Solar Storm Could Cause $330 Billion Loss by University of Cambridge & AIG
Helios Solar Storm – Stress Test Scenario – Executive Summary
The study of solar eruptive phenomena has progressed over the centuries from scholarly recordings of astronomical events, such as sunspots, to advanced modelling of how solar activity may drive geophysical planetary responses, e.g., geomagnetic disturbances. However, there is still a great deal of uncertainty around the potential economic impacts of extreme space weather on modern society.
In this report, we provide a catastrophe scenario for a US-wide power system collapse that is caused by an extreme space weather event affecting Earth: the Helios Solar Storm Scenario.
This scenario is a stress test for managers and policymakers. Stress tests are important for understanding risk exposure across a spectrum of extreme systemic shocks such as those proposed in the Cambridge Taxonomy of Threats, which encompasses a dozen major classes of catastrophes. A suite of scenarios can be used as a basis for calibrating an organisation’s inherent risk, vulnerability and resilience.
Helios Solar Storm Scenario
This scenario describes how an extreme space weather event can cause direct damage and indirect debilitation of high voltage transmission grids in the USA, resulting in power blackouts along with consequential insurance claims and economic losses. Over the past decade, there have been a number of analyses of the potential effects of extreme space weather on the electricity transmission network. This report adds to this literature by providing a transparent economic analysis of the potential costs associated with such an event.
We estimate a range of US insurance industry losses resulting from three variants of the scenario which explore different damage distributions and restoration periods, culminating in losses between $55.0 and $333.7 billion. At the low end, this is roughly double the insurance payouts of either Hurricane Katrina or Superstorm Sandy, and similar to the total insured losses from all catastrophes in 2015.
Overall economic losses are evaluated from two perspectives. First, we estimate global supply chain disruption footprints that stem from suspended business and production activity directly caused by power outages in the US. This perspective provided a detailed industry sector breakdown of potential economic losses but does not account for the dynamic response of the economy. Global supply chain disruptions are conservatively estimated to range from $0.5 to $2.7 trillion across the three scenario variants.
Second we employ a global integrated economic model to estimate losses in global GDP over a five year period relative to a baseline projection – our standard loss metric referred to as GDP@Risk. Importantly, this perspective accounts for post-catastrophe dynamic responses in the global economy, including, for example, changes in monetary policy. For this reason our GDP@Risk estimates are lower than our estimated static losses from supply chain disruptions. The Helios Solar Storm has a global GDP@Risk ranging from $140 to $613 billion across the three scenario variants (representing between 0.15% and 0.7% of global GDP over the projected five year period).
Selection of a space weather scenario as a disruptor of infrastructure
Anomalous behaviour of US telegraph operations in the mid-1800s, especially during the 1859 Carrington Event, brought about the recognition that solar activity can affect human technology.
Although extreme space weather events include a variety of phenomena like Solar Particle Events (SPEs) and bursts of electromagnetic radiation from solar flares, it is very fast Carrington-sized Coronal Mass Ejections (CMEs) which are mostly associated with the geomagnetic disturbances (GMDs) that are severe enough to cause power grid failure. A severe CME has the potential to generate geomagnetically induced currents (GICs) that could cause permanent damage to Extra High Voltage (EHV) transformers. Such high value assets are not easy to procure and replace in the short-term.
Failure in these critical assets could cause system-wide instability issues leading to cascading failure across the electricity system, passed on to other critical interdependent infrastructures such as transportation, digital communications and our vital public health systems. This disruption could also cause considerable
disruption to business activities.
Our impact analysis is underpinned by some key methodological contributions which include deriving bottom- up, state-level restoration curves, which show how long it takes to restore the supply of electricity after the extreme space weather event, based on key risk factors including geomagnetic latitude and deep-earth ground conductivity.
Disruptions in electricity supply are mapped to state-level industrial output by industrial sector, and are aggregated to the US national level. This yields direct economic loss estimates that are then fed into a global multi-regional economic input-output model to assess domestic and international supply chain disruptions. These estimates are themselves a basis for applying a dynamic economic equilibrium model to gauge how the USA and its trading partners recover from this shock over time.
