Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, NATURAL HAZARD SCIENCE (oxfordre.com/naturalhazardscience). (c) Oxford University Press USA, 2018. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 17 December 2018

The Economic Impact of Critical National Infrastructure Failure Due to Space Weather

Summary and Keywords

Space weather is a collective term for different solar or space phenomena that can detrimentally affect technology. However, current understanding of space weather hazards is still relatively embryonic in comparison to terrestrial natural hazards such as hurricanes, earthquakes, or tsunamis. Indeed, certain types of space weather such as large Coronal Mass Ejections (CMEs) are an archetypal example of a low-probability, high-severity hazard. Few major events, short time-series data, and the lack of consensus regarding the potential impacts on critical infrastructure have hampered the economic impact assessment of space weather. Yet, space weather has the potential to disrupt a wide range of Critical National Infrastructure (CNI) systems including electricity transmission, satellite communications and positioning, aviation, and rail transportation.

In the early 21st century, there has been growing interest in these potential economic and societal impacts. Estimates range from millions of dollars of equipment damage from the Quebec 1989 event, to some analysts asserting that losses will be in the billions of dollars in the wider economy from potential future disaster scenarios. Hence, the origin and development of the socioeconomic evaluation of space weather is tracked, from 1989 to 2017, and future research directions for the field are articulated. Since 1989, many economic analyzes of space weather hazards have often completely overlooked the physical impacts on infrastructure assets and the topology of different infrastructure networks. Moreover, too many studies have relied on qualitative assumptions about the vulnerability of CNI. By modeling both the vulnerability of critical infrastructure and the socioeconomic impacts of failure, the total potential impacts of space weather can be estimated, providing vital information for decision makers in government and industry.

Efforts on this subject have historically been relatively piecemeal, which has led to little exploration of model sensitivities, particularly in relation to different assumption sets about infrastructure failure and restoration. Improvements may be expedited in this research area by open-sourcing model code, increasing the existing level of data sharing, and improving multidisciplinary research collaborations between scientists, engineers, and economists.

Keywords: economic analysis of natural hazards, infrastructure, space weather

Introduction

Space weather takes place when disturbances in Earth’s upper atmosphere and near-Earth space are capable of detrimentally affecting a wide range of key technologies (Hapgood et al., 2016), particularly Critical National Infrastructure (CNI) systems such as electricity transmission, satellite communications and positioning, aviation, and rail transportation. Whereas the CNI impacts from terrestrial natural hazards (such as hurricanes and earthquakes) have been widely studied, space weather has received little comparative attention. This disparity is partially due to the low-probability, high-impact nature of this particular threat, and because attributing specific technological problems to space weather can be challenging. The 1859 Carrington event—named after Richard Carrington, who observed the activity from his private observatory in South London—is the most popularized example of a space weather event. Although the Carrington event led to many issues on Earth (e.g., in the telegraph system), our technology has significantly progressed since the mid-19th century. Therefore, it is hard to determine how current CNI systems would respond to a modern-day Carrington event.

Growing recognition of this natural hazard is demonstrated by recent policy developments over the past decade aimed at bolstering national readiness to space weather threats. In the United Kingdom, a Space Weather Preparedness Strategy has been released (Cabinet Office & BIS, 2015) and recognition has been given of this threat on the National Risk Register of Civil Emergencies (Cabinet Office, 2017). On the other side of the Atlantic, in the United States, on October 13th, 2016, President Obama signed an executive order (13744) titled Coordinating Efforts to Prepare the Nation for Space Weather Events, which outlined the roles and responsibilities of different federal government agencies in addressing the risks posed by space weather hazards.

While the study of space weather has progressed significantly, from scholarly recordings of astronomical events many centuries ago to advanced modeling of how solar activity may drive geomagnetic disturbances on Earth, substantial efforts still need to be made to further our understanding of a variety of research areas. These range from fundamental scientific research, to investigating the vulnerability of engineered systems, to assessing the potential socioeconomic impact of CNI failure. While there has been ongoing research in the science and engineering domains for many decades, potential socioeconomic impacts have received little attention. This neglect has been identified as a shortcoming of space weather research for a decade (Lanzerotti, 2008), with little redress. Indeed, although there is a newly invigorated desire to remedy this situation, there has been relatively little published in the peer-reviewed literature. Additionally, a lack of standardized methodology has led to divergent results.

In light of these issues, this article first provides a general introduction to space weather and then considers its potential impact on CNI. Third, the evolution of the economic analysis and impact of space weather is explored. The strength of the analysis undertaken to date is then examined, and this information is put to use to inform the direction of future research. Finally, possible solutions on how to enhance the current level of research regarding space weather and its impacts are presented.

What Is Space Weather?

Space weather arises from many different types of eruptive phenomena associated with solar activity occurring on the surface of the sun (often referred to as “solar storms”). Consequently, the interaction of three primary forms of solar activity with Earth (or in near-Earth space) causes space weather (Figure 2 [see “Impacts on CNI Systems”] provides an illustration of the potential impacts for each type of solar phenomenon), as follows:

  1. 1. Coronal Mass Ejections (CMEs) are massive releases of billions of tons of charged particles and magnetic fields from the surface of the sun (Webb & Howard, 2012).

  2. 2. Solar Energetic Particle Events (SEPs) consist of huge increases in energetic particles, mainly of protons but also of heavy ions, thrown out into space (Shea & Smart, 2012).

  3. 3. Solar flares impart a rapid release of electromagnetic energy previously stored in inductive magnetic fields. Emitted radiation covers most of the electromagnetic spectrum, from radio waves to X-rays (Fletcher et al., 2011).

When these primary forms occur in combination, the time line of impacts is likely to unfold as follows. Earth may first be bombarded with initial radiation (such as X-rays) from a solar flare approximately 8 minutes after the event on the surface of the sun. A second barrage of very high-energy solar particles (SEPs) may then arrive some 10s of minutes later. Finally, a large CME may reach Earth somewhere between 1 and 4 days later, depending on the speed of travel through interplanetary space (see Liu et al., 2014). The magnetic field in the CME is likely to lead to a geomagnetic storm that may also last for multiple days as it drives huge electrical currents, especially at high geomagnetic latitudes, leading to bright auroral displays. Often two CMEs may be released in quick succession, and analysis of past events suggest that this dual occurrence often leads to the most extreme impacts, as indicated by aurora occurring at low latitudes (Vaquero, Valente, Trigo, Ribeiro, & Gallego, 2008; Willis, Armstrong, Ault, & Stephenson, 2005; Ribeiro, Vaquero, & Trigo, 2011).

Although extreme space weather events usually include all three of these solar phenomena, the most concerning is associated with multiple very large and fast Carrington-sized CMEs. CMEs are a key driver of coronal and interplanetary dynamics, particularly if a CME traveling across interplanetary space hits Earth in a southward magnetic field direction (Bz), as it can lead to the most dramatic interaction effect and therefore the largest geomagnetic disturbances (Webb & Howard, 2012). A southward-directed CME interacts with the northward direction of Earth’s magnetic field, leading to a canceling effect, allowing CME energy to enter Earth’s magnetic field. While auroras are often seen during modest forms of geomagnetic activity, they are generally enhanced by large CMEs, at which point the auroral band may expand equatorward to lower latitudes.

Aurora are caused by bands of charged particles being accelerated along Earth’s magnetic field lines into the atmosphere, exciting atmospheric gases that then give off the light we see (see Figure 1). Usually these visual displays occur in the auroral oval regions encircling Earth’s poles and are indicative of geomagnetic activity. The auroral oval regions can be altered in modest forms by the solar wind or in more extreme circumstances by a CME interacting with the planet’s atmosphere.

The Economic Impact of Critical National Infrastructure Failure Due to Space WeatherClick to view larger

Figure 1. Example of the aurora borealis in Yukon Territory, Canada.

Photo from Good Free Photos.

On average, the Sun’s magnetic activity follows an 11-year solar cycle, with variable minimum and maximum periods. The latest solar cycle began in 2008 with minimal activity during the first few years. However, on July 23rd, 2012, an extremely large CME, estimated to be similar in size to the Carrington event, narrowly missed Earth (Baker et al., 2013). While the 2012 London Olympics was generally regarded as a success, had this CME hit Earth the day of the opening ceremony, it would have been a very different story.

The strength and complexity of the Sun’s magnetic field changes throughout the solar cycle, manifesting in visible “sunspots” on the surface due to regions of concentrated magnetic field. The solar atmosphere changes during the sun cycle from a magnetically simple state to a complex configuration producing a larger number of sunspots, which are a key indicator of solar activity (Green & Baker, 2015). While there may be more activity during some parts of the solar cycle, such as the declining phase (Juusola et al., 2015), solar eruptive phenomena are still the result of a random process. Therefore, there is potential for a significant space weather event to affect Earth at any time. Table 1 provides a summarized list of major space weather events, some of the impacts they caused, and literature references for each case.

