House of Commons - Developing Threats: Electro-Magnetic Pulses (EMP)

Developing Threats: Electro-Magnetic Pulses (EMP) - Defence Committee Contents

2 Nature of the threat

Space weather

13. "Space weather" is a naturally occurring phenomenon that can impact upon the Earth's environment in ways that are detrimental to key technologies in operation in space, the atmosphere and the surface of the Earth. "Space weather" generally refers particularly to changes in the space environment near Earth, caused by varying conditions in the sun's atmosphere. Solar activity adheres roughly to an eleven year cycle, with solar activity increasing during a "solar maximum", making space weather events more likely. The next solar maximum is predicted to occur in 2013. However, space weather events do not necessarily obey this cycle; the Carrington event of 1859 (see below) occurred in the middle of a cycle. Space weather events are an everyday occurrence. Indeed, the well-known phenomenon of the aurora borealis, or "Northern Lights", is an effect of charged particles originating from the sun colliding with the Earth's atmosphere.

14. The following table summarises some of the different types of space weather and their potential impacts.

Table 1: Categories of Space Weather

Space Weather Cause Potential impact

Coronal Mass Ejections (CMEs) Plasma ejected violently from the outer atmosphere of the sun Fluctuations in the Earth's magnetic field (geomagnetic storms), driving additional current into power grids, disrupting satellites, GPS (global positioning system) and radars

Solar Energetic Particle (SEP) events High energy particles expelled from sun during solar events like CMEs Damage to electronics, computer chips and power systems in spacecraft [and aircraft] (possibly at ground level too), raised ground radiation levels

Solar radio bursts Intense bursts of radio noise produced by solar events like CMEs Interference with low power wireless radio technologies such as mobile phones, wireless internet and GPS receivers

Solar flares Outburst of radiation and energetic particles Modest effects on Earth

Data Source: Science and Technology Committee, Scientific Advice and Evidence in Emergencies, para 18

15. The largest ever recorded space weather event occurred in September 1859. It is known as the Carrington Event after Richard Carrington, the British astronomer who observed a solar flare so strong that it could be seen with the naked eye. The huge coronal mass ejection (CME) that followed induced enormous electric currents that surged through telegraph systems, causing shocks to telegraph operators and setting fire to papers.[6] Operators were able to disconnect their batteries and continue to send messages using only this induced current. The impact was so wide-ranging that auroras, normally limited to polar regions, were observed as far South as Hawaii and the Caribbean.[7]

16. The Carrington event is thought to have been up to ten times larger (depending on what effect is being measured) than anything seen in the past 50 years.[8] There have been less significant, but still destructive, events in the more recent past. In March 1989 a large CME caused the Canadian province of Quebec's power grid to collapse within 90 seconds, after stabilising equipment failed to cope with the effects of the geomagnetic storm. Around six million people were subsequently without power for nine hours. The same storm caused a transformer to fail in New Jersey and various other effects across the North American grid. It is also thought to have been the cause of damage to two transformers in the UK. The adverse effects of extreme space weather on modern technology are therefore well documented.

17. Resilience to space weather events is routinely built into some components of infrastructure, such as satellites, which are frequently exposed to its effects. However, events vary in intensity, and the potential impact of a severe event could be devastating.

18. Space weather events have the potential adversely to affect human health in some cases. Exposure to even very high levels of electromagnetic activity is not thought to have any direct ill-effects on humans or any other living organisms. However, there could be potential implications for patients with implantable cardiac devices, as highlighted to the Committee by the Royal College of Physicians.[9] Solar energetic particle events may lead to increased exposure to radiation for workers such as pilots and flight attendants on long-haul flights.

THE PROBABILITY

19. The Government told us that the likelihood of a severe space weather event (not necessarily, of course, a Carrington-magnitude event) over the next five years was assessed as being moderate to high, with the potential to cause damage to electrically conducting systems such as power grids, pipelines and signalling circuits.[10]

POTENTIAL IMPACT ON ELECTRONIC INFRASTRUCTURE

20. The maximum possible severity of a space weather event is, of course, impossible to estimate, but witnesses use the Carrington event as a reasonable worst case possibility.[11] Since 1859 the reliance of the world on electrical power has increased enormously. As a result, as the Met Office put it, "the potential effects of space weather are growing rapidly in proportion to our dependence on technology".[12] The impact of an event on the scale of the Carrington Event occurring today would be huge; the US National Research Council estimated the wider societal and economic costs of a severe geomagnetic storm occurring today to be around $1-2 trillion.[13]