Variants of the scenario
The Helios Solar Storm Scenario depicts a geomagnetic disturbance that generates GICs capable of damaging or even destroying EHV transformers. Through direct damage and indirect debilitation of the power grid, an extreme space weather event can cause immediate blackouts, leading to insurance payouts and supply chain interruptions. For the purposes of this report we only account for the impact directly on the USA and indirectly on major trading partners.
Beyond appealing to the scientific and industrial literature, the scope of the extreme space weather event and its expected consequences are based on workshops and interviews with subject matter specialists in space physics, economics, catastrophe modelling, actuarial science, and law; insurance specialists in property, casualty and space insurance; and key representatives from utility companies, government agencies, industry bodies, and engineering consultancies. However, the analysis and determination of impacts is our own and does not imply endorsement of these views by the specialists consulted.
This report proposes three scenario variants (S1, S2 & X1) to span the evidence and expert opinion on electrical damage inflicted by extreme space weather. The S1 variant is considered our basic or baseline scenario. It involves limited damage to EHV transformers in the US, with only 5% of those units suffering any damage, and restoration periods of moderate length. S1 represents an optimistic view that a massive geomagnetic storm would cause limited damage due to an initial grid collapse. This would isolate any further damage to the transmission network, allowing the grid to be re-started after the storm passes. The S2 variant assumes greater damage levels but similar restoration times, reflecting uncertainty about how much damage the components of a power grid might be exposed to in extremis. The X1 scenario is deliberately extreme, and reflects the Kappenman (2010) perspective, with similar damage levels to the S2 scenario but with longer restoration periods.2 The scenario is used to explore the upper bound for the economic and insurance loss estimates. Indeed, this is considered by some to reflect an overly pessimistic view of vulnerability of the electricity transmission system to GICs. Opponents of this perspective do not necessarily disagree with the long replacement times for damaged EHV transformers, but instead disagree with the severity of the damage distribution to the assets themselves.
This is a stress test, not a prediction
This report is one of a series of stress test scenarios that have been developed by the University of Cambridge Centre for Risk Studies to explore management processes for dealing with extreme shocks. It does not predict when a catastrophe may unfold. Indeed, it does not also provide definitive economic and insurance loss estimates, as there is still widespread disagreement between different schools of thought. It does however provide insight into the range of exposure that may be experienced based on different expert opinions of extreme space weather events.
An extreme space weather event
Spots on the surface of the Sun
A cluster of sun spots produces a relatively moderate CME, leading heliophysicists at the U.S. National Oceanic and Atmospheric Administration (NOAA) to predict a moderately sized geomagnetic storm in four days’ time. Three days later, a second very large CME is thrown outwards towards Earth, accompanied by a massive solar flare that produces a radiation storm. This CME reaches a very high speed that is sustained in its path towards Earth due to the previous CME which has lowered the ambient solar wind density.
A Carrington-sized CME slams Earth
The front of the first and moderate CME is preceded by a 30-60 minute warning from space weather satellites. The second CME arrives on the heels of the first but travelling much faster and carrying much higher levels of energy. As the first CME is only moderate in size it is not deemed a major risk and utility operators do not implement their emergency plans.
By the time the second CME is detected there is not enough time to fully implement all mitigation measures by the time it arrives at Earth. Satellites suffer significant damage from solar radiation. On Earth many EHV transformers are affected from the geomagnetic storm, particularly those in high geomagnetic latitudes. Some transformers are damaged or even destroyed, while others are tripped off by the network operators. Transformers are at greatest risk at locations with a high geomagnetic latitude and where there is a highly resistive deepearth ground conductivity structure.
Restoration in the aftermath
The ability to restore power depends on availability of skilled engineers to assess and either re-set, commission repairs, or replace damaged units. Replacing severely damaged transformers can be slow – over a period of months – if many new units are required because of the bespoke nature of large capacity high voltage units and the limited stocks of such units. In the S1 scenario most of the affected population has power restored within a few hours while 15% of those affected remain without power for up to three months or longer. In the X1 scenario, more people are affected and some 10% of those lack power for a ten-month period.
Summary of impacts
Direct economic losses
The damage distributions resulting from the extreme space weather event in the S1 scenario show a quarter of US EHV transformers are tripped off-line with only 5% suffering any form of damage. The loss of these assets leads to a power outage initially affecting 90 million US citizens. The majority of those affected have power restored relatively quickly, with only 5% of the total US population being disconnected for more than three days. The US states most directly affected are Illinois and New York with direct losses from suspended economic activity of roughly $30 billion each in S1.
In both the S2 and X1 scenarios 33% of transformers are tripped off-line, 14% sustain minor damage, 3% sustain major damage and 0.2% are completely destroyed. The loss of these assets leads to a power outage initially affecting 145 million US citizens. The difference in these scenarios is the time it takes for power to be restored. In the X1 scenario, with a significantly longer duration, 15% of the total US population remain disconnected for more than three days. Illinois and New York see direct economic losses of roughly $170 billion and $150 billion respectively in X1.
The total direct shock to value-added activities in the US economy as a result of power failure amounts to $220 billion for S1, $700 billion for S2 and $1.2 trillion for X1, corresponding to 1.4%, 4.6% and 8.1% of US GDP, respectively.
US and international supply chain impacts
The total indirect US supply chain shock is similar size to the direct shock. International supply chain shocks, stemming both upstream via US imports and downstream via US exports, are estimated to be roughly a quarter the size of the overall direct US shock. With overall supply chain disruptions estimated to be as high as $470 billion for S1, $1.5 trillion in S2 and $2.7 trillion in X1 (representing economic losses of 0.7%, 2.2% and 3.9% of global GDP, respectively), the economic impact of these scenario variants is likely to be very significant, potentially leading to major policy interventions such as interest rate adjustments and short-term stimulus measures.
At the industry sector level, US Manufacturing, with the largest gross value added (GVA) of $1.9 trillion, has both the greatest direct ($30 billion in S1 to $170 billion in X1) and indirect ($30 billion in S1 to $180 billion in X1) shocks, with indirect shocks having a roughly equal split between those that have been induced upstream and those induced downstream. China, Canada and Mexico, as the three largest trade partners of the US, collectively account for about a third of all indirect international supply chain impacts, ranging from $20 billion in S1 to $100 billion in X1.
GDP Losses over five years
The scenario variants characterise different shares of the US population experiencing power outage for different durations. We translate these restoration curves into more specific shocks in terms of private and government consumption, productivity (via hours worked), investment, exports, imports and confidence. These variable- specific shocks then become the basis for shocking the overall US economy, within the Global Economic Model (GEM) of Oxford Economics. The integrated economic model solves to find a state of equilibrium with associated deviations in US and global GDP from baseline projections.
In S1, the total estimated US GDP@Risk is $135 billion, or 0.15% of the five-year baseline GDP projection for the US economy, whereas the global GDP@Risk is $140 billion (0.03% of the global GDP projection). In X1, US GDP@Risk is $610 billion (0.7% of the US GDP projection), whereas the global GDP@Risk of $1.1 trillion is roughly a quarter of the five-year baseline global GDP projection.
We estimate the range of US insurance industry losses for the scenarios described in this report to be between $55.0 billion and $333.7 billion. Just over 90% of this loss is from service interruption within property insurance policies for those that loss power. While only 1% is from direct physical property damage. Other insurance lines such as Space, Directors and Officers, Homeowners and Speciality contribute to total insured losses.
By comparison, Swiss Re studies estimate insurances losses from Hurricane Katrina and Superstorm Sandy as $45 and $35 billion, respectively (Swiss Re, 2006; Swiss Re 2013), and total insured losses from catastrophes in 2015 as $85 billion (Swiss Re, 2015). Past events led to relatively short outages that would be within the waiting periods of most property insurance policies. However, the event contemplated in this report assumes a much longer outage that would extend beyond the typical waiting period.
This report contributes to the understanding of the range of economic and insurance impacts of extreme space weather, and makes a number of key methodological contributions relevant for future economic impact assessment. Understanding the economic impact of space weather risks can improve mitigation procedures and practices, and guide where limited resources should be allocated to improve economic resilience. Moreover, in industry it is not just electricity utility companies who are concerned with catastrophe scenarios; the potential losses to consumers of power and to insurance companies due to casualty and business interruption pay-outs are significant.
Narrowing the range of geomagnetic effects following a given solar event, specifically the systemic effect on the electricity transmission network and its key assets, is necessary. Moreover, ongoing work is needed to implement mitigation provisions and response plans which include the rollout of temporary generation facilities and short-term portable replacement transformers. This is consistent with a call by the US National Space Weather Action Plan (National Science and Technology Council, 2015) for improved assessment, modelling, and prediction of the impact of this threat on critical infrastructure systems. Given the uncertainty that exists in this area of research, what is required in the wake of this report is for space physicists, geophysicist and electrical engineers to work collaboratively on improving the estimation of the economic impacts of extreme space weather.
Introduction to Space Weather
What is space weather?
Space weather can be defined as disturbances of the upper atmosphere and near-Earth space that can disrupt a wide range of technological systems (Hapgood et al. 2012). It can arise from many different types of eruptive phenomena associated with solar activity taking place on the surface of the sun (often referred to as a ‘solar storm’). On average, the Sun’s magnetic activity follows an 11 year solar cycle, with variable minimum and maximum sunspot periods. Solar cycle 24 began in 2008 with minimal sunspot activity until 2010. We are now in the declining phase of the solar cycle where intense activity has previously been more prevalent than other periods (Juusola et al. 2015).
The strength and complexity of the Sun’s evolving global magnetic field changes throughout the solar cycle, manifesting as regions of concentrated magnetic field in the photosphere known as sunspots. Through this cycle, the magnetic field in the solar atmosphere alters from a magnetically simple state to a magnetically complex configuration, leading to an increasing number of sunspots (Green and Baker, 2015). While there may be more solar activity during some parts of the solar cycle, solar eruptive phenomena are still the result of a random process. Therefore, there is the potential for this to cause an extreme space weather event affecting Earth at any time.
There are three primary forms of solar activity which drive extreme space weather, as shown in Figure 1.
- Coronal Mass Ejections (CMEs) – CMEs are massive explosions of billions of tonnes of charged particles and magnetic field thrown out into space (Webb and Howard, 2012).
- Solar Proton Events (SPEs) – SPEs are a huge increase in energetic particles, mainly of protons but also heavy ions, thrown out into space (Shea and Smart, 2012). They may be related to CMEs and solar flares.
- Solar flares – Solar flares are 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).
Extreme space weather results from these eruptive solar phenomena. When occurring in combination, Earth may first be bombarded with initial radiation (such as X-rays) from a solar flare from eight minutes after the event on the surface of the sun.
A second barrage of very-high-energy protons (an SPE) may then arrive anywhere between tens-of-minutes later, and may last for days. The SPE may be followed by a large CME reaching Earth some 14.5 hours or more later. The magnetic field in the CME can cause an extreme geomagnetic storm that can also last for days. The storm drives huge electrical current at high latitudes and bright auroral displays.
CMEs are a key aspect of coronal and interplanetary dynamics, and, as they are associated with the vast majority of solar eruptive phenomena (Webb and Howard, 2012). Moreover, CMEs pose the main risk to Earth and its modern, technological society because large (1012kg), dense (100/cm3) and fast (>500kms-1) CMEs hitting Earth with a southward interplanetary magnetic field direction (Bz) can give rise to extreme geomagnetic disturbances (Möstl 2015; Temmer and Nitta, 2015; Balan et al. 2014). This has the potential to damage key electricity network assets and disrupt the aviation, satellites and GPS on which our economy and society depend. This is particularly problematic because failure in the power sector can cascade to other critical interdependent infrastructure systems, disrupting business activities and inducing a range of other economic and social consequences which can affect the global economy (Ouyang, 2014; Anderson et al. 2007; Haimes and Jiang, 2001; Rinaldi et al. 2001).
See full PDF below.