Table 1. Summary of the Historical Storm Catalogue for Major Space Weather Events

Year

Impact

Reference

1847

Spontaneous electrical currents observed in telegraph wires in the British Midlands, along railway corridors from Derby to Rugby, Birmingham, Leeds, and Lincoln.

Barlow, 1849; Prescott, 1860; Cade, 2013

1859

The archetypal example of space weather. Known as the Carrington event, significant disruption occurred to telegraph systems across the globe, and auroras were witnessed down to very low latitudes.

Boteler, 2006; Siscoe et al., 2006; Green & Boardsen, 2006; Ribeiro et al., 2011; Rodger et al., 2008; Saiz et al., 2016; Silverman, 2006; Tsurutani et al., 2003

1870

A large storm produced aurora sightings in Lisbon and Coimbra (Portugal), Greenwich (U.K.), Munich (Germany), and Helsinki (Finland).

Vaquero et al., 2008

1872

Auroras were sighted as low as 10°–20° geomagnetic latitude, with significant recordings in Mumbai.

Moos, 1910a, 1910b; Uberoi, 2011

1882

Strong auroras recorded in Scandinavia and North America.

Rubenson, 1882; Lewis, 1882

1921

Similar in size to the Carrington event, with significant GIC generated in Scandinavia.

Karsberg et al., 1959; Silverman & Cliver, 2001; Kappenman, 2006

1940

Damage caused to the U.S. telephone system and reported effects on the electricity network.

Harang, 1941; Davidson, 1940

1958

Transatlantic communications were disrupted between Newfoundland and Scotland. A blackout occurred in the Toronto area.

Anderson, 1978; Lanzerotti & Gregori, 1986

1989

The Quebec power grid collapsed within 90 seconds. The well-documented Quebec power outage lasted 9 hours.

Bolduc, 2002; Medford et al., 1989

2000

The Bastille Day Event saw a very large CME and flare.

Tsurutani et al., 2005

2003

The Halloween Storms included a mix of CMEs and flares, leading to a 1-hour power outage in Sweden. This storm also led to a radio blackout of high-frequency communications as well as disruption of GPS systems.

Pulkkinen et al., 2005; Tsurutani et al., 2005; Bergeot et al., 2010

Historical accounts record auroral sightings going back millennia, and ground-based magnetograph data have been recorded since the 19th century. However, it is only since the space age that large-scale digital monitoring of space weather events has been undertaken, approximately over the past 50 years. Although considerable focus has been placed on the most extreme space weather events, such as a Carrington-level storm, it has been suggested that the potential economic impact of a prolonged period of moderate activity may be comparable to a single large incident (Schrijver, 2015). Indeed, research has suggested that just dealing with day-to-day space weather can pose a reliability challenge for electricity operators (for further detail, see the work of Forbes & St. Cyr, 2008, 2012, 2017). Given the technological impacts associated with this storm catalogue, it is now pertinent to review the potential impacts to critical infrastructure.

Impacts on CNI Systems

Space weather has the capability of disrupting the critical technologies that comprise the national infrastructure system, although impacts vary by sector. This section provides an overview of the technologies potentially affected by space weather to help inform the economic impact assessment of this hazard. For a further summary of space weather worst-case environments, see Hapgood et al. (2016), or for a detailed analysis of the impacts on engineered systems, see Cannon et al. (2013). As illustrated in an impact tree in Figure 2, the three key types of space weather can affect power grids, satellite systems, radio communications, aviation, rail transport, pipelines, and undersea cables. Each of these impacts are briefly discussed in this section.

The Economic Impact of Critical National Infrastructure Failure Due to Space WeatherClick to view larger

Figure 2. Space weather impact tree.

Adapted from Hapgood et al. (2016).

The largest focus has hitherto been on the threat to electricity transmission infrastructure, partly because energy underpins practically all daily activities and a prolonged loss of power would be disastrous for those affected. During a geomagnetic storm, geomagnetically induced currents (GICs) are generated that are able to subject power grid assets to excessive thermal heating and voltage instability issues, potentially resulting in a loss of power. The speed of change in Earth’s magnetic field affects the generation of GIC, which can lead to immediate or cumulative damage in transformer components (Hutchins & Overbye, 2011). This rate of change of the magnetic field is best measured by dB/dt, which represents the time derivative of magnetic field variations on the ground (Kataoka & Ngwira, 2016). During a geomagnetic storm, many rapid global-scale variations in Earth’s field and current systems repeatedly occur. Known as “substorms,” these cause the most rapid changes in the magnetic field at the surface of Earth and produce the largest GIC. Molinski (2002) states that this phenomenon gives rise to half-cycle saturations in transformers, potential system voltage collapse, a loss in reactive power, as well as the generation of harmonics and excessive transformer heating. Geomagnetic latitude, ground conductivity, and the power system network structure can influence the risk posed by this hazard. Figure 3 illustrates this hazard, including the damage pathway for electricity transmission infrastructure. Disruption to energy infrastructure can also lead to cascading failure, affecting other critical infrastructures such as transportation, digital communications, and vital public health systems.

The Economic Impact of Critical National Infrastructure Failure Due to Space WeatherClick to view larger

Figure 3. Detailed damage pathway for electricity transmission infrastructure.

Adapted from Boteler (2015), and Samuelsson (2013).

Considerable research examining the impacts of geomagnetic activity on electricity transmission transformers has been conducted, particularly in South Africa. This research was intensified as a consequence of geomagnetic activity spanning from October and November, 2003 (the “Halloween Storms”), which caused significant problems to many assets in the South African grid, leading to increased focus by electrical engineering researchers on these issues (see Gaunt & Coetzee, 2007, and Gaunt, 2014 for further detail, as well as Matandirotya, Cilliers, & Van Zyl, 2015, for an example of the GIC measurement and modeling research focusing on electricity transmission infrastructure).

Spacecraft and satellites, including those enabling global positioning systems (GPS), are susceptible to space weather phenomena, particularly radiation bursts (Astafyeva, Yasyukevich, Maksikov, & Zhivetiev, 2014). Indeed, a study of on-orbit spacecraft failures by Tafazoli (2009) found that over 10% of spacecraft anomalies were due to solar or magnetic storms. These impacts include both short-term and long-term effects. First, problems with signal propagation and transmission may occur during the event due to interference caused by ionospheric storms and scintillation (Horne et al., 2013; Hapgood et al., 2016). Indeed, a loss of satellite capability can immediately disrupt many other infrastructure systems and economic sectors that rely on communications, navigation, and timing services. Long-term issues, however, include spacecraft drag, which can cause uncontrolled re-entry for satellites in low orbits, or spacecraft charging issues that can affect onboard electronics. For satellites in higher orbits (e.g., geosynchronous), issues arise from both solar array damage and spacecraft charging (Hastings & Garrett, 2004; Garrett & Whittlesey, 2012; Lai & Tautz, 2006). Solar arrays and other electronic components can be degraded when cosmic rays and solar energetic particles penetrate them (see Koons & Fennell, 2006). Power failures are often critical for spacecraft, as 45% result in complete loss of spacecraft and 80% significantly affect the mission (Tafazoli, 2009).

In particular, high-frequency (HF) radio communications can be temporarily disrupted due to radio absorption (Neal, Rodger, & Green, 2013; Rodger, Verronen, Clilverd, Seppälä, & Turunen, 2008). During large electromagnetic radiation bursts released from the surface of the Sun, HF radio blackouts may occur and last 1 hour or so, with the largest effects taking place in low latitude regions where the Sun is highest in the sky. During a barrage of very high-energy solar particles (SEPs), the greatest impacts can occur in the polar regions, sometimes lasting for several days. In addition, during periods of extreme space weather, aviation routes may need to be rerouted to avoid high latitude regions due to probable disruption to HF communications (Neal et al., 2013) and to avoid radiation risks to passengers and crew. Moreover, airline operations suffer problems with avionics and GPS/GNSS navigation systems during extreme events (Jones et al., 2005), which could potentially cause delays at major airports around the globe, particularly where there are a large number of flights that use the high latitude regions as an aviation corridor (e.g., New York to Tokyo or Toronto to Hong Kong). New ambitions to send human missions to Mars and beyond also raise concerns about potential radiation exposure in space (Cougnet et al., 2004).

Rail transportation can be affected, mainly from signals and tracks being subjected to GIC in two different ways. First, signaling and train control system anomalies may occur during periods of high geomagnetic activity. Indeed, failures not related to recognized technical malfunctions can, on average, be up to seven times more likely (Ptitsyna et al., 2008; Eroshenko et al., 2010; Wik et al., 2009). Second, the structural integrity of rail infrastructure may be affected if repeatedly subjected to extreme conditions, as this may increase the rate of corrosion. However, less evidence has been found to support this impact. This type of exposure is similar in nature to the risk to pipelines, particularly if there is cumulative long-term damage due to GIC, as this damage can increase the likelihood of corrosion, shortening an asset’s life (Pulkkinen, Pirjola, Boteler, Viljanen, & Ngwira, 2001a; Pulkkinen, Viljanen, Pajunpää, & Pirjola, 2001b; Gummow & Eng, 2002). Uncertainty exists regarding the magnitude of repeated exposure and the amount of time before an asset becomes affected, as exposure would not lead to immediate failure. Hence, it may not be possible to attribute the damage caused by exposure to high GIC to a space weather event if the asset eventually fails many months or years after.

