6.3.1 Kinetic Impactors

A kinetic impactor works by transferring momentum from the spacecraft’s motion to the target asteroid through a collision, which also causes secondary momentum ‘kicks’ through debris ejection during crater formation. Momentum is a conserved property in physics that depends on an object’s mass and velocity. Like velocity, momentum is a vector quantity, meaning it has a magnitude (‘how much’) and a direction.Footnote ³⁰ The overall effect of the kinetic impactor will depend on the total amount of momentum change imparted onto the asteroid and the direction in which the momentum change is applied relative to the asteroid’s current motion.

To develop this idea further, first consider the effect of the spacecraft’s collision alone. For simplicity, we only consider the case of a spacecraft hitting the asteroid directly along or against the asteroid’s instantaneous direction of motion (i.e. its track).

Modelling the collision such that the spacecraft’s momentum is perfectly absorbed by the asteroid, we can write the change in the asteroid’s velocity immediately after the impact as

$\mathit{\Delta v}=v_r\frac{m_{\mathit{sc}}}{m_{\mathit{\text{astr}}}}$

for asteroid mass $m_{\mathit{\text{astr}}},$ spacecraft mass $m_{\mathit{sc}},$ and spacecraft velocity relative to the asteroid $v_r.$ The direction of the change is governed by the direction of $v_r,$ with the against motion being represented by a negative value for the spacecraft’s relative velocity.

The spacecraft’s mass will be much smaller than the asteroid’s mass, making very high-velocity impacts necessary for even modest deflections. Fortunately, as we saw above, even small changes can have a large effect when given enough time. Moreover, the momentum imparted onto the asteroid is larger than that given by the spacecraft alone. This is because the high speeds of the collision will result in a cratering event, causing the ejection of asteroid material during crater formation. The mass loss due to cratering provides a rocket-reaction-like ‘kick’ on the asteroid itself. Assuming the ejecta is released predominantly in the direction opposite to that taken by the incoming spacecraft, the crater ejecta enhances the overall $\mathit{\Delta v}.$ However, it is also possible that waves will propagate through the asteroid and cause mass to be lost on the side opposite the impact, which would work against the desired momentum change and reduce the overall $\mathit{\Delta v}.$

Cratering effects in the kinetic impact method are taken into account using a parameterised approach, modifying the equation above by including an additional factor:

$\mathit{\Delta v}=\beta v_r\frac{m_{\mathit{sc}}}{m_{\mathit{\text{astr}}}},$

where $\beta$ (pronounced ‘beta’) is greater than unity if cratering enhances the kinetic impact method and less than unity if it works against it. Simulations show that $\beta >1$ are very feasible,Footnote ³¹ but this depends on the type of asteroid and exactly how the shock waves propagate through the asteroid. To truly know, we need to perform tests.

NASA recently sent a spacecraft to an asteroid to do just that. The Double Asteroid Redirection Test (DART) mission targetted Didymos, a binary asteroid system consisting of a large asteroid accompanied by a ‘moonlet’: a smaller asteroid that orbits the larger one.Footnote ³² This moonlet, now officially designated Dimorphos (but sometimes called ‘Didymoon’) is about 170 metres in diameter, a size scale that would pose a considerable threat if a comparable object were to hit Earth. (There is, it is important to note, no current threat to Earth from either Didymos or Dimorphos).

We can assume that Dimorphos has a mass of about 5 billion kilograms.Footnote ³³ The DART spacecraft mass was 500 kilograms and was planned to collide with Dimorphos at about 6.6 kilometres per second. To get a feel for the numbers, if we assume that $\beta =2$, then the expected $\mathit{\Delta v}$ would be approximately one millimetre per second. Why use a double asteroid for the test? Didymos has a very well-characterised light curve (variation of brightness over time) with easily discernible variation due to the passage of Dimorphos across and behind Didymos, as seen from Earth. A change in the period of Dimorphos due to DART was, indeed, quite noticeable using observations from the ground. In contrast, measuring the velocity change for a single asteroid orbiting the Sun would have been extremely difficult, at least without precise ranging equipment or years of meticulous observations.Footnote ³⁴

The kinetic impactor method for asteroid redirection has some clear advantages: the technology is relatively simple and the legal issues surrounding its implementation are (as we will see) largely uncontroversial. The downside is that the way the kinetic impactor strikes the asteroid matters significantly. As discussed in the previous section, the impact is most likely to be designed to be along or against the track of the asteroid. However, due to constraints set by the details of orbital dynamics, one of the directions will be much easier to accommodate than the other. This means that the kinetic impactor method has a preferred direction for deflection, a direction that might not be the same as what is needed on the B-plane. Consider Figure 6.5 again. Suppose we know that an asteroid will strike Earth at the first dot below the centre position. The best deflection strategy would accordingly be to perturb the asteroid such that it moves downward on the B-plane, only needing to go a bit more than half of an Earth radius. But say such a deflection is not practical due to the details of the orbits, and instead the asteroid needs to be moved up on the diagram. This would mean that a larger perturbation is required (over 1.5 Earth radii), which in turn requires a more massive spacecraft, a higher impact speed, multiple impactors or a combination of two or more of these. In fact, multiple impactors might be the safest approach in this situation, since using a large $\mathit{\Delta v}$ all at once risks fragmenting the asteroid, which would in turn increase the collisional cross section of the material and could, potentially, cause multiple destructive airbursts when the fragments reach Earth.Footnote ³⁵ Then there is uncertainty in the effective $\beta ,$ which can have a significant effect on the strength and number of impacts required. When looking at a larger motion on the B-plane, all these uncertainties could amount to an unsuccessful deflection. For these reasons – more flexibility and control over the perturbations – we should now consider the use of a nuclear explosive device.

6.3.2 Nuclear Explosive Devices

Nuclear explosive devices (NEDs) deflect an asteroid by vaporising a region of its ‘regolith’ – a layer of unconsolidated rock and dust found on the surface of most asteroids (as well as other celestial bodies such as the Moon). The newly formed vapour, bounded by the asteroid on one side and Space on the other, expands rapidly away from the asteroid’s surface. The result is that the vapour acts just like the exhaust from a rocket, imparting a $\mathit{\Delta v}$ onto the asteroid, with the direction set by the location of the NED (and hence the location of regolith that becomes vaporised).

