun to Earth. Power infrastructure may be damaged when CMEs interact with Earth’s magnetic field — the magnetosphere. Satellites, GPS applications, telecommunications equipment and aviation components may also be affected.
After the CME impacts the magnetosphere and ionosphere, currents are transmitted to the surface of Earth. The geomagnetically induced current (GIC) can enter power systems through the neutral earth connections and have adverse effects on critical grid infrastructure.
At its most benign, geomagnetic storms are visible as the northern lights; at its most damaging, geomagnetic storms generate GIC, which can result in costly damage to key grid assets or even blackout. Grid capacitors may be tripped and transformers may be damaged by GIC.
GIC causes half-cycle saturation in exposed transformers. While no transformer is totally immune from GIC effects, older single-phase transformers are generally considered to be more susceptible than newer three-phase transformers. A reasonable sample of transformers on the extra-high voltage transmission system suggests that the average age is greater than 30 years out of an anticipated 40-year economic life.
In addition, a large percentage of these transformers are of the single-phase design. Design and topology of the extra-high and high-voltage systems also make these systems more susceptible to the effects of GIC. This puts the equipment on these systems more at risk than equipment on lower voltage systems.
Half-cycle saturation produces harmonics, requires additional reactive current and can create damaging hotspots, which can result in immediate or delayed transformer failure or shorten the lifespan of equipment by damaging or prematurely aging the insulation.
Harmonics, produced at levels not normally seen under normal operation, can impact capacitor banks, protective relays, generators and other equipment, while the additional reactive current required by the transformers can lead to system voltage fluctuations. The severe damage of significant assets and complications resulting from the effects related to transformer saturation directly poses problems for system-wide stability, which may lead to localized or extensive grid blackouts.
Insurers may not be aware that power operators in exposed areas have detailed protocols to protect infrastructure from GIC. Protocols are varied and specific to the location and power infrastructure in question and may involve temporary reconfigurations.
Other operators have relied on more comprehensive measures: Some use series capacitors to block GIC, though some believe that series capacitors present additional unique problems. Only two transmission lines on the U.S. East Coast have series capacitors.
The varied and, arguably, piecemeal approach to adopting protective measures against geomagnetic storms should be contrasted with the billions of dollars spent to protect power infrastructure from lightning strikes.
Damage and Attritional Losses
Most power risks are insured under a form of all-risks insurance. Unless damage is excluded, sublimited or somehow limited in the wording, losses are payable up to policy limits. Physical damage and business interruption exposures to GIC are significant. Geomagnetic risk is rarely sublimited, excluded or even specifically contemplated by policy wordings.
However, some policy terms, conditions and exclusions may have an application in connection with space weather losses. The lifespan of transformers can be shortened by GIC, and insulation can be damaged by half-cycle saturation caused by GIC.
A number of transformers survived the 1989 geomagnetic storm that damaged the Hydro-Quebec system but were taken out of service in the years after the event. And in 2003, a number of Eskom transformers in South Africa survived the Halloween storm but were taken out of service in the year after the event.
Power insurers may not be liable for prematurely aged equipment. Space weather losses may not always be sudden and unforeseen. Wear and tear, gradual deterioration and other exclusions may also afford insurers some comfort.
Potential Impact of Large Event
Using modest modeling assumptions for a mid-latitude solar flare impact, Metatech’s report identifies more than 300 large power transformers in the U.S. that are at risk of failure from a severe event.
It should be noted that CMEs can impact wide areas of both the Northern and Southern Hemisphere as it was seen during the 2003 Halloween storm when it impacted areas in Sweden and South Africa. Power infrastructure in other areas around the world faces similar risks, but vulnerability varies widely due to design differences.
Large transformers are often bespoke equipment. It may take up to 24 months to build, test, transport and install a new transformer. Typically few, if any, compatible spare transformers are kept by power operators, and transporting transformers is a complex and expensive logistical exercise.
Until recently, there were no U.S. manufacturers of large power transformers. The combined global capacity to produce the number of transformers that may be required in a severe event is uncertain. The 24-month average lead time assumes, of course, that there has not been a widespread damaging solar event and does not take into account the increase in worldwide demand from other countries that may be similarly affected by the event.
In the event of a widespread event, repairs to power infrastructure in North America could take years. It is hard to overstate the threat to the global economy posed by a multiyear cessation or interruption of power to the East Coast of the U.S.
Beyond Power and Utilities
Like power infrastructure, satellites, telecommunications and aviation are uniquely exposed to solar flares. And almost every first-party risk that relies on the regular supply of electricity is exposed to widespread power disruption caused by space weather.
Without specific policy wording to the contrary, U.S. courts often liberally construe claims for business interruption losses that can be directly or indirectly attributed to power outages.
The potential business interruption and contingent business interruption losses from a solar event could rival or outpace the volume and quantum of recent Sandy, Thailand, Japan, New Zealand, Australian, various Gulf Coast hurricanes and 9/11 losses.
And while the industry has had recent cause to reflect on the significance of contingent business interruption CBI) wordings, it is not clear that CBI claims linked to space weather have been thoroughly addressed.
Risk Surveys, Modeling and Loss Investigations
Risk surveys for power infrastructure (and other risks with unique exposure to space weather such as aviation, telecoms and satellites) should provide information on exposure to space weather. This may include analysis of space weather protocols, reviewing the impact of recent solar events, reporting on the geology, commenting on the adequacy of protective measures, inspecting insulation and connections and modeling the impact of solar events of varying magnitudes and latitudes.
For more generic first-party risks, risk surveys and modeling should contemplate the impact of widespread, sustained power grid disruptions. Insurers should expect loss notifications, investigations and reports from experts to address these issues.
An improved awareness is the best first step in responding constructively to space weather risks. Modeling and reporting will at least place insurers in a position equal to their insureds. And policies usually include exclusions for nuclear war, terrorism and even electromagnetic pulses — risks that are arguably more remote than a recurring natural phenomenon.
Insurers incorporate sublimits for earthquakes, floods and other catastrophe risks. Such exclusions acknowledge that the magnitude of some losses is beyond the scope and purpose of insurance. Sublimits also provide some certainty and cap exposure to manageable levels.
Like hurricanes, earthquakes and other natural catastrophe phenomena, space weather and geomagnetic storms are routine occurrences — it is a question of when, not if, losses will be caused by space weather. Incorporating policy exclusions and sublimits would be an easy way for insurers to limit exposure.
Every loss investigation should contemplate a link to damage caused by solar flares, which may trigger exclusions, concern condition precedents or confirm that a loss may be for an earlier policy period. With better reporting, risk surveys and modeling, there may even be an appetite for new solar flare and power outage insurance products.
The opinions expressed are those of the authors and do not necessarily reflect the views of the firm, its clients, or Portfolio Media Inc., or any of its or their respective affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.