Effects to communications cables can take place in three different ways: (1) historical impacts on regional copper wire electric telegraph and telephone systems; (2) historical impacts on copper wire transoceanic cables; and (3) modern impacts on power systems in optical-fiber transoceanic cables. During the Carrington event of 1859, many telegraph operators reported strange electrical effects. In fact, telegraph communications were still able to transmit and receive information even after systems were disconnected from the power supply due to GIC running through the cables. More recently, a documented case endeavored to understand the problems caused by a large storm in 1958 whereby businesses and consumers in Finland were disrupted by the failures of two coaxial phone cable systems in the southern part of the country (Nevanlinna, Tenhunen, Pirjola, Annanpalo, & Pulkkinen, 2001). The event was caused by blown fuses associated with the AC power supplies at repeater stations. Moreover, much like power grids, the submarine equivalent is equally affected by certain geographic and technical factors which, in this case, include the depth of the cable (Meloni, Lanzerotti, & Gregori, 1983). However, we have seen in recent decades a revolution in the technologies used to transmit information in digital communications networks. Modern systems rely on fiber optics, with glass fiber being far less conductive than copper. Hence, electrical cables that power fiber-optic equipment are far more at risk (Medford, Lanzerotti, Kraus, & Maclennan, 1989). The next section focuses purely on potential impacts to the electricity transmission infrastructure, as this area has been the key subject of study for almost three decades.

The Evolution of the Economic Analysis of Space Weather

In this section, different chronological periods are analyzed based on how the study of the economic impacts of space weather has evolved over time. Prior to the geomagnetic storm of 1989 and the voltage collapse of the Hydro-Quebec electricity transmission grid, there were few major examples of critical infrastructure failure attributed to space weather events. Hence, there had been limited analysis of the consequential economic impacts. Therefore, developments are tracked beginning after this event, with analysis broken down into three key temporal periods: 1989–2007, 2008–2013, and 2014–2017. The justification for these three chronological periods is provided in each subsection. The papers cited in this part of the article are summarized in Table 2.

Table 2. Summary of Studies Focusing on the Economic Impact Assessment of Space Weather

Year

Author

Infrastructure Type

Geography

Spatiotemporal Impact

Economic Methodology

Economic Impact

Peer Reviewed?

Country

Region

Population Affected

Restoration Period

Asset Damage

Direct Economic Impact

Indirect Economic Impact

Total Economic Impact

1990

Barnes and Van Dyke

Electricity transmission infrastructure

United States

Northeast

Not stated

50% connected in 16 hours, 75% in 24 hours, 100% in 48 hours

Value of Lost Load estimation

$16 million (1988 USD)

$3-6 billion (1988 USD)

Not modeled

Not modeled

Yes

2002

Bolduc

Electricity transmission infrastructure

Canada

Quebec

9 million

N/A

Not stated

$13.2 million (Canadian dollars)

Not modeled

Not modeled

Not modeled

Yes

2005

Pulkkinen et al.

Electricity transmission infrastructure

Finland

Malmö

50,000

1 hour

Not stated

Not stated

$0.5 million (USD)

Not modeled

Not modeled

Yes

2008

Kappenman (in Space Studies Board)

Electricity transmission infrastructure

United States

National assessment

Not stated

4 to 10 years

Not stated

Not stated

$1–2 trillion (USD)

Not stated

Not stated

No

2008

Forbes and St. Cyr

Electricity transmission infrastructure

Multiple countries

National assessments

N/A

N/A

Econometrics

N/A

N/A

N/A

N/A

Yes

2012

Forbes and St. Cyr

Electricity transmission infrastructure

United States

National assessments

N/A

N/A

Econometrics

N/A

N/A

N/A

N/A

Yes

2013

Atmospheric Environmental Research for Lloyd’s of London

Electricity transmission infrastructure

North America

N/A

20–40 million

16 days to 1–2 years

Value of Lost Load estimation

Not stated

$0.6–2.6 trillion (USD)

Not modeled

Not modeled

No

2014

Schulte in den Bäumen et al.

Electricity transmission infrastructure

Global

National assessment

Not stated

5 months to 1 year

Multi-Regional Input–Output analysis

Not modeled

Not stated

Not stated

$3.4 trillion (USD)

Yes

2014

Schrijver et al.

Electricity transmission infrastructure

North America

N/A

N/A

N/A

Retrospective cohort exposure study with controls

Not stated

~4% of claims are statistically associated with geomagnetic activity

Not modeled

Not modeled

Yes

2017

Forbes and St. Cyr

Electricity transmission infrastructure

England and Wales

N/A

N/A

N/A

Econometrics

N/A

N/A

N/A

N/A

Yes

2017

Oughton et al.

Electricity transmission infrastructure

United States

National assessment

8–66%

24 hours

Multi-Regional Input–Output analysis

Not modeled

$3–28.2 billion (USD)

$1.4–7.2 billion (USD)

$4.4–35.4 billion (USD)

Yes

1989–2007

After the 1989 geomagnetic storm, various researchers began to focus on the potential impacts on the power grid, including calculations on the economic costs of both asset damage, the unserved electricity load, and replacement power. Mitigation costs were first examined by Douglas (1989), who explored investment into neutral blocking and grounding devices for the electricity transmission grid in the wake of the Quebec event, where six million people lost power (Boteler, 1991). Then cost information surfaced in a paper by Bolduc (2002), a researcher at the Hydro-Québec Research Institute, focusing on GIC observations in the Hydro-Québec power system. The cost of damaged equipment to Hydro-Québec from overvoltages was $6.5 million (Canadian) in material damages alone, among a total of $13.2 million (although this amount is likely to be relatively insignificant when compared to the wider economic impact). Indeed, the damages to transmission equipment were quite severe as it took many months to get some assets repaired and fully operational again.

In the same March 1989 event, utilities in the northern United States also experienced problems. Consequently, the U.S. Department of Energy funded research by Barnes and Van Dyke (1990) at the Oak Ridge National Laboratory, who undertook an analysis of a geomagnetic disturbance affecting the power grid in the northeast of the United States for a 48-hour period. It is stated in the analysis that several transformers were damaged and removed from service during the 1989 event, in particular at the Salem nuclear power plant in New Jersey, where the replacement cost was reported as “several million dollars, and the replacement energy cost was about $400,000 a day for 6 weeks, while the plant was shut down” (Barnes & Van Dyke, 1990, p. 3). Fortunately, replacement transformers were available, otherwise the plant could have been closed for up to a year due to the prolonged delivery times of extra high voltage (EHV) transformers.

In the analysis by Barnes and Van Dyke, a hypothetical scenario was explored where 95% of those states served by the Northeast Power Coordinating Council (Maine, Vermont, New Hampshire, Massachusetts, New York, Connecticut, and Rhode Island), as well as New Jersey and the majority of Pennsylvania, were without power due to a blackout caused by a geomagnetic disturbance. The authors used a power restoration process where 50% of the population was reconnected after 16 hours, 75% after a day, and 100% after 2 days. This event was assumed to cause damage at two nuclear power plants to four single-phase power transformers. The methodology focused on estimating the Value of the Lost Load (VOLL) during the period of the blackout. VOLL is a monetary indicator expressing the costs associated with electricity supply interruption (Schröder & Kuckshinrichs, 2015); however, it only captures the direct effects of electricity supply interruption and does not factor in the multiplier impacts that may accrue throughout the economy. Following a literature review, the estimated cost impacts were determined to be between $1.87 and $3.33 per unserved kWh of demand (1988 USD) for both residential and industrial and commercial customers. Using estimates for the regional average hourly energy load, the lost load of unserved electricity was then multiplied by the lower and upper costs per kWh to obtain an estimated range.

The costs of replacement power were also calculated based on damaged transformers, as replacement power was assumed to be required for 12 months while new transformers were constructed, given bespoke spares were unlikely to be found. A loss of two nuclear 1100 MW power plants (operating at 65% capacity) led to a power replacement cost of $313 million to $1.253 billion (1988 USD) for 12 months, depending on the cost per kWh. Based on these assumptions, the direct economic costs ranged from $3.0 billion to $6.1 billion. The value associated with the unserved load ranged between 79% and 89% of the cost; the electricity replacement cost was smaller at between 10% and 20%, and the transformer cost was negligible. Kappenman (1996) later quoted this study and related these impacts to other natural hazards that occurred in 1989, stating it was equivalent to Hurricane Hugo or the San Francisco earthquake.

In comparison to the March 1989 event, the October 2003 geomagnetic storm was less severe and led to less of an economic impact. Due to ongoing maintenance work on the transmission grid in southern Sweden, the Malmö region underwent a large-scale blackout because of significant GICs. According to Pulkkinen et al. (2005), for approximately 1 hour about 50,000 people were left without power, although this outage led to a relatively minor economic impact of approximately $0.5 million US dollars from the unserved electricity. The loss of power was reported to have caused significant local issues, however, delaying many trains, and leaving many people stranded in elevators.