An NED spacecraft only needs to rendezvous with the asteroid, removing the directional bias inherent in the kinetic impactor method; the NED can be detonated along or against track. In addition, the spacecraft carrying the NED can first study the asteroid to best determine how far away the NED should be detonated, which will control the amount of regolith that is vaporised and thus the resulting $\mathit{\Delta v}$. In 2007, a NASA report prepared for the US Congress concluded, ‘Nuclear standoff explosions are assessed to be 10–100 times more effective than the non-nuclear alternatives analyzed in this study.’Footnote ³⁶

Despite the clear advantages, NEDs have their own challenges. Just as in the kinetic impactor method, too large a single $\mathit{\Delta v}$ could fragment the asteroid; to reduce this risk, multiple, sequential NEDs may be necessary. Moreover, the perturbation will depend on the actual yield of regolith vaporisation, which may not behave as expected. A nuclear explosion close to Earth, such as an attempt to break up a small but still destructive asteroid, could also have unintended consequences, for instance delivering radioactive debris to Earth’s atmosphere and possibly even to the surface.Footnote ³⁷ Perhaps most pressing from an implementation standpoint, NEDs have legal and security implications, as will be discussed below.

Still, there are several reasons why NEDs are widely considered to be an attractive option for asteroid deflection: the necessary technology already exists in the form of nuclear warheads and large Space rockets, they can deliver far more energy than other conceivable methods, and they offer more flexibility in timing. The latter is significant, as perturbations have a maximum effectiveness if applied during certain parts of an asteroid’s orbit; specifically, you get more orbital change for your $\mathit{\Delta v}$ if the ‘kick’ is applied at perihelion (the closest approach of the asteroid to the Sun), where the asteroid is moving the fastest in its orbit. This well-known ‘Oberth effect’ can thus cause larger changes to the asteroid location on the B-plane than an otherwise equivalent mission that perturbs the asteroid away from perihelion.

6.3.3 The Long Game: Mass Drivers and Gravity Tractors

The previous two sections explored ‘impulsive’ asteroid redirection methods, in which the desired $\mathit{\Delta v}$ is achieved more or less instantaneously. With enough lead time, however, a gentler approach could be taken, with a continuous application of small nudges accumulating over time into the desired movement of the asteroid on the B-plane. Although there are several such methods, we focus on two very different approaches to highlight the range of possibilities.

A mass driver is potentially the crudest approach. It involves landing one or more spacecraft on an asteroid and throwing mass in the opposite direction of the desired deflection.Footnote ³⁸ With every throw, the ‘recoil’ due to Newton’s third law (equivalent momentum conservation) imparts a small $\mathit{\delta v}$ (‘little delta v’) to the asteroid. This by itself is insufficient to redirect the asteroid, but after time all the little $\mathit{\delta v}$’s deliver a cumulative $\mathit{\Delta v}$ that is large enough to produce the desired effect.

The asteroid itself is the source of the mass. In the most basic form, you might imagine a mining-like apparatus that is continuously scooping up material and jettisoning it into Space. In a more sophisticated form, the material might be sorted, processed and used in a high-velocity ion engine.

Essentially, a mass driver turns the asteroid into a rocket. Under the assumption of rocket motion only, the change in speed of a rocket after throwing out some mass $\mathit{\Delta M}$ is

$\mathit{\Delta v}=-v_e\mathit{ln}\left(\frac{M_0-\mathit{\Delta M}}{M_0}\right),$

where $M_0$ is the initial mass and $v_e$ is the exhaust speed of the propellant. And here we run into the harsh reality of rocketry: the velocity change is proportional to the logarithm of the mass change. This means we need to either throw out a lot of mass, or have high exhaust speeds, or both. Flipping the equation around, we see that

$\mathit{\Delta M}=M_0\left(1-\mathit{exp}\left(-\mathit{\Delta v}/v_e\right)\right).$

If we want to achieve a $\mathit{\Delta v}$ of approximately one millimetre per second (comparable to the kinetic impactor scenario discussed above), then for $v_e=$ 10 metres per second, two kilometres per second, and 40 kilometres per second, we need to respectively use $\frac{\mathit{\Delta M}}{M_0}=10^{-4},$ $5\times 10^{-7},$ and $2.5\times 10^{-8}$ of the asteroid’s mass as fuel (note this is useful mass – the amount that needs to be processed may be much higher). The different speeds are chosen to highlight those that (1) might be comparable to what will be used to expel waste during mining, (2) reflect conventional rocketry nozzle speeds and (3) are comparable to ion drive exhaust speeds. Although these mass fractions are small, they might be challenging to mine. For an asteroid such as Dimorphos, even the ion drive scenario requires consuming 125 kilograms of useful mass from the asteroid. The conventional rocket scenario requires 2.5 tonnes (again this is useful mass), and the low-speed mining ejection scenario requires 500 tonnes. The last approach literally involves throwing rocks, so while large amounts are needed, no processing is required.

The difficulty of the mass driver method rises quickly both with the size of the asteroid and with the size of the desired $\mathit{\Delta v}.$ It should also be kept in mind that Dimorphos is fairly small. All other things being equal, an asteroid twice as large would require eight times as much mass as fuel to achieve comparable $\mathit{\Delta v}$’s. Achieving a higher $\mathit{\Delta v}$ would also require considerably more mass: ten times more mass as fuel in the case of a Dimorphos-like asteroid that we wanted to give a $\mathit{\Delta v}$ of about one centimetre per second.

The benefit of a mass driver is that the spacecraft turns the asteroid into a fuel source. Since each $\mathit{\delta v}$ is quite small, there is also no risk of fragmenting the asteroid from the impulses themselves. Moreover, should Space resource utilisation of asteroids become commonplace, mining spacecraft could be repurposed for planetary defence, in a fortuitous application of dual-use technology. There are, nonetheless, numerous and potentially quite serious challenges. Physically touching the asteroid is required, which could disturb the surface layers and potentially cause instabilities, including an unwanted outburst of material. Moreover, throwing any material off the asteroid will naturally cause a debris stream, which in turn could have long-term unintended consequences – as we explain in Chapter 5 on Space mining. The asteroid will also be spinning and may need to be de-spun before mass drivers can be landed or operated effectively.

Gravity tractors, by contrast, employ Newton’s third law and momentum conservation in a manner that avoids having to physically touch the asteroid. In this approach, gravity is a finicky tether that connects the asteroid and a nearby spacecraft. Just as the mutual gravity between the spacecraft and the asteroid accelerates the spacecraft towards the asteroid, it also accelerates the much more massive asteroid towards the spacecraft. The gravity tractor’s job is to fly in formation with the asteroid, using thrusters to maintain a steady distance and direction. Gravity then provides a continuous supply of $\mathit{\delta v}$’s to the asteroid.

The benefits of a gravity tractor are clear. There is no need to physically touch the asteroid and therefore no risk of generating debris or inducing surface activity. Details such as the asteroid’s spin rate or surface composition do not matter. But there are also at least two major challenges: a large spacecraft is needed, and it needs to carry enough fuel for a lengthy period of formation flying with the asteroid.