21. The cause of the damage would be geomagnetically induced currents (GICs). These are electric currents driven by electric fields that are induced in the surface layers of the Earth crust by rapid changes in the geomagnetic field (such as those occurring during magnetic storms). These currents are most evident in long metal structures, such as power grids, pipelines and railway circuits, with earth connections to the surface layers, so that currents can flow between the Earth and these structures. Research Councils UK highlighted in particular the risk of GICs causing damage to electrical grid transformers:

GICs pose a threat to electricity distribution grids extending over long distances which can cause blackouts and damage. Permanent damage to transformers caused by GIC is a major concern. Transformers are costly, not available as "off-the-shelf" items, and replacing one is a major exercise. The consequences of a prolonged loss of electrical power are potentially catastrophic as the infrastructures and services that modern developed societies rely on are entirely dependent on electricity. Examples include heating, lighting, refrigeration, communications, pumping of fuel, water and sewage.[14]

22. The US National Academy of Sciences has estimated that if a magnetic storm that occurred in May 1921 was repeated today then 130 million people in the US would lose their electricity and more than 350 transformers would be at risk of permanent damage.[15] Avi Schnurr, Chair of the EIS Council added that the Federal Energy Regulatory Commission (FERC) had estimated that the duration of the impact would be five to 10 years.[16] He thought it would be fair to say that the conclusions about the UK would not be better.

23. While experience of, and forecasts about the likely impact of, severe space weather on the USA, are relevant, it cannot automatically be assumed that the effect of the UK would be the same. FERC wrote "while the threats posed by EMP and the vulnerabilities of electrical infrastructure to EMP are not unique to the US, differences between the US and UK power grids should be considered when reviewing the applicability of these responses to the UK".[17] Chris Train of National Grid thought that the consequences in the UK would be less because the UK infrastructure was differently formulated:

We have been working along with our partners—the BGS and Manchester University and others—looking at what the potential impacts would be here in the UK.

Because of the meshed infrastructure here in the UK, we believe that the impact would not be as great here. The effort that we have been putting in has been around operation mitigations following a coronal mass ejection to understand how we might manage the system to minimise the impact of the potential in such an event. But I do not think that it would have the same catastrophic cascading effect that would happen in the United States because of the different nature of the configuration and development of the networks.[18]

24. In response to Chris Train, Avi Schnurr made clear that he thought that the National Grid assessment was, for a variety of reasons, too optimistic.[19] Both witnesses did, however, agree that there was more work to be done on the modelling.[20]

25. National Grid, while noting that contingency plans were in place to react to damage, admitted that in the event of an extremely severe storm, long-term blackouts could be a possibility:

If all transformers at a node are damaged then, depending on the location of the node within the network, this could result in a local area being disconnected until replacement transformers could be installed. Replacing a transformer can take two or more months depending on the availability and location of spares. In this extreme event scenario National Grid estimates that the probability there would be a disconnection event is 62% for England and Wales and 91% for GB as a whole.[21]

26. If, as might be expected, several countries were affected simultaneously and needed new transformers, this would, presumably, affect the availability of spares and hence extend the delay in restoring power supplies.

27. As noted by several commentators, including Research Councils UK, modern developments in technology have actually led to greater vulnerability to the effects of space weather, as microchips become smaller and more advanced. This is a particular hazard for satellites which are exposed to the greatest effects of space weather.

28. The risks posed by space weather are known and significant, though there is argument about the likely extent of their impact: a severe event could potentially have serious impacts upon UK infrastructure and society more widely. It is essential that this hazard is sufficiently recognised and addressed by the Government and relevant civil bodies.

29. We recommend that work proceed as a matter of urgency to identify how seriously a future Carrington event would affect the UK infrastructure. It is clear that more modelling is required to establish the likely effect of a major space weather event on the National Grid. This should be independently validated and compared with the results of observations of Grid behaviour during space weather events.