In reviewing these studies, it appears that the economic impact assessments undertaken between 1987 and 2007 were relatively basic and focused first on the cost to damaged infrastructure assets and, second, on utilizing VOLL techniques to calculate direct economic impact. The wider economic impacts were hence not considered.

2008–2013

Almost two decades after the Quebec 1989 incident, a workshop was held in Washington DC on May 22, 2008 under the auspices of the U.S. National Research Council’s Space Studies Board. Bringing together industry, federal government representatives, and social scientists, the workshop focused on the potential societal and economic impacts of severe space weather and produced an account of the event in the form of an extended summary report (see Space Studies Board, 2008). This report raised the profile of the potential socioeconomic impacts of space weather by examining the level of disruption to CNI. In particular, Kappenman’s contribution within the report points to an estimate by the Metatech Corporation that “the total cost of a long-term, wide-area blackout caused by an extreme space weather event could be as much as $1 trillion to $2 trillion during the first year, with full recovery requiring 4 to 10 years depending on the extent of the damage” (Space Studies Board, 2008, p. 13). This figure became widely quoted following the event despite little evidence to support the claim. Indeed, tracing the spatial and temporal assumptions used to arrive at this figure was challenging, but due to the size of the proposed impact, one assumed this figure covered both direct and indirect costs to the economy. Within the report, the author made comparative reference to the 2003 blackout that took place in the northeast of the United States and Ontario, Canada (a non-space weather-induced critical infrastructure failure), which affected 50 million people and led to an estimated cost of between $4 billion and $10 billion, according to the US–Canada Power System Outage Task Force (2004).

A Metatech report was later prepared by Kappenman (2010) for Oak Ridge National Laboratory. Essentially a vulnerability assessment, the report provided a detailed overview of the modeling and analysis undertaken. It also explored why the replacement of extra high voltage transformers would take such a protracted period of time. Later, a JASON report (2011) was specifically tasked with assessing the worst-case scenario put forward by Kappenman, which proposed that the U.S. transmission grid would undergo catastrophic damage, leaving millions without essential electricity services for anywhere between a few months to a few years. Tasked by the U.S. Department of Homeland Security, one of the primary objectives of the study was to assess “the plausibility of Mr. Kappenman’s worst-case scenario” (JASON, 2011, p. 1), including (1) catastrophic damage to >300 extra high voltage transformers, (2) 130 million people without power for several years, and (3) a $1–2 trillion (USD) economic impact.

However, the findings of the report highlight some key issues regarding the original analysis. First, both data and algorithms used in the analysis were stated to be proprietary and were therefore unavailable for examination by the JASON investigators; hence the report emphasized that national policy should not be based on methods that were not fully transparent and available to key decision makers. Second, the nature and characteristics of the extra high voltage transformers were not well known, as exemplified in the analysis by the extrapolation of small samples of data. Therefore, it was incredibly challenging to robustly quantify how many might fail when exposed to different levels of GIC. Finally, experiences of other transmission infrastructure systems exposed to high GIC had not necessarily experienced catastrophic damage (e.g., in both Quebec and Finland), making the authors doubt the suggested impacts in the worst-case scenario. The study concludes by stating:

We agree that the U.S. electric grid remains vulnerable but are not convinced that Kappenman’s worst-case scenario [26] is plausible, i.e. that a severe solar storm will probably destroy up to 300 EHV transformers, leaving as many as 130 million people without power for years while replacement transformers are manufactured and installed.

(JASON, 2011, pp. 64–65)

This conclusion was an important turning point in the narrative associated with the economic impacts of space weather, as it raised doubts over extremely long restoration periods (and hence trillion-dollar impacts). However, the idea that a space weather event could cause trillions of lost dollars was alarming for many risk bearers, who consequently undertook their own scenario-based analysis. One example related to the insurance industry who not only insure the physical assets of many critical infrastructure operators, but also cover property, casualty, and supply chain insurance more generally throughout society and the economy. Hence, the worry was that if this type of tail event actually took place, it could be damaging enough to put many insurers out of business, especially as existing vehicles for spreading risk may not address space weather.

Consequently, Lloyd’s of London (2013) commissioned Atmospheric and Environmental Research (AER) to undertake an assessment of the risk to the North American grid. The analysis can be seen as a primer for corporate risk managers to begin to understand how CMEs can drive geomagnetic storms on Earth. The assessment initially covers risk factors that may increase exposure such as geomagnetic latitude, ground conductivity, coastal effects, and transmission system characteristics. The assessment uses Carrington-level geomagnetic storm simulations for the geomagnetic field and then relates these to local ground conductivity structures (see Wei et al., 2013). A power grid model is then used to assess the level of GIC flowing through transformers (identified using commercially available transformer data) to assess transformer vulnerability due to thermal heating. Grid instability impacts are therefore not modeled. Using estimated transformer age distributions and temperature information, outage scenarios are estimated at the county level.

The AER assessment appendix details how the economic costs are derived, using a similar VOLL method as that of Barnes and Van Dyke (1990), by calculating the cost of the unserved electricity in 2001 USD. A linear restoration is assumed using $2.00/kW, $19.38/kW, and $8.40/kW for residential, commercial, and industrial users. The study finds that for a Carrington-level storm, whereby 20-40 million people are affected for between 16 days and up to 1-2 years, the total economic cost is estimated at between $0.6 trillion and $2.6 trillion USD. The logic for this analysis in terms of impact zone size and restoration time is similar to that used in the proposed Kappenman worst-case scenario, but again transformer age distributions and characteristics are essentially assumed. In the recommendations of the JASON (2011) report, it is emphasized that the actual transformer asset data should be collected to enable comprehensive simulation of the entire grid. Although improvements were made in the analysis of electricity transmission infrastructure in the period between 2008 and 2013, key disagreements arose due to diverging views on how engineered systems would respond. In some cases, these differences led to some very large economic impacts but with little detail on the economic methodology used to justify the proposed numbers.

2014–2017

In 2014, a multidisciplinary group of scientists and economists published a paper on how severe space weather can disrupt global supply chains (Schulte in den Bäumen, Moran, Lenzen, Cairns, & Steenge, 2014). The key contribution of the analysis was that for the first time a more robust economic methodology was applied to space weather impact assessment, superior to the previous method of roughly calculating the lost value of the unsupplied electricity load.

None of the previously published estimates considered global impacts to international trade; therefore, this analysis was an important step forward in utilizing standard macroeconomic methods. The methodology applied, known as input–output economics, is a field of analysis developed by Wassily Leontief in the late 1930s, for which he was awarded the Nobel Prize in Economic Science in 1973 (see Miller & Blair [2009] for a comprehensive overview of the method). Importantly, the techniques provide a formal framework for analyzing interindustry transactions (often monetary flows), and allow one to model the impacts of changes within the economy, both directly and indirectly. This method has been a frequent workhorse used to understand the potential economic ramifications of infrastructure failure (see Haimes & Jiang, 2001; Anderson, Santos, & Haimes, 2007; Leung, Haimes, & Santos, 2007; Setola, De Porcellinis, & Sforna, 2009; Pant, Barker, Hank, & Landers, 2011; Pant, Barker, & Zobel, 2014; Jonkeren & Giannopoulos, 2014). In this example, Schulte in den Bäumen et al. (2014) first applied a physical model calibrated to the latitudinal (80°) and longitudinal (8°) width of the auroral electrojet, and then applied a set of scenarios across different continents, in both the northern and southern hemisphere, for a storm similar to the Quebec 1989 event. This application was the first time a physical model had been coupled with a global macroeconomic model and was a key contribution to the literature. Moreover, a set of Multi-Regional Input–Output (MRIO) tables were utilized, called Eora, developed by the coauthor Lenzen, which, depending on the country, covered between 25 and 400 industrial sectors, for a total of 187 countries of the world. These countries represented 99.99% of global trade. A total grid shutdown was then modeled in those countries within the storm impact zone footprint.

Importantly, the authors made the key point that, as all economic assessments to date had not considered indirect trade effects, they had not included domestic and international supply chain linkages in the estimates. This observation led to a total economic loss in the United States of $25 billion USD per day, similar to the Lloyd’s of London estimate of approximately $30 billion USD per day. Moreover, a Carrington-level event taking place over North America was estimated to reach $1.2 trillion USD over five months. Indeed, in the discussion the authors state that a “severe space weather event could be the worst natural disaster in modern history with global cost estimates to be over 5% of world Gross Domestic Product (GDP) and impacts reaching across every industry and every segment of society” (Schulte in den Bäumen et al., 2014, p. 2756). Indeed, the results of the paper indicated that the total economic impact could be $3.4 trillion USD over a year, which is approximately 5.6% of global GDP.