Since there is essentially no difference between the mass driver and the gravity tractor in terms of satisfying the rocketry equations, well over 100 kilograms of fuel would be needed for an effective $\mathit{\Delta v}$ of approximately one millimetre per second using an ion thruster (or something similar) on an asteroid like Dimorphos. Note that this is for the gravity ‘tug’ only. For further context, if there were a need to give a bigger asteroid like Apophis a larger $\mathit{\Delta v}$ of approximately one centimetre per second, then about ten or more tonnes of fuel would be required. In all likelihood, however, even more fuel would be needed – because the rockets cannot be fired in the optimal direction, since this would place the asteroid in the path of the exhaust. Moreover, we have again used the term ‘effective’ $\mathit{\Delta v}$ to signify that the impulse is not instantaneously applied, which can lead to some important differences in the detailed orbital evolution.

There is one further point to mention. With an impulsive technique such as a kinetic impactor or an NED, the full $\mathit{\Delta v}$ can be applied at an optimal orbital configuration, providing the maximal movement on the B-plane for the given momentum change. In contrast, low-impulse, long-duration techniques apply ${\mathit{\delta v}}^{’}\mathrm{s}$ throughout the orbit. For this reason, the total $\mathit{\Delta v}$ for a mass driver or a gravity tractor could be larger than that needed for an impulsive technique, all other things being equal. Still, the low-impulse methods have many advantages – if there is sufficient lead time to implement them! A practical scientific demonstration of such a method, similar to the demonstration of a kinetic impactor being provided by DART, would be a major contribution to planetary defence.

6.5.1 International Asteroid Warning Network

The International Asteroid Warning Network (IAWN) connects astronomers, observatories and other institutions that were already engaged in identifying and studying potentially hazardous NEOs. By pooling existing capabilities, IAWN aims to ‘discover, monitor, and physically characterize’ the entire population of potentially hazardous NEOs using ‘optical and radar facilities and other assets based in both the northern and southern hemispheres and in space’.Footnote ⁵⁶ It also serves as an international clearing house for NEO observations,Footnote ⁵⁷ co-ordinates campaigns for the observation of NEOs of particular concern, and recommends criteria and thresholds for when emerging impact threats should be communicated to national governments and general publics. Finally, IAWN aims to develop a database of potential ‘impact consequences’, to assess ‘hazard analysis results’, to communicate them to governments and to assist in the planning of ‘mitigation responses’.Footnote ⁵⁸ These latter activities, it should be noted, are directed at dealing with the effects of an impact after it occurs.

Participation in IAWN is open to all governmental and non-governmental entities with relevant capabilities, including survey telescopes, follow-up observations, orbit computations, hazard analysis, data distribution, processing and archiving. However, participants must accept a policy of free and open communication. If someone identifies an NEO threat, they must tell everyone else about it! The network’s ‘Statement of Intent’ currently has more than 40 signatories, ranging from highly skilled ‘amateur’ astronomers to NASA, ESA, the China National Space Administration and the Special Astrophysical Observatory of the Russian Academy of Sciences.Footnote ⁵⁹ In terms of participation in IAWN, a shared interest in knowing about cataclysmic threats has superseded national rivalries.

6.5.2 Space Mission Planning Advisory Group

The Space Mission Planning Advisory Group (SMPAG, generally pronounced ‘same page’) was created to ‘prepare for an international response to an NEO impact threat through the exchange of information, development of options for collaborative research and mission opportunities, and NEO threat mitigation planning activities’.Footnote ⁶⁰ Currently composed of representatives from 18 Space agencies, including NASA, ESA, Roscosmos and the China National Space Administration, SMPAG addresses issues such as the feasibility of and options for mitigating an impact threat through a Space mission, and the length of time that it would take to build and launch a spacecraft to deflect an NEO. SMPAG is grounded on a shared conviction that the ‘threat of an asteroid or comet impact is a real and global issue demanding an international response’. Recognising that states ‘already share a number of common interests in NEO threat identification and mitigation’, SMPAG aims ‘to develop cooperative activities among its members and to build consensus on recommendations for planetary defense measures’.Footnote ⁶¹ In other words, unlike IAWN, which focuses on international co-operation in the detection of potentially hazardous NEOs, SMPAG focuses on co-ordinating the capabilities that might be needed to deflect or destroy them.

That said, SMPAG is not working to marshal a fleet of asteroid deflection spacecraft and rockets in preparation for a planetary emergency. Nor would it fulfil any decision-making role should such an emergency arise. Rather, SMPAG would respond to a credible impact threat by proposing ‘mitigation options and implementation plans for consideration by the international community’.Footnote ⁶² This means that the decision makers would be national governments, whether acting unilaterally, in some ad hoc coalition, or through an existing international mechanism such as the United Nations Security Council. No predeterminations have been made as to who would contribute, and what they would contribute, in the event of an Earth impact emergency. These issues, of who decides and who acts, will be discussed below. First, however, we should consider the kinds of decisions that would have to be taken.

6.7.1 Information sharing

If an NEO with a potentially dangerous orbit is discovered, it is almost inconceivable that the astronomers involved would not promptly inform the global astronomical community of their find. There are strong ethical obligations to share information that could potentially save millions of lives. Moreover, science relies on the international circulation of discoveries and data, and careers are made through peer-reviewed publications leading to global reputations. The astronomers who discovered a significant NEO threat would thus have powerful incentives to share that information; indeed, if the NEO were a comet, it would be named after them. But even in the absence of strong ethical and professional motivations, keeping secret the existence of a possible Earth-impacting asteroid or comet is not a real possibility. After observations of an NEO are taken, the data are almost universally submitted to the International Astronomical Union’s Minor Planet Center and made publicly available, and for good reason. Initial observations are typically unable to yield reliable orbit solutions. Publicly accessible announcements of new objects are circulated so that astronomers worldwide, including highly skilled ‘amateurs’, can acquire more observations of the object. Several independent groups also focus on computing orbital solutions with known data. Even independently of formal collaborations, the detection and characterisation of NEOs is a team effort and thus involves many people with open information.

National governments are unlikely to interfere because they, too, would have powerful incentives to share the information. Even if the data are sufficient to determine whether an impact will happen, the resulting orbit solutions are unlikely, at first, to be accurate enough to determine an impact location with actionable certainty. With everyone (or at least many) at equal risk, all states would have an equal interest in seeing the full deployment of the international astronomical community’s capabilities to determine whether an impact were forthcoming, and where exactly it would take place.

International law augments these reasons for information sharing with a binding legal obligation that can be inferred from two articles of the 1967 Outer Space Treaty. Article IX reads,

As we explained in Chapter 3, observatories are engaged in the ‘exploration’ of Space. It is also clearly in the interests of all parties to the treaty to be promptly informed of any new NEO threat.

Further to this, Article XI sets out a general obligation to share information:

This treaty obligation is uncontroversial and may well have contributed to the development of a parallel rule of customary international law that binds all states and not just the parties to the Outer Space Treaty. This process, of treaties contributing to parallel customary obligations, is well established in the international legal system – and will be discussed at greater length below.

Treaties and customary international law are the first two ‘sources of international law’, as identified by the Statute of the International Court of Justice. The third source of international law – ‘the general principles of law recognized by civilized nations’ – is also relevant here.Footnote ⁷³ This is because the International Court of Justice (ICJ) has held that the obligation to share information in life-and-death situations is supported by ‘elementary considerations of humanity’ that constitute ‘general and well-recognized principles’ of law. As the Ad-Hoc Working Group on Legal Issues explains, in the 1949 Corfu Channel Case,

This led the Ad-Hoc Working Group to conclude, ‘While the case does not address the specific situation of an NEO impact threat, it can nevertheless support the argument that elementary considerations of humanity can form the basis of a duty to share information in order to avoid the loss of human lives.’Footnote ⁷⁷

The obligation to share information promptly and publicly about NEO threats is thus strongly supported on ethical, professional, practical and legal grounds.

6.7.2 Assisting Other States

If an asteroid on an Earth impact trajectory is identified, and if the asteroid is small enough that the damage will be limited to one state or a small number of states, those states clearly have the right to attempt a deflection mission. They might then be responsible for any damage caused to third states, for instance if the mission altered the trajectory of the asteroid only slightly, causing it to strike a state or states which had not initially been threatened. This issue of ‘state responsibility’ will be addressed below, along with the question whether this damage could be excused by either a United Nations Security Council resolution or ‘circumstances precluding wrongfulness’.

The Ad-Hoc Working Group raises another issue, namely whether states having the capability to mount a deflection mission are legally obligated to come to the assistance of states that lack this capability but discover that they will be the location of an NEO impact. We will turn to this issue of a possible legal obligation in a moment, but first there are compelling reasons to believe that the issue is unlikely ever to arise. Indeed, there are at least three reasons to believe that spacefaring states would always seek to prevent an NEO impact even if their own territories and populations were not directly threatened. First, an asteroid large enough to cause significant damage in one state will have indirect effects in other states. These effects could include alterations to the climate, if large amounts of material are lofted into the atmosphere, leading to a consequential diminishment of global food supplies. They could impact the broader economy, if international trade, investment and travel are disrupted. They could also lead to migration, if an asteroid strike in one country forced large numbers of people to flee to other countries, either before or after the impact. Such sudden and dramatic changes could, in turn, affect the political stability or national security of multiple states. Second, the random nature of NEO threats means that an impeding strike on a non-spacefaring state or states would provide an excellent opportunity for spacefaring states to test their deflection capabilities, knowing that another NEO will, sooner or later, eventually threaten them. Third, governments everywhere are responsive to public opinion. It is difficult to imagine that the public in the United States, Europe, Russia or China, would – if accurately informed about the situation – abide their leaders abandoning millions of fellow human beings to a preventable NEO threat.

Now we turn to the Ad-Hoc Working Group’s analysis of the issue, which concluded that, ‘in the absence of specific and clear obligations under international law, States are free to decide whether they provide assistance to other States that are threatened by a possible NEO impact’.Footnote ⁷⁸ Unfortunately, by limiting its analysis to the search for ‘specific and clear obligations under international law’,Footnote ⁷⁹ the Ad-Hoc Working Group did not consider the third source of international law, i.e. ‘the general principles of law recognized by civilized nations’.Footnote ⁸⁰ Interestingly, this is the same Ad-Hoc Working Group that, on the issue of information sharing, referred approvingly to the International Court of Justice’s finding in the Corfu Channel Case that ‘elementary considerations of humanity’ constitute ‘general and well-recognized principles’ of law.Footnote ⁸¹ Had the Ad-Hoc Working Group conducted a similar and necessary analysis with regard to a duty to rescue human beings in distress, they might have reached a very different conclusion, which we will expand on now.

The duty to rescue exists in many national legal systems. For example, section 323c(1) of the German Civil Code states,

Similarly, in the Canadian province of Quebec, its provincial Charter of Human Rights and Freedoms states,

Many other Civil Law systems, from France to Argentina to Egypt, contain the same duty to rescue.

Common Law systems do not have a general duty to rescue, although such a duty has been found in the context of pre-existing relationships, for instance teachers vis-à-vis their students, or parents vis-à-vis their children.Footnote ⁸⁴ In the United States, numerous states have ‘Good Samaritan’ statutes, and some of these contain a duty to rescue.Footnote ⁸⁵ In Vermont, for instance,

Other US states, however, only go so far as to provide immunity from civil liability to a person who acts to rescue another and, in doing so, inadvertently causes harm.

Internationally, the duty to rescue is included in numerous treaties. The International Convention for the Safety of Life at Sea (SOLAS Convention) was adopted in 1914, with the negotiations having been prompted by the sinking of the Titanic two years earlier. Although it has been updated many times since then, the SOLAS Convention has always required each party ‘to ensure that any necessary arrangements are made for coast watching and for the rescue of persons in distress at sea round its coasts’.Footnote ⁸⁷ The 1944 Convention on International Civil Aviation (Chicago Convention) has an entire annex devoted to search and rescue. Parties to the Chicago Convention are required to assist survivors of accidents regardless of nationality.Footnote ⁸⁸ The 1979 International Convention on Maritime Search and Rescue (SAR Convention) requires states parties, individually or co-operatively, to ‘participate in the development of search and rescue services to ensure that assistance is rendered to any person in distress at sea’.Footnote ⁸⁹

The 1982 United Nations Convention on the Law of the Sea reinforces these earlier treaties, with Article 98(1) reading,

The duty to rescue is also found in numerous regional and bilateral treaties, and not just between allies. For instance, in 1988 the United States and the Soviet Union concluded a bilateral treaty on maritime search and rescue.Footnote ⁹¹ Then, after the loss of the Russian nuclear attack submarine Kursk in the year 2000, Russia and NATO signed an agreement on submarine rescues in 2003.Footnote ⁹² Two years later, a British submersible was used to free seven Russian sailors whose mini submarine had become tangled in a fishing net 190 metres below the surface of the Pacific Ocean off the Kamchatka peninsula.Footnote ⁹³ There is also the 2011 Agreement on Cooperation on Aeronautical and Maritime Search and Rescue in the Arctic, which reiterates the obligations of the Chicago Convention and the SAR Convention in a regional context among the eight Arctic states, and includes five NATO states as well as Russia.Footnote ⁹⁴

The duty to rescue is found in the 1967 Outer Space Treaty, with the first sentence of Article V reading, ‘States Parties to the Treaty shall regard astronauts as envoys of mankind in outer space and shall render to them all possible assistance in the event of accident, distress, or emergency landing on the territory of another State Party or on the high seas’.Footnote ⁹⁵ Indeed, the duty to rescue was considered so fundamental that, the very next year, the same states concluded the 1968 Agreement on the Rescue of Astronauts, the Return of Astronauts and the Return of Objects Launched into Outer Space (Rescue Agreement).Footnote ⁹⁶ The Rescue Agreement elaborates on Article V of the Outer Space Treaty. It explains that, in any given situation, the duty to rescue requires that ‘those Contracting Parties which are in a position to do so shall, if necessary, extend assistance in search and rescue operations for such personnel to assure their speedy rescue’.Footnote ⁹⁷ As we explain in Chapter 1, this duty applies everywhere: within the jurisdiction of each respective state party as well as in areas beyond national jurisdiction, such as the high seas and Space.

For all these reasons, the Ad-Hoc Working Group was wrong to conclude that ‘States are free to decide whether they provide assistance to other States that are threatened by a possible NEO impact’.Footnote ⁹⁸ States have a general duty to rescue people in distress that would most certainly be engaged by an impending asteroid or comet strike. Of course, the duty is not absolute: no state would be required to develop a deflection capability in order to come to the assistance of another state. However, what if a state already had a deflection capability, including both spacecraft and rockets, on standby? Balancing the duty to rescue against any risks and expenses associated with acting is a fact-specific determination, one that does not detract from the existence of this duty as a general principle of law among the community of nations.

6.7.3 Nuclear Explosive Devices

The potential use of NEDs for deflecting or destroying asteroids is debated among international lawyers, with this debate connecting to a larger one about the legality of using or even possessing nuclear weapons.Footnote ⁹⁹ In this section, we will demonstrate that most of the legal discussion about using NEDs for planetary defence is of limited relevance. This is because a nuclear explosion in Space would constitute a clear violation of the 1963 Limited Test Ban Treaty,Footnote ¹⁰⁰ which binds the two states most likely to attempt such an action, namely the United States and Russia. It is also possible that the prohibition on nuclear explosions in Space has become a rule of customary international law, in which case it would bind non-parties to the Limited Test Ban Treaty, most notably China – which tested its first atomic bomb in 1964 but has never conducted a nuclear test in Space.Footnote ¹⁰¹ China has signed but not ratified the 1996 Comprehensive Test Ban Treaty, which, if it ever comes into force, will ban all nuclear tests including in Space.Footnote ¹⁰²

This is not the end of the discussion, however, since the United Nations Security Council could still authorise the use of an NED, with resolutions adopted under Chapter VII of the UN Charter prevailing over conflicting rules of international law. Moreover, if the Security Council failed to adopt a resolution – for instance because of a veto cast by one of its five permanent members – a state could still use an NED and claim ‘necessity’.Footnote ¹⁰³ Necessity, as we will explain below, is a ‘circumstance precluding wrongfulness’ within the international law of state responsibility. The relevant questions, at that point, would concern whether the criteria for necessity had been fulfilled.

6.7.4 Nuclear Explosive Devices and the Outer Space Treaty

Most legal discussions concerning the use of NEDs in planetary defence start with the first paragraph of Article IV of the Outer Space Treaty, which reads,

It is important to note that this paragraph does not prohibit the launch of nuclear weapons into Space if they do not make an orbit around the Earth. Nor does it say anything about whether NEDs used for planetary defence should be distinguished from nuclear weapons. However, the absence of any such provisions has not stopped international lawyers from debating whether an NED should be considered a weapon. The Ad-Hoc Working Group looks to dictionaries, writing, ‘Generally, the term “weapon” can be defined as “any object used in fighting or war, such as a gun, bomb, knife” (Cambridge English Dictionary) or as “an instrument of any kind used in warfare or in combat to attack and overcome an enemy” (Oxford English Dictionary)’.Footnote ¹⁰⁵

Both these definitions require that the object be used to fight or wage war, which an NED would not. The Ad-Hoc Working Group concedes this point,Footnote ¹⁰⁶ but then moves past the dictionary definitions to come to the opposite conclusion:

This then leads the Ad-Hoc Working Group to conclude, ‘Since, following the analysis above, NEDs can be qualified as “nuclear weapons”, their use in the context of planetary defence missions falls under the scope of this provision’ (i.e. Article IV, first paragraph).Footnote ¹⁰⁸

Again, this conclusion is contestable. As terrorists have demonstrated, cars and passenger planes can be used as weapons, even though they are not designed or considered as such. More to the point, a kitchen knife is not considered a weapon, unless wielded with hostile intent. Even firearms used for hunting – especially subsistence hunting (i.e. for food) – are not generally considered weapons.

Other international lawyers have taken a more nuanced approach. James Green engages in a lengthy exercise in treaty interpretation, including a foray into the negotiating records (travaux préparatoires) of the Outer Space Treaty, before concluding that an NED launched directly from Earth towards an asteroid would be permissible, but an NED stationed in Space in anticipation of an Earth impact emergency would not.Footnote ¹⁰⁹ He bases the former conclusion on the text of Article IV’s first paragraph , as well as the fact that it was negotiated at the height of the Cold War when both the United States and the Soviet Union would have wished ‘to retain the possibility of undertaking nuclear strikes against each other via intercontinental ballistic missiles launched out of the atmosphere on a trajectory that then returned them to their terrestrial target’.Footnote ¹¹⁰ ICBMs existed before the Outer Space Treaty was negotiated. However, permanently stationing nuclear weapons in Space would have escalated the Cold War, and this, Green explains, was something that both superpowers were cognisant to avoid.

One could quibble with some of Green’s analysis. Like the Ad-Hoc Working Group, he argues that an NED cannot be distinguished from a nuclear weapon for the purposes of the Outer Space Treaty, since the components would be identical. But again, these arguments may not actually matter, since another treaty is much clearer on the key point in question.

6.7.5 The Limited Test Ban Treaty

Although the Outer Space Treaty may not necessarily pose an obstacle to the use of an NED against an NEO, Article I(1)(a) of the 1963 Limited Test Ban Treaty is unequivocal:

1. 1. Each of the Parties to this Treaty undertakes to prohibit, to prevent, and not to carry out any nuclear weapon test explosion, or any other nuclear explosion, at any place under its jurisdiction or control:

   1. (a) in the atmosphere; beyond its limits, including outer space; or under water, including territorial waters or high seas …Footnote ¹¹¹

Note that the prohibition is not limited to nuclear weapon tests but encompasses ‘any other nuclear explosion’, including in Space. This conclusion is supported by the Preamble, which states that ‘the principal aim’ of the Limited Test Ban Treaty is ‘the speediest possible achievement of an agreement on general and complete disarmament’, and that a further aim is ‘to put an end to the contamination of man’s environment by radioactive substances.’ Moreover, the travaux préparatoires reveal that the words ‘or any other nuclear explosion’ were inserted into Article I(1)(a) to prevent the prohibition being circumvented through an assertion of ‘peaceful use’.Footnote ¹¹²

Russia, the United States, the United Kingdom and India are all parties to the Limited Test Ban Treaty; China, France and North Korea are not. The Limited Test Ban Treaty thus poses a legal obstacle to four of the states currently able to attempt an NEO deflection with an NED. But China also has nuclear warheads and large Space rockets, while France has warheads and potential access to rockets via ArianeSpace – the European launch provider.

6.7.6 Nuclear Explosive Devices and Customary International Law

It is well established that treaty provisions can contribute to parallel rules of customary international law. In the North Sea Continental Shelf Cases, the International Court of Justice considered whether a provision in the 1958 Geneva Convention on the Continental Shelf could have given rise to a parallel rule of customary international law.Footnote ¹¹³ It wrote,

It is significant that, while several states tested nuclear weapons in Space before the Limited Test Ban Treaty was adopted in 1963, none has done so since. Moreover, those states that acquired the capability to test nuclear weapons in Space after 1963 have refrained from doing so. Most significantly, two of these states – China and France – have refrained from nuclear weapon testing in Space despite never having acceded to the Limited Test Ban Treaty. However, it is uncertain whether this avoidance of nuclear weapon testing in Space has been driven by ‘a general recognition that a rule of law or legal obligation is involved’ (in the words of the ICJ),Footnote ¹¹⁵ or whether states simply became aware that such tests have dangerous consequences.

The Soviet Union conducted five nuclear weapon tests in Space in 1961 and 1962. The latter of these, which occurred at an altitude of 230 kilometres, generated such a strong electromagnetic pulse that it caused a fire in a power plant on the ground and disabled hundreds of kilometres of telephone lines. That same year – 1962 – the United States detonated a 1.44 megaton thermonuclear weapon at an altitude of 400 kilometres over the Pacific Ocean.Footnote ¹¹⁶ The test, dubbed Starfish Prime, was one of five tests in Operation Fishbowl, which sought to determine whether an artificial intensification of the Van Allen radiation belts could disable intercontinental ballistic missiles.Footnote ¹¹⁷ It provided a surprising result: the electromagnetic pulse from the explosion shut down power grids in Hawaii and disabled Telstar 1, which had just begun broadcasting live television between the United States and Europe, as well as five other satellites from the US, the United Kingdom and the Soviet Union.Footnote ¹¹⁸

However, even were the subsequent avoidance of nuclear explosions in Space the result of an awareness of risk rather than a specific legal obligation, this avoidance is occurring in a legal context that includes not only the Limited Test Ban Treaty, but also a general obligation in customary international law to not cause damage to the property of another state. The existence of this obligation is recognised in Article VII of the Outer Space Treaty, which provides that launch states are ‘internationally liable for damage to another State Party to the Treaty or to its natural or juridical persons by such [Space] object or its component parts on the Earth, in air space or in outer space, including the moon and other celestial bodies’. The 1972 Convention on the International Liability for Damage Caused by Space Objects (Liability Convention) elaborates on this provision by providing for absolute liability for damage on Earth, and fault-based liability for damage in Space.Footnote ¹¹⁹ It is easy to imagine a nuclear explosion that is caused by one state damaging or disabling satellites owned by others – as, indeed, occurred with Starfish Prime. States that can conduct nuclear weapon tests in Space and choose not to do so based on a recognition of the risks to others are, therefore, engaged in state practice and demonstrating opinio juris in support of a rule of customary international law prohibiting such explosions.

Further support for the existence of a rule of customary international law prohibiting nuclear explosions in Space can be found in the 1996 Comprehensive Test Ban Treaty, Article I(1) which reads, ‘Each State Party undertakes not to carry out any nuclear weapon test explosion or any other nuclear explosion, and to prohibit and prevent any such nuclear explosion at any place under its jurisdiction or control’.Footnote ¹²⁰ The inclusion of Space within this prohibition is confirmed by the Preamble, where the states parties are: ‘Noting the aspirations expressed by the Parties to the 1963 Treaty Banning Nuclear Weapon Tests in the Atmosphere, in Outer Space and Under Water to seek to achieve the discontinuance of all test explosions of nuclear weapons for all time’.

According to Article XIV, the Comprehensive Test Ban Treaty will not come into force until it is ratified by 44 specific states. These ‘Annex 2 states’ are those that participated in the negotiation of the treaty in 1994–1996 and possessed nuclear reactors at that time. Of these 44 states, eight have not yet ratified the treaty, although five of them – the United States, China, Egypt, Iran and Israel – have signed it. They are therefore obligated to ‘refrain from acts which would defeat the object and purpose’ of the treaty, with this general obligation in customary international law (binding on the signatories of any treaty) being recognised in Article 18 of the Vienna Convention on the Law of Treaties.Footnote ¹²¹

The three remaining Annex 2 states – India, North Korea and Pakistan – have neither signed nor ratified the Comprehensive Test Ban Treaty and seem unlikely to do so. But what is more relevant for the purposes of customary international law is that 168 states have ratified the Comprehensive Test Ban Treaty, while another 16 have signed but not yet ratified. This means that 95 per cent of countries – 184 of 193 member states of the United Nations – have thereby indicated their support for a prohibition on nuclear weapon tests that extends to Space. Considering all this, we conclude that a rule of customary international law prohibiting nuclear explosions in Space exists, with North Korea (but not India or Pakistan, because they have ratified the Limited Test Ban Treaty) perhaps remaining outside its scope as a so-called ‘persistent objector’.Footnote ¹²²

6.8.1 Space Agencies Acting Collectively

The ideal response to an NEO threat would be for Space agencies to collectively determine the feasibility and risks of different mitigation options, decide on the best approach and implement it. As discussed above, Space agencies are already working together on these issues through an International Asteroid Warning Network that reports to the Space Mission Planning Advisory Group, currently made up of representatives from 18 Space agencies. Both bodies were created in 2013 at the recommendation of the UN Committee on the Peaceful Uses of Outer Space.Footnote ¹²³ The IAWN connects Space agencies, observatories and other groups engaged in discovering, monitoring and characterising potentially hazardous NEOs; it also serves as a ‘clearing house’ for NEO observations and recommends criteria and thresholds for notification of emerging threats.Footnote ¹²⁴ The SMPAG prepares for an international response to an NEO threat by exchanging information, developing options for collaborative research and mission opportunities, and conducting threat mitigation planning activities. Yet it is unclear whether Space agencies would be allowed to lead on these issues in the event of an actual NEO threat, given that militaries are also active in Space and often have much greater political influence. However, as we will explain, militaries are poorly suited for planetary defence. For this reason, states should commit in advance to keeping Space agencies in charge during such eventualities, with this commitment expressed in a multilateral declaration or, better yet, treaty. This commitment could then be implemented in national legal systems, to bind militaries directly.

6.8.2 A Decision-Making Matrix

The Ad-Hoc Working Group recognised that responses to NEO threats will be complicated because of the absence of international legal instruments explicitly addressing this issue and because of the short time that might be available to make decisions and to act at the international level. For these reasons, it suggested the drafting of a template for decision making that could quickly be adapted and adopted by the international community in the face of a specific threat. This template could include ‘modalities for the organization of cooperation among States and intergovernmental organizations’, as well as for the dissemination of information regarding NEO threats.Footnote ¹²⁵ As the Ad-Hoc Working Group explained, any such decision-making matrix would have to balance between the need for transparency and inclusion and ‘the importance of avoiding a lengthy process that inhibits an effective response and of providing the flexibility and the resilience that is required’.Footnote ¹²⁶

The Ad-Hoc Working Group issued its report before the COVID-19 pandemic exposed the weakness of multilateral co-operation in the field of global health. Given the rapid sidelining of the World Health Organization, how confident can we be that SMPAG – another voluntary mechanism designed to co-ordinate state responses – will operate effectively during an equally significant global crisis? Will national political struggles or leadership failings impair collective decision making? Will powerful states turn away from international co-operation as they prioritise national preservation? These last questions become even more relevant with regard to the most powerful international body, the United Nations Security Council, where the five permanent members each hold a veto over decision making.

6.8.3 The UN Security Council

One hundred and ninety-three states have ratified the 1945 United Nations Charter and are consequently bound to accept Security Council decisions made under Chapter VII of that treaty. Chapter VII gives the Security Council the power to ‘determine the existence of any threat to the peace, breach of the peace, or act of aggression’ and to ‘decide what measures shall be taken … to maintain or restore international peace and security’. Such measures can include ‘action by air, sea, or land forces’, including within the territory of non-consenting states.Footnote ¹²⁷

For decades, the Security Council has taken a broad view of international security. It has invoked Chapter VII to impose an embargo on the racist government of Southern Rhodesia in 1968,Footnote ¹²⁸ to deliver humanitarian aid in Somalia in 1992,Footnote ¹²⁹ and to require measures against terrorist financing in all national legal systems in 2001.Footnote ¹³⁰ In 2010, the Security Council invoked Chapter VII in response to an earthquake in Haiti, authorising the deployment of 680 police to augment the existing UN peacekeeping and humanitarian mission in that country.Footnote ¹³¹ There is no question that the Security Council could decide that an NEO was a threat to international peace and security and make decisions – such as authorising the use of an NED – that would otherwise violate international law. The Security Council could even provide the acting states with a waiver of liability for any resulting damage, liability that might otherwise flow from the 1967 Outer Space Treaty, the 1972 Liability Convention or customary international law.

The peremptory effect of Chapter VII resolutions comes from Article 103 of the UN Charter, which reads, ‘In the event of a conflict between the obligations of the Members of the United Nations under the present Charter and their obligations under any other international agreement, their obligations under the present Charter shall prevail.’ The only limiting factor here is that Security Council resolutions require the support of at least nine of its 15 members, including the concurring votes (or abstentions) of all five permanent members. In other words, China, France, Russia, the United Kingdom and the United States all hold vetoes over Security Council resolutions. In the literature on planetary defence, the veto is generally regarded as a bad thing, since it could prevent the Security Council from authorising necessary action. However, the veto can also serve as a check on precipitous or incautious action, such as a deflection mission that is not supported by a careful scientific assessment of the risks involved.

In any event, if Security Council decision making is blocked because of a veto, it is conceivable that one or more states might proceed with an unauthorised deflection mission.

6.8.4 Individual States

The DART mission (discussed above) was led by NASA, with some participation from the European Space Agency and the Italian Space Agency.Footnote ¹³² Although there was no risk that the targeted moonlet would inadvertently be directed onto an Earth impact trajectory, it is noteworthy that the United States did not seek the consent of other states during the mission-planning process. It did consult with them, however, including by informing the SMPAG.

If an NEO is discovered on an actual Earth impact trajectory, one or more states might take it upon themselves to mount a deflection mission. Some experts have pointed to the right of self-defence, a rule of customary international law that is codified in Article 51 of the UN Charter, as providing a legal avenue for unilateral action. However, as Green correctly explains,

No rule of international law directly prohibits a unilateral deflection mission directed against an asteroid or comet. Rather, the legal issues concern the rights of other states to be consulted, not to be exposed to increased risk, and to obtain compensation in the event of an accident. There are other legal issues concerning the use of NEDs, concerning the role of the UN Security Council in this regard (and in relation to possible waivers of liability), and concerning the possibility that ‘circumstances precluding wrongfulness’ might excuse a breach of international law. We will delve deeper into some of these issues below.

The possibility of a unilateral deflection mission is real, given the panic and selfishness that can arise in crisis situations. Adding to the risk, militaries could insist on being involved in national decision making concerning an NEO threat, because of the scale of the threat, the kind of equipment that could be used – especially NEDs – and the fact that the political influence of militaries always exceeds that of Space agencies.

Militaries, however, are poorly suited for planetary defence. They tend to favour forceful actions over more subtle interventions such as diplomacy. In the context of planetary defence, they might favour kinetic impactors or NEDs over slow-moving mass drivers or gravity tractors. Militaries are not involved in current NEO detection and mission-planning exercises at the international level, notably IAWN and SMPAG, and therefore might not be fully informed on these matters – and more likely to make mistakes. Finally, militaries tend to co-operate with smaller circles of states than Space agencies, making them poorly suited for multilateral initiatives that include non-allies. There is a reason why the International Space Station, where Americans and Russians work side by side, is operated by civilian Space agencies rather than militaries.

Fortunately, even a unilateral military-led response to an NEO threat would be constrained by international law. First, any state planning a unilateral deflection mission would have to consult with other states. Article IX of the Outer Space Treaty reads, in part,

Consultation offers many benefits, one of which is an increased likelihood that careful scientific analyses of the risks of any proposed action will take place. Militaries acting unilaterally have done incautious things in the past in Space. For example, between 1961 and 1963, the US military launched 480 million copper needles into orbit, in an attempt to create an artificial ring around Earth for relaying radio signals.Footnote ¹³⁵ Most of these needles have long since re-entered the atmosphere, driven by the effects of solar radiation, but clumps of them remain in orbit today – contributing to the serious and growing problem of Space debris. Consultation is critical in the context of an NEO threat since it might well lead to an initial rendezvous mission to determine the asteroid’s precise orbit, shape and composition, and thus serve as the basis for a considered response.

In lieu of consulting other states and taking their interests into account, what would happen if a military decided to act unilaterally? What if the NED were launched without pooling scientific knowledge and mission capabilities with other states, and before carefully considering other methods? Does international law empower other states to prevent a powerful state from acting in this way? One possible step would be economic countermeasures, up to and including sanctions, especially if the acting state was a party to the 1963 Limited Test Ban Treaty (which, again, clearly prohibits the use of an NED).Footnote ¹³⁶ However, economic countermeasures are unlikely to have any immediate effect on a powerful state that sees itself responding to a serious threat. A more interesting question concerns pre-emptive self-defence. If other states had scientific evidence that the unilateral deflection mission would increase the risk to their territory and populations, for instance by changing the impact location, could they take pre-emptive military action to prevent the launch?

This scenario, it should be noted, is fundamentally different from discussions of self-defence as a justification for deflection missions, as dismissed by Green at the beginning of this section. Here, the threat comes from another state interfering with the asteroid.

The existence and extent of a right of pre-emptive self-defence is hotly contested in international law, with the United States leading the push for a more expansive approach, and many smaller, less powerful states resisting its efforts.Footnote ¹³⁷ We will not repeat that debate here. It is, however, readily conceivable that a state might claim pre-emptive self-defence in circumstances where it feels threatened by another state interfering with a natural force. Imagine if a state had reason to believe that another state was preparing to attack a hydroelectric dam upstream from a population centre. However, if a right of pre-emptive self-defence exists, and if it were invoked to justify military action to prevent the launch of an asteroid deflection mission by another state, that action would still be subject to the usual constraints imposed on self-defence under customary international law, namely that the response must be both ‘necessary’ and ‘proportionate’.Footnote ¹³⁸

Although the issue of pre-emptive self-defence is interesting in this context, it is unlikely ever to arise. Militaries cannot simply launch their existing nuclear missiles against incoming NEOs, since ballistic missiles cannot achieve escape velocity. There will always be time to consult with other states, receive additional scientific input and engage in sober second thought. As this happens, international law will move to the forefront of the deliberation process, not least because of rules on state responsibility and liability, rules that would apply to any damage caused on Earth by a failed deflection mission.

6.9.1 Circumstances Precluding Wrongfulness

Sometimes in the international realm, just as in the domestic, lawbreakers are excused for their actions because of the unusual circumstances they found themselves in. If a state chose to violate international law while engaging in planetary defence, for example by using an NED, it is possible that the violation would be excused because it took place under ‘circumstances precluding wrongfulness’. The different circumstances that can preclude wrongfulness are identified in the International Law Commission’s Draft Articles on State Responsibility, with ‘consent’ and ‘necessity’ being of greatest potential relevance here.Footnote ¹⁴¹ It is also important to note that, according to Article 27 of the Draft Articles, the invocation of a circumstance precluding wrongfulness does not relieve the acting state of any obligation to provide compensation for any material loss caused by the otherwise illegal act in question. In other words, being excused for the wrong does not relieve the state of any obligation to pay compensation.

6.9.2 States Are Responsible for Non-State Actors

There is one important difference in the international law of state responsibility as it applies in Space as compared to elsewhere, and this concerns non-state actors. Generally speaking, the conduct of private actors is not attributable to a state under international law. But according to Article VI of the Outer Space Treaty,

It is easy to imagine SpaceX mounting an asteroid-deflection mission if Elon Musk felt that national governments were moving too slowly or taking the wrong approach. But if SpaceX undertook such a mission, responsibility for complying with international law and providing compensation for any damage would rest with the US government, since SpaceX is incorporated in the United States. The same rule would apply, self-evidently, to the conduct of any private contractor taking part in a state-led mission. The point, however, is to ensure that national governments have a strong incentive to regulate private companies, and exercise strong oversight, because it is the governments that will carry the legal and financial burdens internationally if something goes wrong.

6.10.1 Liability for False Alarms?

Another liability issue identified by the Ad-Hoc Working Group concerns false alarms. A warning about an NEO threat could cause a government to launch a deflection mission or evacuate a city or region. Is anyone legally responsible for the associated expenses and disruptions if the warning proves to be a false alarm? But while this question is interesting, it is only of limited relevance, since the astronomical community quickly verifies or disproves any newly identified NEO threat, thus reducing the time during which an alarm which turns out to be false is perceived to be real enough to entail unnecessary expenses or disruptions.

For more than three decades, Brian Marsden directed the Central Bureau for Astronomical Telegrams, the body created by the International Astronomical Union to serve as an international clearing house for information on transient astronomical events. In 1992, Marsden warned of a possible Earth impact from Comet Swift Tuttle in 2126.Footnote ¹⁵⁸ This would have been an extinction-level event, because Swift Tuttle has a nucleus with a diameter of about 26 kilometres, making it the largest object in the Solar System that repeatedly passes close to Earth, doing so with a relative velocity of about 60 kilometres per second. Fortunately, information obtained from historical Chinese reports in 69 BCE and 188 CE, combined with further observations, enabled the astronomical community to rapidly determine that there is, in fact, no impact risk from Comet Swift Tuttle for the next two millennia.Footnote ¹⁵⁹

In 1998, Marsden issued another warning that also turned out to be a false alarm, this time concerning a possible impact with the asteroid 1997 XF₁₁ in 2028. The warning caused a media frenzy, partly because it was issued by press release rather than circulated within the astronomical community, and partly because 1997 XF₁₁ has a diameter of approximately one kilometre. Fortunately, it took less than a day for other astronomers to resolve the asteroid’s orbit with greater precision, after one of them found an eight-year-old image of the same asteroid. It will indeed pass by Earth in 2028, but at about 2.4 lunar distances (930,000 kilometres) from us.Footnote ¹⁶⁰

These false alarms demonstrated how quickly the astronomical community reacts to possible Earth impacts, by providing new data and analysis and almost immediately verifying or disproving the existence of a threat. The cost of a false alarm is therefore quite limited, while the benefit of an early warning of an actual strike – i.e. more time in which to act – would be preserved.

In any case, as the Ad-Hoc Working Group points out, the issue of liability for false alarms does not fall within the scope of the Outer Space Treaty or the Liability Convention, since there is no ‘space object’ – in the legal sense of a human-made spacecraft or parts thereof – involved here. Rather, the issue is governed by the general rules of international law on state responsibility and liability, at least in cases where the warning has been issued by a state, including a national Space agency, or experts acting on behalf of a group of states or Space agencies, as with IAWN. These rules involve fault-based rather than strict liability, leading the Ad-Hoc Working Group to conclude,

But if states and Space agencies are protected, what about false alarms issued by non-governmental organisations? Brian Marsden, for example, worked for a non-governmental organisation, namely the International Astronomical Union. Many other astronomers work for universities, some of them private. What about amateur astronomers, hundreds of whom are actively and very ably involved in NEO detection?

International law principally applies among nation states. As a result, the question of liability for false alarms issued by non-state groups and individuals is one of national law, which varies from state to state. The good news is that liability in most national legal systems is fault-based, and as a result a warning issued in good faith will not generate liability. Just as importantly, a warning issued by a reputable non-governmental organisation or amateur astronomer will attract the attention of the astronomical community almost as fast as a warning issued by a state, meaning that new data and analysis will be brought to bear quickly – before a false alarm can result in unnecessary expenses or disruptions. For all these reasons, liability for false alarms is not much of an issue. Astronomers, it turns out, have each other’s backs.