High Altitude Nuclear EMP Weapons (HEMP)

30. Space weather events have always been with us, though their likely effects have become more significant as technology advances. A newer threat, though one which has been the subject, until recently, of more research, is the possibility of deliberate attack, perhaps with little or no warning. A single nuclear weapon detonated between 25-500 miles above the Earth could create an electromagnetic pulse (EMP) - i.e. a high density electrical field-- with the potential to cause severe damage to technology over a wide geographical area, the area depending on the height of the detonation.

31. The effect of such a high-altitude nuclear EMP (HEMP) burst would not be identical to those of severe space weather because it would create a series of electromagnetic "waveforms" that each has a slightly different effect on Earth. The overall effect, however, would be similar in nature to the effects of naturally occurring space weather, but faster and more intense, which makes HEMP potentially highly destructive.

THE EMP COMPONENTS

"E1" or "Fast" component occurs within a few billionths of a second of detonation. It produces a very brief but intense electromagnetic field that can induce very high voltages in electrical conductors. Unlike naturally occurring geomagnetic storms, which pose the greatest risk to long electrical conductors like power line transformers, this effect has the power to disrupt or damage micro-electronic systems, electronics-based control systems, sensors, communication systems, protective systems, computers and similar devices. Damage and disruption could occur almost simultaneously over a very large area.[22]
"E2" covers roughly the same geographic area as the first component and is similar to lightning in its effect, though far more geographically widespread and somewhat lower in amplitude. In itself this component would not be an issue for critical infrastructure systems since they have existing protective measures for defence against lightning strikes. The most significant risk derives from the fact that this component follows a small fraction of a second after the first (E1) component, which may have already impaired or destroyed protective and control features. The energy associated with the second component therefore may be allowed to pass into and damage systems.
"E3", or "slow" component is a slower-rising, longer-duration pulse that creates disruptive currents in long electricity transmission lines, similar in effect to that of a severe geomagnetic storm.

32. The sequence of E1, E2, and E3 components of EMP is important because each can cause damage which can allow subsequent components to cause greater damage than they might independently. The combined effect of a nuclear EMP is therefore very difficult to mitigate,[23] and witnesses agreed that such an event would be "a different kettle of fish" from a Carrington-sized event, and a "truly catastrophic event".[24]

PRACTICAL EXPERIENCE

33. In July 1962, the United States conducted an experiment with the detonation of a 1.4 megaton nuclear weapon (Starfish Prime) 250 miles above the Pacific Ocean, around 900 miles from Hawaii. The effect of this explosion was the generation of an electromagnetic pulse that was far larger than expected. The EMP caused damage to electrical equipment in Hawaii, knocking out streetlights, setting off fire alarms and damaging telephone equipment. There were also visible auroras in the sky following the detonation. In 1962, the Soviet Union also performed a series of three high-altitude nuclear tests in space over Kazakhstan. The weapons used were smaller than that of the US's Starfish Prime test, but the EMP effects were reportedly more significant, as the detonations occurred over a populated area. While there is little information available about these tests at least in the public domain, what there is suggests . that there was significant damage to telephone wires and power cables.

POTENTIAL IMPACT ON ELECTRONIC INFRASTRUCTURE

34. An EMP Commission was established in the United States in 2001 to look expressly at the potential impact of a high altitude EMP attack on key US infrastructure. It published a preliminary report in 2004 which established the nature of the threat being faced, followed by a final report in 2008 which went into much greater detail as to the potential impact on critical national infrastructures and recommended courses of action to address the threat. Its overall conclusions were that US society is vulnerable to an EMP attack, the consequences of which might be long-term, widespread and catastrophic, and because of the interdependency of the systems which are likely to be affected, the current recovery plans may be of little value.[25]

35. National Grid's written evidence reiterated the concerns of the EMP Commission:

The effect of E1 and E3 pulses from HEMP would be considerably more extreme [than space weather events]. For these effects we have no practical experience to fall back on, although the Commission to Assess the Threat to the United States from EMP Attack did conduct a number of experiments on E1 and its effect on SCADA. They concluded that "Large-Scale load losses in excess of 10% are likely at EMP threat levels" and that "widespread collapse of the electrical power system [...] is virtually inevitable.[26]

Risk?

36. While the probability of a HEMP attack is judged by the Government to be low, it is accepted that its impact would be high, and in view of this the Government includes it in the second tier of priorities for UK national security.

37. The Government does not differentiate in its risk assessment between HEMP and other elements of a nuclear attack:

The National Security Strategy and National Security Risk Assessment assessed the risk from a nuclear attack and thus HEMP, as part of the risk of an attack on the UK by another state or proxy using CBRN (chemical, biological, radiological and nuclear) weapons. This risk was judged to be low likelihood but in view of the impact, to be considered in the second tier of priorities for UK National Security.[27]

It views the threat as potential rather than actual.

Currently no state has both the intent to threaten our vital interests and the capability to do so with nuclear weapons. MOD's view is that over the next decade, existing space launch vehicle technology could theoretically be adapted by states to deliver a nuclear device; however, the MOD does not currently see the UK or Western Europe as a target for such an EMP attack. MOD does not believe that any non-state actors can currently produce improvised nuclear devices and none are likely to be able to make a sufficiently robust warhead for missile delivery in the foreseeable future.

And

To generate more widespread damage from EMP, a nuclear warhead would have to be detonated at high altitude to generate the EMP from the interaction between the radiation from the weapon and the outer layers of the atmosphere. This could only be achieved by launching a device by missile to an altitude of several tens of kilometres. A limited number of States possess this capability.[28]

38. The US EMP Commission found that some "rogue states", including Iran and North Korea, are aware of the potential of such an attack. Iran, in particular, is reported to have been conducting what appear to be missile tests to simulate a nuclear EMP strike.[29] According to US Senate testimony, an Iranian military journal publicly discussed using EMP against the West:[30]

Once you confuse the enemy communication network […] you will, in effect, disrupt all the affairs of that country. If the world's industrial countries fail to devise effective ways to defend themselves against dangerous electronic assaults then they will disintegrate within a few years.[31]

39. We asked Avi Schnurr if Iran had the capability to launch a weapon to an altitude of several tens of kilometres. He replied that there was a serious risk that rogue states and nations, some of which already possessed the launch capacity, would acquire the necessary weapons:

There are only two things that stand right now between us—by "us" I mean the United Kingdom, the United States and other allies—and having this level of catastrophe. One is that although they have the ships and the missiles, terrorists groups and rogue nations do not today, necessarily, have access to nuclear weapons. That is a very thin boundary, because, for example, North Korea has nuclear weapons. Could North Korea sell its nuclear weapons? Could there be destabilisation in a state that has nuclear weapons? I could name a few. Those are things that could happen in the future. The other thin boundary is will. We are dependent on keeping any warhead of any size out of the hands of transnational terrorists or rogue nations and on their good will if they acquire them.[32]

40. We asked the Ministry of Defence if Iran could launch a missile from a ship. Following a rather confusing exchange,[33] we asked for written clarification. They explained:

A number of elements are required to enable a state or non-state actor to successfully launch a nuclear EMP attack:

A delivery system capable of adequate range and altitude, with the capacity to carry a significant payload. A ballistic missile is, therefore, the most likely delivery system and, given the weight of a HEMP device, it must be capable of carrying a payload significantly heavier than a high explosive warhead.

A nuclear device is also required to deliver a HEMP. Successful uranium enrichment and sophisticated weapons engineering are required to manufacture a viable nuclear device. To be delivered at high altitude to generate a HEMP, the nuclear device must also be ruggedised sufficiently to withstand: the harsh conditions of launch; the high velocity journey through the atmosphere and into space; and, perhaps, depending on where on the flight path the nuclear device is detonated, a period of re-entry.

As well as manufacturing a robust nuclear device, it must then be successfully integrated into the ballistic missile to create a weapon system.

The development of all these elements is technically very challenging and expensive, with progress likely to be made in small incremental steps over a period of many years, and we judge this to be within the grasp of only a limited number of state actors.[34]

41. Consistent with its view of the threat of an HEMP attack as potential rather than actual, the policy of the Government is to try to ensure that no attack occurs. David Ferbrache said:

We are very much focused on trying to ensure that that event does not occur in the first place, which is all about counter-proliferation action to prevent the acquisition of nuclear weapons or ballistic missile capabilities. Deterrent capability is one of the areas that we make absolutely certain is protected against EMP, in terms of our ability then to retaliate against such an aggressive act.[35]

42. On the basis of the evidence received, it seems likely that at present only those states with a known nuclear capability would be able to utilise an HEMP weapon. However, certain states such as Iran could potentially pose a realistic threat in the future, even if it does not currently do so, if nuclear non-proliferation efforts are not successful. Non-state actors could also pose a threat. While the risk may at present be low, the potential impact of such a weapon could be devastating and long-lasting for UK infrastructure. The Government cannot therefore be complacent about this threat and must keep its assessment of the risk under review. It is therefore vitally important that the work of hardening UK infrastructure is begun now and carried out as a matter of urgency.

Non-Nuclear EMP

43. It is also possible to build non-nuclear devices which can disrupt electronic systems, though so far only over a limited area. The Chair of the US EMP Commission wrote:

Non-nuclear EMP weapons, like radiofrequency weapons, can damage and destroy electronics locally. Such weapons have short ranges, kilometers for some military systems to meters for devices improvised by terrorists or criminals. Industrial EMP simulators, intended to test commercial systems for hardness against interference from stray electronic and radio emissions, are on the open market and can be purchased by anyone. At least one such EMP simulator is designed to look like a suitcase, can be operated by an individual, and is powerful enough to damage or destroy the electronic controls that regulate the operation of transformers and other components of the power grid. Armed with such a device, and with some knowledge about the electric grid, a terrorist or lunatic could blackout a city.[36]

44. Avi Schnurr said:

The biggest issue with non-nuclear EMP weapons is that the complexity and threshold required to produce them is minimal, to say the most. At the summit meeting in Washington DC, for example, there were two Assistant Secretaries of Defence, a Deputy Under-Secretary and the Pentagon's chief lawyer, all of whom expressed grave concerns over this risk—the non-nuclear EMP risk in particular, but the risk of EMP in general. The non-nuclear EMP risk is much shorter-range. However, that range, which could be 100 metres, a fraction of a kilometre or a kilometre—under certain circumstances, which I could discuss separately, it could be multiple kilometres—includes the risk of having a field strength that would be even greater, although limited in extent, than a nuclear EMP [...]. We had a speaker at that summit who described, to the extent he was allowed to describe it, a device that he built from hardware he acquired from retail stores in the United States, which he had built into a van.[37]

45. A number of nations are thought to be undertaking research into the development of non-nuclear EMP attack weapons, but the Government does not currently regard them as a serious risk.[38] Nick Harvey said "it is certainly considered a potential threat. It is not considered a particularly likely one, certainly in the foreseeable future; but we keep that constantly under review. It is a material risk that we need to consider, but we do not think there is any imminent likelihood or threat from it".[39]

46. While in the UK material relating to non-nuclear EMP is highly classified, other nations, particularly in the EU, are apparently much more open about devices in existence and in development. We asked the MoD why the UK kept information about non-nuclear EMP under a higher security classification than did other countries, and whether this affected its ability to share best practice with allies. The response was:

We collaborate with our allies on non-nuclear EMP effects, including research and development into countermeasures, through the NATO research and technology organisation which has a working group looking at those issues—so that is quite a close linkage.

In terms of classification, there is quite a bit of material on the internet. We routinely monitor that and assess it. Some of the devices are potentially viable; some are not. Most of them are rather short-range; for instance, with modified microwave sources, you are talking about ranges in the category of hundreds of metres. We keep an eye on those threats. Is it classified? There are some classified areas. We do not want to share our view on what viable devices might be at the high end of non-nuclear EMP, so we protect that very sensitive area, because we do not wish to see further proliferation of those competent devices. That is the classification reason.[40]

47. While existing non-nuclear EMP devices may be crude and limited, the fact that viable devices could be produced by non-state actors is a cause for concern. Even localised damage could have the potential to disrupt activity, especially if combined with other forms of attack.

6 "A Super Solar Flare", NASA Science News, 6 May 2008, science.nasa.gov/science-news/science-at-nasa Back

7 Severe Space Weather Events - Understanding Societal and Economic Impacts Back

8 Ev 22 Back

9 Ev 46 Back

10 Ev 30 Back

11 Q 17 Back

12 Ev 50 Back

13 Severe Space Weather Events - Understanding Societal and Economic Impacts, p 4 Back

14 Ev 25-26 Back

15 Severe Space Weather Events - Understanding Societal and Economic Impacts, p 3 Back

16 Q 3 Back

17 Ev 48 Back