In another key contribution, Schrijver, Dobbins, Murtagh, and Petrinec (2014) analyzed a novel dataset of insurance claims belonging to the insurance company Zurich. For January 2000 up to December 2010, 11, 242 insurance claims were analyzed for equipment loss and affiliated business interruption for corporates located in North America. The key finding was that on days of elevated geomagnetic activity, the claim rate increased by approximately 20%, for the top 5% of most active days. Overall, claim rates were elevated by approximately 10% for the top third of the most active days when ranked by the maximum variation in Earth’s geomagnetic field. This finding suggests that large-scale geomagnetic activity causing variations in the quality of power provided can induce equipment faults in electrical and electronic devices, potentially leading to an estimated 500 additional claims on average per year. Importantly, this analysis used data from a relatively low activity period for space weather, so this number could rise if a particular increase in solar activity occurred. The financial implications of these claims are substantial. Indeed, they are unlikely to be related to space weather by both the firms affected and the wider insurance industry, demonstrating that many of the cost impacts go unattributed. Research by Forbes and St. Cyr (2008, 2012, 2017) also focuses on day-to-day space weather impacts on power grids, which can still pose a challenge for network operators, potentially incurring substantial operational costs.

One particular problem with the level of analysis to date is the challenge of reconciling global or continental-scale geophysical activity with local impacts on critical infrastructure. Indeed, while the use of macroeconomic modeling methodologies is an important step forward for estimating total economic impact (e.g., Schulte in den Bäumen et al., 2014), the local impacts of critical infrastructure failure on firms, labor, and value-added activity are lost. Within the United States, there is such heterogeneity among different states, particularly in industrial composition, that modeling impacts at the macroeconomic level can produce quite coarse results. In a study by Oughton, Skelton, Horne, Thomson, and Gaunt (2017), an assessment of electricity transmission failure due to space weather was undertaken, focusing on the United States. Different scenario-based storm footprints were tested to explore the variation in direct and indirect economic impacts (see Figure 4). The weighted population centroid of each state was represented by a black dot in Figure 4, with the state being included in each scenario if each centroid fell within a set of electrojet footprints (see Oughton et al. [2017] for further details), These scenarios are exploratory in nature, with S1 being the most probable, and S4 being highly improbable.

The Economic Impact of Critical National Infrastructure Failure Due to Space WeatherClick to view larger

Figure 4. Tested blackout zones, customer disruptions, and daily direct economic impact. S1–S4 represent four scenarios reported in Oughton et al. (2017).

The methodology was similar to that used by Schulte in den Bäumen et al. (2014) in that it utilizes MRIO data to quantify domestic and international supply chain impacts, although it does build on a different dataset, called the World Input Output Database (WIOD). However, the method was supplemented by using state-level GDP output data for 20 industrial sectors and used these data to drive national shocks in the macroeconomic model, allowing the local heterogeneity in economic activity to be quantified. Due to disagreement within the scientific and engineering communities on the length of potential power outages, the analysis focused on the economic impact for a 24-hour period. Depending on the storm footprint tested, the total economic impact ranged from $6.2 billion to $41.5 billion USD per day to the U.S. economy.

Reflecting on these analyses, both Oughton et al. (2017) and Schulte in den Bäumen et al. (2014) approximately scaled the impact footprint based on the latitudinal and longitudinal width of the electrojet. However, recent research by Ngwira et al. (2015) and Pulkkinen, Bernabeu, Eichner, Viljanen, and Ngwira (2015) suggests that zones of extreme activity have significantly smaller footprints, with substorms taking place in localized areas within the electrojet, indicating that these analyses could overestimate impacts. Like much of the literature to date, the shortcomings of these analyses are that they lack robust scientific and engineering inputs, hence they pay little attention to ground conductivity and underlying grid structure. Indeed, this exemplifies the fact that many analyses on the economic impacts of space weather have lacked the level of rigor required by the scientific and engineering community.

Now that the literature over the past three decades has been critiqued, the strengths of different approaches are next discussed, and an ontology is provided for assessing the economic impacts of space weather in future research.

An Ontology for Assessing the Economic Impacts of Space Weather

Reflecting on the relative strengths and weaknesses of different approaches, this section proposes how future research may progress. Assessing the economic impacts of natural hazards due to critical infrastructure failure is a well-developed field and does not need to be revisited (see Ouyang [2014] for a comprehensive overview on modeling infrastructure systems, including economic methods). However, the economic analysis of space weather is yet to bring together the strengths of different approaches to comprehensively assess the total impacts; such information is vitally needed by decision makers in government and industry to help inform cost-benefit assessments for resilience.

First, the actual economic costs of space weather have been calculated in a variety of ways in the literature. Many studies have focused purely on a single type of cost, such as damage to network assets, the direct cost of unserved electricity, or the wider economic impacts via supply chains. No study has yet convincingly quantified the implications of each of these economic costs, despite this information being essential to gain a comprehensive understanding. Moreover, some have called for a greater focus on assessing both direct and indirect losses (see Eastwood et al., 2017, p. 213), but this activity should not be undertaken in isolation. Indeed, the conjecture proposed here is that mitigation costs should also be considered in order to enable the type of resilience analytics actually needed to support real decisions (see Rose [2017] for further background on the benefit-cost analysis of economic resilience actions).

Finally, economic impacts have sometimes been ambiguously defined in the literature, and therefore it is important to identify to whom they relate. For example, there could be costs to infrastructure network operators as service providers, to firms as the key actors carrying out production activities in the wider economy, and also to households as consumers of final goods and services. Based on the ideas articulated here, Table 3 presents the different types of economic costs associated with space weather, and which stakeholders bear them.

Table 3. The Economic Costs Associated with Space Weather

Entity

Cost TYPE

Type of Damage or Mitigation Costs

Infrastructure network operator

Direct

Damage to assets

Idle resources

Lost sales of electricity

Indirect

Delayed supply of replacement assets

Increased cost of replacement assets

Mitigation

Operational mitigation measures

Asset upgrades and blocking devices

System upgrades including islanding

Commercial and industrial customers

Direct

Production downtime

Delayed scheduling of activities

Indirect

Delayed supply of upstream goods and services

Delayed delivery of downstream goods and services

Mitigation

Backup power

Inventory stockpiling

Higher electricity tariffs to fund network upgrades

Households

Direct

Lost leisure time

Property and casualty damages

Indirect

Constrained consumer spending due to unavailable goods and services

Price increases due to short supply of goods and services

Mitigation

Backup power

Higher electricity tariffs to fund network upgrades

As many studies have focused on different types of costs, a large divergence in proposed estimates has often been seen (e.g., estimates for equipment damage costs reported after the Quebec event vis-à-vis scenario-based estimations of wider economic impacts). These divergent estimates are exacerbated by a high level of uncertainty associated with the spatial and temporal impacts of potential space weather events.

Within the catastrophe modeling paradigm, techniques for addressing uncertainty already exist with regard to quantifying the potential impact of earthquakes, hurricanes, and other natural hazards by capturing both primary and secondary causes of uncertainty. Primary uncertainty relates to the size and location of the storm impact zone for a space weather event and the problems in predicting its occurrence. For example, the initial impact of the most intense substorm could affect North America, Europe, or East Asia, depending on the daily rotation of the Earth. But this uncertainty also reflects the fact that we have quite poor time-series data and only a partial understanding of the physical processes driving space weather. Subsequently, the variability in local intensity, asset damage, and different economic impacts pertain to the level of secondary uncertainty. Figure 5 illustrates an ontology for assessing the economic impacts of space weather, recognizing both primary and secondary drivers of uncertainty and potential direct and indirect economic impacts.

The Economic Impact of Critical National Infrastructure Failure Due to Space WeatherClick to view larger

Figure 5. Ontology for assessing the economic impacts of space weather.

Local risk factors can either increase or decrease potential risk. As already identified, ground conductivity, geomagnetic latitude, and coastal proximity are important geophysical parameters that affect the amount of GIC produced, although realistically the amount is difficult to predict. Many economic assessments have not utilized actual infrastructure network data, which has led to an overly simplistic estimation of impacts. Thus, Figure 5 highlights the need to use network exposure data, which has the advantage of being able to anchor vulnerability and risks to assets within a spatial context. This anchoring is vital for robustly quantifying potential economic costs to the network operator as well as to households and firms. Damage to assets can be quantified as replacement or repair costs to estimate the economic impact on the network operator.

A variety of mitigation options are available which have the potential to limit damage to infrastructure assets. In general, mitigation options are different from those for other natural hazards due to the characteristics of space weather. For instance, earthquake risk requires assets to be made more structurally robust in order to withstand increased physical vibration, and flooding risk requires assets to be elevated or surrounded by embankments. Yet with space weather, specific technological components within infrastructure assets can be most affected and require hardening or protecting in other ways, such as the winding conductors of EHV transformers. As well as physical installation of mitigation equipment (e.g., in the power grid) and enhancement of space weather forecasting and early warning systems, a variety of operational mitigation measures can be implemented across the network. These include (1) increasing spinning reserve, (2) canceling maintenance in order to make as many lines as possible operational, and (3) introducing islanding and microgrid configurations. However, many operational measures are highly dependent on satellite capabilities to provide early warnings that space weather events are actually taking place. Therefore, ensuring this satellite capability is crucial for ensuring operational awareness and protecting critical infrastructure.

Finally, wider economic impacts occur to both households and firms. Where asset data for infrastructure networks can be challenging to obtain, population and business statistics are relatively ubiquitous across national statistical bureaus. Often these statistics are available at a high level of granularity, and if no specific network structure is available, then data can be spatially joined to the closest network asset by using open-source software. Importantly, firms are often part of both domestic and international supply chains and therefore these impacts need to be quantified for a comprehensive cost assessment. Indeed, upstream economic impacts take place indirectly because firms directly affected by critical infrastructure failure buy fewer goods and services. Moreover, downstream economic impacts also accrue as firms directly affected by the event are unable to sell their goods and services to other firms, which may often be critical for their production processes. In the case of final products, consumers may not be able to purchase their desired goods and services, which would lead to reduced aggregate demand in the economy.

Each of the areas within Figure 5 has been more or less addressed within the literature already; however, each area has often been addressed in isolation, preventing assessment of the true economic cost of space weather. The concepts of primary and secondary uncertainty, while cornerstones of 21st-century catastrophe modeling, have yet to percolate into the economics of space weather. This is partly because the level of analysis as of 2018 has been relatively piecemeal, with few researchers focusing their attention on this topic for a prolonged period, which inevitably means there is little consistent analysis, limited incremental refinement of assessment models, and no examination of the sensitivities of assumption sets. These are all areas relevant for future research.

Conclusions

This article tracks how the economic impact assessment of space weather has evolved from 1989 to 2017, focusing mainly on the risks posed to electricity transmission infrastructure. Although a number of key contributions have taken place, the economic analysis of this natural hazard still lags behind other natural catastrophe threats such as hurricanes or earthquakes, partially due to the low-probability, high-impact nature of extreme space weather. While we have not experienced a Carrington-sized event in recent times, we do regularly experience moderate space weather activity. Yet, often we may fail to properly attribute problems induced by this hazard, partially due to a lack of awareness, but also because failure of infrastructure assets and other electrical equipment may take place in the days, weeks, or months following an event. Despite space weather continuing to receive growing attention across industry and government, much progress needs to be made in trying to understand the degree of secondary uncertainty associated with the potential economic impacts of space weather. This includes exposure, vulnerability, and potential damage to key infrastructure assets.

As well as charting the evolution of the literature, this article proposes an ontology for the comprehensive economic assessment of space weather. This ontology has been based on the relative strengths of certain methods and is motivated by the need for evidence when considering mitigation measure decisions. Hitherto, key studies have independently addressed economic impacts to key entities involved, such as infrastructure network operators, firms concerned with impacts to the wider supply chain, or players in the insurance market. However, future work needs to bring together the potential detrimental cost implications for each of the entities within the economy and assess these impacts in relation to investment in mitigation measures, allowing a more advanced understanding of the cost-benefit trade-offs associated with this natural hazard.

A critical issue with regard to assessments carried out over the past three decades is that many completely overlook actual physical impacts on infrastructure assets and the topology of infrastructure networks, often relying far too heavily on qualitative assumptions about Critical National Infrastructure (CNI) vulnerability. This oversight is problematic because it gives rise to a relatively weak evidence base. Although data on infrastructure assets and the network topology can sometimes be challenging to obtain, the tools are available to explore the sensitivity of the uncertainty associated with CNI vulnerability. Indeed, the JASON (2011) report states that researchers should be making more use of simulation techniques to increase the evidence base on this matter. This situation also relates to data sharing, which can be a perennial issue, especially when dealing with asset information that is potentially business sensitive.

In terms of best practice, as efforts on this matter have historically been relatively piecemeal, with few authors publishing more than once on the economic impacts of space weather, little explicit exploration of model sensitivities, including examination of different assumption sets, has taken place. Increasingly, open-source model code is becoming the gold standard, although the author found little evidence of researchers moving in this direction. Given the importance of this under-researched hazard and the need for national policy to be based on fully transparent data, models, and tools, taking these steps toward best practices would be a worthy contribution to the economic impact assessment of space weather.

Acknowledgments

The author would like to express his gratitude to the U.K. Engineering and Physical Science Research Council for financially supporting this researach under grant EP/N017064/1: Multi-scale InfraSTRucture systems AnaLytics.

References

Anderson, C. W. (1978). Magnetic Storms and Cable Communications. In C.F. Kennel, L. J. Lanzerotti, & E.N. Parker (Eds.), Solar System Plasma Physics (pp. 323–327). Amsterdam: North-Holland.Find this resource:

Anderson, C. W., Santos, J. R., & Haimes, Y. Y. (2007). A risk-based input-output methodology for measuring the effects of the August 2003 northeast blackout. Economic Systems Research, 19(2), 183–204.Find this resource:

Astafyeva, E., Yasyukevich, Y., Maksikov, A., & Zhivetiev, I. (2014). Geomagnetic storms, super-storms, and their impacts on GPS-based navigation systems. Space Weather, 12(7), 2014SW001072.Find this resource:

Baker, D. N., Li, X., Pulkkinen, A., Ngwira, C. M., Mays, M. L., Galvin, A. B., & Simunac, K. D. C. (2013). A Major Solar Eruptive Event in July 2012: Defining Extreme Space Weather Scenarios. Space Weather, 11(10), 585–591.Find this resource:

Barlow, W. H. (1849). VI. On the Spontaneous Electrical Currents Observed in the Wires of the Electric Telegraph. Philosophical Transactions of the Royal Society of London, 139 (January), 61–72.Find this resource:

Barnes, P. R., & Van Dyke, J. W. (1990). Economic consequences of geomagnetic storms (a summary). IEEE Power Engineering Review, 10(11), 3–4.Find this resource:

Bergeot, N., Bruyninx, C., Defraigne, P., Pireaux, S., Legrand, J., Pottiaux, E., & Baire, Q. (2010). Impact of the Halloween 2003 ionospheric storm on kinematic GPS positioning in Europe. GPS Solutions, 15(2), 171–180.Find this resource:

Bolduc, L. (2002). GIC observations and studies in the hydro-Québec power system. Journal of Atmospheric and Solar-Terrestrial Physics: Space Weather Effects on Technological Systems, 64(16), 1793–1802.Find this resource:

Boteler, D. H. (1991). Predicting geomagnetic disturbances of power systems. Eos, Transactions American Geophysical Union, 72(14), 159–160.Find this resource:

Boteler, D. H. (2006). The super storms of August/September 1859 and their effects on the telegraph system. In C. R. Clauer & S. Siscoe (Eds.), The great historical geomagnetic storm of 1859–A modern look [Special issue]. Advances in Space Research, 38(2), 159–172.Find this resource:

Boteler, D. H. (2015). The Impact of Space Weather on the Electric Power Grid. In C. J. Schrijver, F. Bagenal, & Sojka (Eds.), Heliophysics V. Space Weather and Society. Palo Alto, CA: Lockheed Martin Solar & Astrophysics Laboratory.Find this resource:

Cabinet Office. (2017). National risk register of civil emergencies—2017 edition. London, UK: Cabinet Office.

Cabinet Office, and Department for Business, Innovation & Skills. (2015). Space weather preparedness strategy. Version 2.1. London, UK: Cabinet Office and BIS.Find this resource:

Cade, W. B. (2013). The first recorded space weather impact? Space Weather, 11(9), 489.Find this resource:

Cannon, P., Angling, M., Barclay, L., Curry, C., Dyer, C., Edwards, R., . . . Greene, G. (2013). Extreme space weather: Impacts on engineered systems and infrastructure. London, UK: Royal Academy of Engineering.

Cougnet, C., Crosby, N. B., Eckersley, S., Foullon, C., Guarnieri, V., Guatelli, S., . . . Heynderickx, D. (2004). Radiation exposure and mission strategies for interplanetary manned missions (REMSIM). Earth, Moon, and Planets, 94(3–4), 279–285.Find this resource:

Davidson, W. F. (1940). The magnetic storm of March 24, 1940—Effects in the power system. Edison Electric Institute Bulletin, 8, 365–366.Find this resource:

Douglas, J. (1989). A storm from the sun. IEEE Power Engineering Review, 9(12), 11–13.Find this resource:

Eastwood, J. P., Biffis, E., Hapgood, M. A., Green, L., Bisi, M. M., Bentley, R. D., . . .Burnett, C. (2017). The Economic Impact of Space Weather: Where Do We Stand? Risk Analysis, 37(2), 206–18.Find this resource:

Eroshenko, E. A., Belov, A. V., Boteler, D., Gaidash, S. P., Lobkov, S. L., Pirjola, R., & Trichtchenko, L. (2010). Effects of strong geomagnetic storms on northern railways in Russia. Advances in Space Research, 46(9), 1102–1110.Find this resource:

Fletcher, L., Dennis, B. R., Hudson, H. S., Krucker, S., Phillips, K., Veronig, A., . . . Battaglia, M. (2011). An observational overview of solar flares. Space Science Reviews, 159(1–4), 19–106.Find this resource:

Forbes, K. F., & O. C. St. Cyr. (2008). Solar Activity and Economic Fundamentals: Evidence from 12 Geographically Disparate Power Grids. Space Weather, 6(10).Find this resource:

Forbes, K. F., & St. Cyr, O.C. (2012). Did Geomagnetic Activity Challenge Electric Power Reliability during Solar Cycle 23? Evidence from the PJM Regional Transmission Organization in North America. Space Weather, May.Find this resource:

Forbes, K. F., & St. Cyr, O. C. (2017). The challenge posed by geomagnetic activity to electric power reliability: Evidence from England and Wales. Space Weather, 15(10), 2017SW001668.Find this resource:

Garrett, H. B., & Whittlesey, A. C. (2012). Guide to mitigating spacecraft charging effects. Hoboken, NJ: Wiley.Find this resource:

Gaunt, C. T. (2014). Reducing uncertainty—Responses for electricity utilities to severe solar storms. Journal of Space Weather and Space Climate, 4, A01.Find this resource:

Gaunt, C. T., & Coetzee, G. (2007). Transformer Failures in Regions Incorrectly Considered to Have Low GIC-Risk. In Power Tech, 2007 IEEE Lausanne, 807–12. Lausanne, Switzerland: IEEE.Find this resource:

Green, J. L., & Boardsen, S. (2006). Duration and extent of the great auroral storm of 1859. In C. R. Clauer & S. Siscoe (Eds.), The great historical geomagnetic storm of 1859–A modern look [Special issue]. Advances in Space Research, 38(2), 130–135.Find this resource:

Green, L., & Baker, D. (2015). Coronal mass ejections: A driver of severe space weather. Weather, 70(1), 31–35.Find this resource:

Gummow, R. A., & Eng, P. (2002). GIC effects on pipeline corrosion and corrosion control systems. Journal of Atmospheric and Solar-Terrestrial Physics: Space Weather Effects on Technological Systems, 64(16), 1755–1764.Find this resource:

Haimes, Y., & Jiang, P. (2001). Leontief-based model of risk in complex interconnected infrastructures. Journal of Infrastructure Systems, 7(1), 1–12.Find this resource:

Hapgood, M., Angling, M., Attrill, G., Burnett, C., Cannon, P., Gibbs, M., . . . Harrison, R. (2016). Summary of space weather worst-case environments. Revised edition (Report RAL-TR-2016-006).Find this resource:

Harang, L. (1941). Maximalwerte Der Erdstromspannungen in Der Nähe Der Nordlichtzone Während Sehr Intensiver Erdmagnetischer Störungen [Maximum values of Earth current near the auroral zone during very intensive geomagnetic disturbances] (in German, with English summary). Gerlands Beitrage zur Geophysik, 57, 310.Find this resource:

Hastings, D., & Garrett, H. (2004). Spacecraft–environment interactions. Cambridge, UK: Cambridge University Press.Find this resource:

Horne, R. B., Glauert, S. A., Meredith, N. P., Boscher, D., Maget, V., Heynderickx, D., & Pitchford, D. (2013). Space weather impacts on satellites and forecasting the Earth’s electron radiation belts with SPACECAST. Space Weather, 11(4), 169–186.Find this resource:

Hutchins, T. R., & Overbye, T. J. (2011). The Effect of Geomagnetic Disturbances on the Electric Grid and Appropriate Mitigation Strategies. North American Power Symposium (NAPS), 2011, 1–5. IEEE.Find this resource:

JASON. (2011). Impacts of Severe Space Weather on the Electric Grid. JSR-11-320. Mclean, Virginia: JASON.

Jones, J. B. L., Bentley, R. D., Hunter, R., Iles, R. H. A., Taylor, G. C., & Thomas. D. J. (2005). Space weather and commercial airlines. Advances in Space Research, Space Weather, 36(12), 2258–2267.Find this resource:

Jonkeren, O., & Giannopoulos, G. (2014). Analysing critical infrastructure failure with a resilience inoperability input–output model. Economic Systems Research, 26(1), 39–59.Find this resource:

Juusola, L., Viljanen, A., van de Kamp, M., Tanskanen, E. I., Vanhamäki, H., Partamies, N., & Kauristie, K. (2015). High-latitude ionospheric equivalent currents during strong space storms: Regional perspective. Space Weather, 13(1), 2014SW001139.Find this resource:

Kappenman, J. (2010). Geomagnetic storms and their impacts on the U.S. power grid. Goleta, CA: Metatech. Prepared for Oak Ridge National Laboratoy.

Kappenman, J. G. (1996). Geomagnetic storms and their impact on power systems. IEEE Power Engineering Review, 16(5), 5.Find this resource:

Kappenman, J. G. (2006). Great geomagnetic storms and extreme impulsive geomagnetic field disturbance events—An analysis of observational evidence including the great storm of May 1921. In C. R. Clauer & S. Siscoe (Eds.), The great historical geomagnetic storm of 1859–A modern look [Special issue]. Advances in Space Research, 38(2), 188–199.Find this resource:

Karsberg, A., Swedenborg, G., & Wyke, K. (1959). The influences of Earth magnetic currents on telecommunication lines. Tele (English Edition), Televerket (Swedish Telecom), 1, 1–21.Find this resource:

Kataoka, R., & Ngwira, C. (2016). Extreme geomagnetically induced currents. Progress in Earth and Planetary Science, 3(August), 23.Find this resource:

Koons, H. C., & Fennell, J. F. (2006). Space Weather Effects on Communications Satellites. Radio Science Bulletin, 316, 27–41.Find this resource:

Lai, S. T., & Tautz, M. (2006). High-level spacecraft charging in eclipse at geosynchronous altitudes: A statistical study. Journal of Geophysical Research: Space Physics, 111(A9), A09201.Find this resource:

Lanzerotti, L. J., & Gregori, G. P. (1986). In Telluric Currents: The Natural Environment and Interactions with Man-Made Systems, by National Academy Press. Washington, DC: National Academy Press.Find this resource:

Lanzerotti, L. J. (2008). Economic impacts of space weather. Space Weather, 6(1), S01002.Find this resource:

Leung, M., Haimes, Y., & Santos, J. (2007). Supply- and Output-Side Extensions to the Inoperability Input-Output Model for Interdependent Infrastructures. Journal of Infrastructure Systems, 13(4), 299–310.Find this resource:

Lewis, H. C. (1882). Note on the aurora of April 16–17, 1882. Proceedings of the American Philosophical Society, 20(111), 283–291.Find this resource:

Liu, Y. D., Luhmann, J. G., Kajdič, P., Kilpua, E. K. J., Lugaz, N., Nitta, N. V., . . . Möstl, C. (2014). Observations of an extreme storm in interplanetary space caused by successive coronal mass ejections. Nature Communications, 5(March).Find this resource:

Lloyd’s of London. (2013). Solar storm risk to the North American electric grid. London, UK: Lloyd’s of London.Find this resource:

Matandirotya, E., Cilliers, P. J., & Van Zyl, R. R. (2015). Modeling geomagnetically induced currents in the South African power transmission network using the finite element method. Space Weather, 13(3), 185–195.Find this resource:

Medford, L. V., Lanzerotti, L. J., Kraus, J. S., & Maclennan, C. G. (1989). Transatlantic Earth potential variations during the March 1989 magnetic storms. Geophysical Research Letters, 16(10), 1145–1148.Find this resource:

Meloni, A., Lanzerotti, L. J., & Gregori, G. P. (1983). Induction of currents in long submarine cables by natural phenomena. Reviews of Geophysics, 21(4), 795–803.Find this resource:

Miller, R. E., & Blair, P. D. (2009). Input–output analysis: Foundations and extensions. Cambridge, UK: Cambridge University Press.Find this resource:

Molinski, T. S. (2002). Why utilities respect geomagnetically induced currents. Journal of Atmospheric and Solar-Terrestrial Physics, Space Weather Effects on Technological Systems, 64(16), 1765–1778.Find this resource:

Moos, N. A. F. (1910a). Magnetic observations made at the government observatory Bombay, 1847 to 1905, and their discussion: Part II. The phenomenon and its discussion. Bombay, India: Government Central Press.Find this resource:

Moos, N. A. F. (1910b). Magnetic observations made at the government observatory, Bombay, for the period 1846 to 1905, and their discussion: Part I. Magnetic data and instruments. Bombay, India: Government Central Press.Find this resource:

Neal, J. J., Rodger, C. J., & Green, J. C. (2013). Empirical determination of solar proton access to the atmosphere: Impact on polar flight paths. Space Weather, 11(7), 420–433.Find this resource:

Nevanlinna, H., Tenhunen, P., Pirjola, R., Annanpalo, J., & Pulkkinen, A. (2001). Breakdown caused by a geomagnetically induced current in the Finnish telesystem in 1958. Journal of Atmospheric and Solar-Terrestrial Physics, 63(10), 1099–1103.Find this resource:

Ngwira, C. M., Pulkkinen, A. A., Bernabeu, E., Eichner, J., Viljanen, A., & Crowley, G. (2015). Characteristics of extreme geoelectric fields and their possible causes: Localized peak enhancements. Geophysical Research Letters, 42(17), 2015GL065061.Find this resource:

Oughton, E. J., Skelton, A., Horne, R. B., Thomson, A. W. P., & Gaunt, C. T. (2017). Quantifying the Daily Economic Impact of Extreme Space Weather Due to Failure in Electricity Transmission Infrastructure. Space Weather, 15(1), 65–83.Find this resource:

Ouyang, M. (2014). Review on modeling and simulation of interdependent critical infrastructure systems. Reliability Engineering & System Safety, 121(January), 43–60.Find this resource:

Overbye, T. J., Shetye, K. S., Hutchins, T. R., Qiu, Q., & Weber, J. D. (2013). Power grid sensitivity analysis of geomagnetically induced currents. IEEE Transactions on Power Systems, 28(4), 4821–4828.Find this resource:

Pant, R., Barker, K., Hank G., F., & Landers, T. L. (2011). Interdependent impacts of inoperability at multi-modal transportation container terminals. Transportation Research Part E: Logistics and Transportation Review, 47(5), 722–737.Find this resource:

Pant, R., Barker, K., & Zobel, C. W. (2014). Static and dynamic metrics of economic resilience for interdependent infrastructure and industry sectors. Reliability Engineering & System Safety [Special issue of selected articles from ESREL 2012], 125(Suppl. C), 92–102.Find this resource:

Prescott, G. B. (1860). History, theory, and practice of the electric telegraph. Boston, MA: Ticknor and Fields.

Ptitsyna, N. G., Kasinskii, V. V., Villoresi, G., Lyahov, N. N., Dorman, L. I., & Iucci, N. (2008). Geomagnetic effects on mid-latitude railways: A statistical study of anomalies in the operation of signaling and train control equipment on the east-Siberian railway. Advances in Space Research, 42(9), 1510–1514.Find this resource:

Pulkkinen, A., Lindahl, S., Viljanen, A., & Pirjola, R. (2005). Geomagnetic Storm of 29–31 October 2003: Geomagnetically Induced Currents and Their Relation to Problems in the Swedish High-Voltage Power Transmission System. Space Weather, 3(8), S08C03.Find this resource:

Pulkkinen, A., Bernabeu, E., Eichner, J., Viljanen, A., & Ngwira, C. (2015). Regional-scale high-latitude extreme geoelectric fields pertaining to geomagnetically induced currents. Earth, Planets and Space, 67(1), 93.Find this resource:

Pulkkinen, A., Pirjola, R., Boteler, D., Viljanen, A., & Yegorov, I. (2001a). Modelling of space weather effects on pipelines. Journal of Applied Geophysics, 48(4), 233–256.Find this resource:

Pulkkinen, A., Viljanen, A., Pajunpää, K., & Pirjola. R. (2001b). Recordings and occurrence of geomagnetically induced currents in the Finnish natural gas pipeline network. Journal of Applied Geophysics, 48(4), 219–231.Find this resource:

Ribeiro, P., Vaquero, J. M., & Trigo, R. M. (2011). Geomagnetic Records of Carrington’s Storm from Guatemala. Journal of Atmospheric and Solar-Terrestrial Physics, Space Climate, 73(2–3), 308–15.Find this resource:

Rodger, C. J., Verronen, P. T., Clilverd, M. A., Seppälä, A., & Turunen, E. (2008). Atmospheric impact of the Carrington event solar protons. Journal of Geophysical Research: Atmospheres, 113(D23), D23302.Find this resource:

Rose, A. (2017, April). Benefit-cost analysis of economic resilience actions. In Oxford research encyclopedias: Natural hazard science.Find this resource:

Rubenson, R. (1882). Catalogue Des Aurores Boréales Observées En Suéde, Kgl. Sv. Vetenskaps Akad. Handl, 15.Find this resource:

Saiz, E., Guerrero, A., Cid, C., Palacios, J., & Cerrato, Y. (2016). Searching for Carrington-like events and their signatures and triggers. Journal of Space Weather and Space Climate, 6, A6.Find this resource:

Samuelsson, O. (2013). Geomagnetic Disturbances and Their Impact on Power Systems—Status Report 2013. Ind. Elec. Eng. Auto., Lund Univ., Lund, Sweden. Sweden: Lund University.Find this resource:

Schrijver, C. J. (2015). Socio-economic hazards and impacts of space weather: The important range between mild and extreme. Space Weather, 13(9), 2015SW001252.Find this resource:

Schröder, T., & Kuckshinrichs, W. (2015). Value of lost load: An efficient economic indicator for power supply security? A literature review. Frontiers in Energy Research, 3.Find this resource:

Schrijver, C. J., Dobbins, R., Murtagh, W., & Petrinec, S. M. (2014). Assessing the Impact of Space Weather on the Electric Power Grid Based on Insurance Claims for Industrial Electrical Equipment. Space Weather, 12(7), 487–98.Find this resource:

Schulte in den Bäumen, H., Moran, D., Lenzen, M., Cairns, I., & Steenge, A. (2014). How severe space weather can disrupt global supply chains. Natural Hazards and Earth Systems Science, 14(10), 2749–2759.Find this resource:

Setola, R., De Porcellinis, S., & Sforna, M. (2009). Critical infrastructure dependency assessment using the input–output inoperability model. International Journal of Critical Infrastructure Protection, 2(4), 170–178.Find this resource:

Shea, M. A., & Smart, D. F. (2012). Space weather and the ground-level solar proton events of the 23rd solar cycle. Space Science Reviews, 171(1–4), 161–188.Find this resource:

Silverman, S. M. (2006). Comparison of the aurora of September 1/2, 1859 with other great auroras. In C. R. Clauer & S. Siscoe (Eds.), The great historical geomagnetic storm of 1859–A modern look [Special issue]. Advances in Space Research, 38(2), 136–144.Find this resource:

Silverman, S. M, & Cliver, E. W. (2001). Low-latitude auroras: The magnetic storm of 14–15 May 1921. In V. Papitashvili, B. T. Tsuratani, & Y. I. Feldstein (Eds.), Interplanetary medium and geophysical phenomena during geomagnetic storms [Special issue]. Journal of Atmospheric and Solar-Terrestrial Physics, 63(5), 523–535.Find this resource:

Siscoe, G., Crooker, N. U., & Clauer, C. R. (2006). Dst of the Carrington storm of 1859. In C. R. Clauer & S. Siscoe (Eds.), The great historical geomagnetic storm of 1859–A modern look [Special issue]. Advances in Space Research, 38(2), 173–179.Find this resource:

Space Studies Board. (2008). Severe Space Weather Events–Understanding Societal and Economic Impacts: A Workshop Report. Washington, DC: National Academies Press.Find this resource:

Tafazoli, M. (2009). A study of on-orbit spacecraft failures. Acta Astronautica, 64(2–3), 195–205.Find this resource:

Tsurutani, B. T., Gonzalez, W. D., Lakhina, G. S., & Alex, S. (2003). The extreme magnetic storm of 1–2 September 1859. Journal of Geophysical Research: Space Physics, 108(A7), 1268.Find this resource:

Tsurutani, B. T., Judge, D. L., Guarnieri, F. L., Gangopadhyay, P., Jones, A. R., Nuttall, J., . . . Zambon, G. A. (2005). The October 28, 2003 extreme EUV solar flare and resultant extreme ionospheric effects: Comparison to other Halloween events and the Bastille Day event. Geophysical Research Letters, 32(3), L03S09.Find this resource:

Uberoi, C. (2011). Geomagnetic storms over India: A close look at historic space weather at Mumbai. Space Weather, 9(8), S08005.Find this resource:

US–Canada Power System Outage Task Force. (2004). Final report on the August 14, 2003 blackout in the United States and Canada: Causes and recommendations. US–Canada Power System Outage Task Force.Find this resource:

Vaquero, J. M., Valente, M. A., Trigo, R. M., Ribeiro, P., & Gallego, M. C. (2008). The 1870 space weather event: Geomagnetic and auroral records. Journal of Geophysical Research: Space Physics, 113(A8), A08230.Find this resource:

Webb, D. F., & Howard, T. A. (2012). Coronal mass ejections: Observations. Living Reviews in Solar Physics, 9, 3.Find this resource:

Wei, L. H., Homeier, N., & Gannon, J. L. (2013). Surface electric fields for North America during historical geomagnetic storms. Space Weather, 11(8), 451–462.Find this resource:

Wik, M., Pirjola, R., Lundstedt, H., Viljanen, A., Wintoft, P., & Pulkkinen, A. (2009). Space weather events in July 1982 and October 2003 and the effects of geomagnetically induced currents on Swedish technical systems. Annals of Geophysics, 27(4), 1775–1787.Find this resource:

Willis, D. M., Armstrong, G. M., Ault, C. E., & Stephenson, R. F. (2005). Identification of possible intense historical geomagnetic storms using combined sunspot and auroral observations from East Asia. Annales Geophysicae, 23, 945–971.Find this resource: