CORDIS - Forschungsergebnisse der EU

A Global Approach to Near-Earth Object Impact Threat Mitigation

Final Report Summary - NEOSHIELD (A Global Approach to Near-Earth Object Impact Threat Mitigation)

Executive Summary:
NEOShield was conceived to address realistic options for preventing the collision of a naturally occurring celestial body (near-Earth object, NEO) with the Earth. Three deflection techniques, which appeared to be the most realistic and feasible at the time of the European Commission’s call in 2010, form the focus of NEOShield efforts: the kinetic impactor, in which a spacecraft transfers momentum to an asteroid by impacting it at a very high velocity; blast deflection, in which an explosive, such as a nuclear device, is detonated near, on, or just beneath the surface of the object; and the gravity tractor, in which a spacecraft hovering under power in close proximity to an asteroid uses the gravitational force between the asteroid and itself to tow the asteroid onto a safe trajectory relative to the Earth.

Detailed test-mission designs are a primary product of the NEOShield project, and are intended to enable rapid development of actual test missions at a later stage.

A prerequisite for the successful deflection of a NEO is relevant knowledge of its physical characteristics. The knowledge required depends on the deflection technique in question, but would include such parameters as mass, shape, spin vector, albedo, and in some cases surface porosity, near-surface structure, and mineralogy. Ideally we wish to know the most likely properties of a future potential impactor that could trigger a space-borne deflection action. Furthermore, by narrowing the range of expected properties, a rational basis for the choice of objects to serve as targets in deflection test missions can be provided. Investigations of the mineralogy and structure of NEOs also aid in predicting the effects of an airburst in the atmosphere or an impact on the ground. Important scientific components of the NEOShield project are: analyses of observational data of NEOs, especially infrared data which provide insight into sizes, albedos, mineralogy, and surface thermal characteristics; laboratory experiments in which projectiles are fired at materials thought to be analogous to those in asteroids; and computer modelling and simulations to incorporate the laboratory results into scaled-up investigations of impulsive NEO deflection techniques. The results of the scientific work provide insight into how asteroids would respond to deflection attempts. Related scientific tasks are studies of the most appropriate observations, instrumentation, and types of space mission to efficiently provide mitigation-relevant information on a threatening NEO, and the identification of potential NEO targets for deflection demonstration missions.

The development and optimisation of some key technologies are essential preparation for a deflection test on a small asteroid. In circumstances in which a kinetic impactor is the technique of choice, it might be necessary to impact a small, only partially illuminated, dark asteroid at high velocity, a challenging task with currently available technology. In order to increase the probability of success, the NEOShield program of work included improvements to spacecraft guidance, navigation and control systems. Likewise, in the case of the gravity tractor, studies have been carried out of complex systems that would allow a spacecraft to manoeuvre and hold its position close to a low-gravity, irregularly-shaped, rotating asteroid, during a deflection attempt that might last years to a decade. In the case of blast deflection, NEOShield work concentrated on theoretical studies of the effects of a nuclear explosion on a typical small asteroid. Obvious military and political issues currently preclude research into optimised explosive devices and tests of this method in space.

Testing deflection techniques, such as the kinetic impactor and gravity tractor, is a vital prerequisite to a reliable international NEO defence system. Results from the type of studies carried out by NEOShield obviously serve to reduce the scientific and technical preparatory work required to bring an appropriate and viable deflection mission to the launch pad in an emergency situation. Two new UN-sanctioned bodies, the International Asteroid Warning Network (IAWN) and the Space Mission Planning Advisory Group (SMPAG), of which NEOShield personnel are members, were established under the auspices of the UN Committee on the Peaceful Uses of Outer Space (COPUOS ) during the active lifetime of the NEOShield project. The participation of NEOShield personnel in the most prominent current international efforts addressing the impact hazard, ensures a broad international stage for the dissemination of NEOShield research.
Project Context and Objectives:
(Note: Figure and table numbers refer to the material uploaded as an attached pdf file.)

Collisions of celestial objects with the Earth have taken place frequently over geological history and major collisions of asteroids and comets with the Earth will continue to occur at irregular, unpredictable intervals in the future. As a result of modern observing techniques and directed efforts thousands of near-Earth objects (NEOs) have been discovered over the past 20 years and the reality of the impact hazard has been laid bare. Even relatively small impactors can cause considerable damage: the asteroid that exploded over the Russian city of Chelyabinsk in February 2013 had a diameter of only 18 m yet produced a blast wave that damaged buildings and caused injuries to some 1500 people (Fig. 1). The potentially devastating effects of an impact of a large asteroid or comet are now well recognized.

Asteroids and comets are considered to be remnant bodies from the epoch of planet formation. Planet embryos formed in the protoplanetary disk about 4.5 billion years ago via the accretion of dust grains and collisions with smaller bodies (planetesimals). A number of planet embryos succeeded in developing into the planets we observe today; the growth of other planet embryos and planetesimals was terminated by catastrophic collisions or a lack of material in their orbital zones to accrete. Most asteroids are thought to be the fragments of bodies that formed in the inner Solar System and were subsequently broken up in collisions.

As a result of collisions, subtle thermal effects and the very strong gravitational field of Jupiter, small main-belt asteroids can drift into certain orbital zones from which they may be ejected under the influence of Jupiter into the inner Solar System. The population of NEOs is thought to consist mainly of such objects, together with an unknown smaller number of old, inactive cometary nuclei. At the time of writing the number of known NEOs exceeds 12000; over 1500 of these are so-called potentially hazardous objects (PHOs), i.e. those having orbits that can bring them within 7.5 million kilometres of the Earth’s orbit and are large enough (diameter ≥ 100 m) to destroy a large city or urban area and kill millions of people if they were to impact the Earth. Smaller objects can also present a significant threat: the Chelyabinsk event is a very recent example (see above); a somewhat larger object caused the Tunguska event of 1908 in Siberia, in which an area of over 2000 square km was devastated and some 80 million trees felled. The Tunguska event is thought to have been due to the airburst of an object with a diameter of 30 - 50 m at a height of 5 - 10 km. The estimated impact frequency of NEOs on the Earth depends on size. The impact frequency increases with decreasing size due to the size distribution of the asteroid population: there are many more small objects than large ones. Current, albeit uncertain, statistical knowledge of the NEO size and orbital distributions indicates that NEOs with diameters of 50, 100, 300 m, for example, impact roughly every 1000, 10,000, and 70,000 years, respectively.

The known NEO population contains objects with a confusing variety of physical properties. Some NEOs are thought to be largely metallic, indicative of material of high density and strength, while some others are carbonaceous, of lower density, and less robust. A number of NEOs appear to be evolved cometary nuclei that are presumably porous and of low density but otherwise with essentially unknown physical characteristics. In terms of large-scale structure NEOs range from monolithic slabs to re-accumulated masses of collisional fragments (so-called rubble piles) and binary systems (objects with moons). More than 50 NEOs in the currently known population have been identified as binary or ternary systems and many more are probably awaiting discovery.

The phenomenon of collisions in the history of our Solar System is a fundamental process, having played the major role in forming the planets we observe today. Collisions of asteroids and comets with the Earth have taken place frequently over geological history and probably contributed to the development of life. In contrast, later impacts of asteroids and comets most likely played a role in mass extinctions. NEOs present a scientifically well-founded threat to the future of our civilisation. While past impacts have probably altered the evolutionary course of life on Earth, and paved the way for the dominance of mankind, we would now rather not remain at the mercy of this natural process.

Can we protect our civilisation from the next major impact?

In contrast to other natural disasters, such as earthquakes and tsunamis, the impact of an asteroid discovered early enough can be predicted and prevented. Partners in the NEOShield project (Table 1) are confident the basic technology necessary to prevent an impact is available now. But how do we implement it and what do we need to know about the hazardous object to maximize our chances of success? Preventing a collision with a NEO on course for the Earth would require either total destruction of the object, to the extent that remaining debris posed little hazard to the Earth or, perhaps more realistically, deflecting it slightly from its catastrophic course. In either case accurate knowledge of the object’s mass would be of prime importance. In order to mount an effective mission to destroy the object, knowledge of its density, internal structure, and strength would also be highly desirable. Deflection of the object from its course would require the application of an impulse or continuous or periodic thrust, the magnitude and positioning of which may depend on the mass and its distribution throughout the (irregularly-shaped) body, the surface characteristics, and the spin vector, depending on the strategy deployed. It is crucial to ensure that the deflection operation does not simply move the object to another hazardous trajectory. Mitigation planning takes on a higher level of complexity if the Earth-threatening object is a rubble pile or binary system.

In the case of an object with a diameter of about 50 m or less, the best course of action may be to simply evacuate the region around the predicted impact point, assuming there would be sufficient advance warning (however, only a small fraction of the asteroids in this size category have been discovered to date). For objects larger than 50 m a number of mitigation strategies may be considered, depending on circumstances. The NASA Report to Congress, “Near-Earth object survey and deflection. Analysis of alternatives” ( 2007) on the surveying and deflection of near-Earth objects concluded that nuclear devices offer the most effective means of applying a deflecting force to an asteroid. While they may offer the only feasible solution in desperate circumstances, e.g. in the case of very little advanced warning, it is clear that the geopolitical issues associated with launching nuclear devices and testing them in space seriously compromise the practicability of this technique. The NASA report concluded that the most effective non-nuclear option is the kinetic impactor, which involves applying an impulsive force to the asteroid by means of a large mass in the form of a spacecraft accurately guided to the target at a high relative velocity. The gravity tractor is a “slow-pull” approach that may require a long period of time to achieve the required amount of deflection, but is a promising technique for cases in which there are many years of advance warning, the target NEO is relatively small, and/or a very slight, precise deflection is required to prevent an impact on the Earth.

The report of the National Research Council of the US “Defending Planet Earth: Near-Earth Object Surveys and Hazard Mitigation Strategies: Final Report” ( 2010) contains the following findings and recommendation:

“Finding: Mitigation of the threat from NEOs benefits dramatically from in-situ characterization of the NEO prior to mitigation, if there is time to do so.”

“Finding: Kinetic impactors are adequate to prevent impacts on Earth by moderately sized NEOs (many hundreds of meters to 1 kilometer) with decades of advance warning. The concept has been demonstrated in space, but the result is sensitive to the properties of the NEO and requires further study.”

“Recommendation: If Congress chooses to fund mitigation research at an appropriately high level, the first priority for a space mission in the mitigation area is an experimental test of a kinetic impactor along with a characterization, monitoring and verification system, such as the Don Quijote mission that was previously considered, but not funded, by ESA. This mission would produce the most significant advances in understanding and provide an ideal chance for international collaboration in a realistic mitigation scenario.”

The objectives of NEOShield broadly reflect the above findings and recommendation. The project work packages are designed around the following tasks: NEO physical characterisation, laboratory experiments to investigate the material properties of asteroid analogue materials, modelling and computer simulations to incorporate the laboratory results into realistic size-scaled investigations of impacts into NEOs, a trade-off study of different deflection techniques, and detailed designs of deflection test missions. In the light of results arising from our research into the feasibility of the various mitigation approaches and the mission design work, a further objective was to formulate a “global response campaign roadmap” that may be implemented when an actual significant impact threat arises. The roadmap considers the necessary international decision-making milestones, required reconnaissance observations, both from the ground and from rendezvous spacecraft, and a space-mission deflection campaign.

Project Results:
The following is a brief description of highlights from the scientific and technical work packages (Table 2) following the logical structure of the project: starting with scientific results from investigations of NEO physical properties, through target selection for deflection test missions, the monitoring of a deflection attempt, post-deflection orbit evolution, to the results from the technical work packages concerned with the design of feasible deflection test missions, and tools for an international strategy or “roadmap” for responding to an impact threat.

A full list of resulting NEOShield scientific and technical documents ("deliverables") is appended to this report.

Hypervelocity gas-gun experiments and associated computer modelling:
An important component of the research into the physical properties of NEOs is laboratory work with hypervelocity gas guns and associated modelling and computer simulations. The gas-gun experiments provide data on the behaviour of asteroid surface analogue materials when impacted by a projectile at high velocity. Modelling and numerical simulations of the impact process required a detailed characterisation of the target materials covering a wide area of strain rates. These material tests, which included determination of porosity, density and chemical composition, were conducted using different types of testing facilities; the tests and results are described in NEOShield Deliverable 4.2.

Following the completion of work to specify the overall strategy and provide detailed requirements for the experiment parameters and the target materials (NEOShield Deliverable 2.4) impact experiments were conducted using a two-stage light-gas accelerator. For NEOShield purposes mm-sized spherical projectiles were fired into four different types of materials: quartzite, sandstone, limestone and aerated concrete. The momentum transfer was measured by means of a ballistic pendulum. A high-speed camera was used to record the highly transient ejection process. Three-dimensional models of the resulting craters were created providing morphological information, and crater volumes were calculated using the digital data. Targets were weighed before and after each experiment to determine the ejected mass.

Experiments were also carried out using an all-angle light-gas gun, which allowed vertical impacts of projectiles into loose material representative of asteroid regolith. The results of the NEOShield experiments (NEOShield Deliverables 4.1 and 4.3; see Figs. 2a and 2b) have provided vivid demonstrations of the dependency of the momentum multiplication factor beta (the ratio between the target momentum change as a result of the impact and the momentum of the projectile) on the target material, especially its density and porosity. An unanticipated, albeit preliminary, finding is that the results for layered targets, with a thin layer of regolith over a high-porosity solid, indicate somewhat higher momentum enhancement than for the bare high-porosity solid material. This can be understood as a result of compaction of the solid material reducing the amount of ejecta, whereas the unconsolidated regolith is more easily released.

The results of the NEOShield gas-gun experiments have demonstrated that knowledge of the physical properties of a potentially hazardous NEO is of fundamental importance for the determination of its response to a kinetic impactor and other impulsive deflection techniques. While the results from the small-scale experiments in Earth gravity cannot be directly applied to large-scale impacts into asteroids with microgravity, they provide useful inferences for impact mitigation:

- For a given material composition, the greatest momentum enhancement is from solid surfaces if the porosity is low, but from deep regolith if the porosity is high.

- The projectile properties (i.e. shape/density of impacting spacecraft) have a larger influence on the momentum enhancement than the regolith particle size/shape.

- The laboratory experiments provide valuable constraints for hydrocode simulations on small scales.

Computer modelling of the effects of various material properties on the beta factor have shown that using a higher tensile strength leads to less ejecta and therefore a smaller beta value. Using a target with a strength corresponding to laboratory scales (cm-sized bodies) leads to a significantly smaller beta value than that found for a target with a strength corresponding to a 300 m diameter object (which is about 20 times smaller). On the other hand, assuming a lower strength against crushing leads to more compaction and therefore also less ejecta and a smaller beta. Decreasing the porosity leads to increased beta.

The modelling and results are described in NEOShield Deliverables 3.2 and 3.3; see Fig. 3.

First main-journal peer-reviewed papers on the NEOShield hypervelocity gas-gun experiments and modelling work have been published:

Jutzi, M., Michel, P. (2014) Hypervelocity impacts on asteroids and momentum transfer I. Numerical simulations using porous targets. Icarus 229:247-253.
Hoerth, T., et al. (2015) Momentum transfer in hypervelocity impact experiments on rock targets. Procedia Engineering 103:197-204.

Investigation of mitigation-relevant properties of NEOs from existing observational data:
Objects with diameters greater than 300 m are relatively rare and impacts of such objects are expected to occur less than once every 70,000 years. Objects with diameters less than 50 m are expected to impact much more frequently (e.g. the Tunguska event of 1908 and the Chelyabinsk superbolide of 2013) but are likely to dissipate the bulk of their energy in the atmosphere causing relatively minor damage on the ground. For the purposes of the NEOShield project we consider the NEO diameter range of interest to be D = 50 – 300 m. Analysis of available observational data suggests that small NEOs can have very irregular shapes, rapid rotation rates of up to 4 rev/min, and, depending on models used, structures ranging from solid monoliths to aggregates (“rubble piles”) with wide ranges of possible bulk densities, porosities, and degrees of cohesion. Realistic ranges of mitigation-relevant NEO parameters have been established to guide the selection of representative targets for demonstration missions, and as an aid to mission planners (see below). We have also shown that there is little danger of a previously benign NEO being deflected onto an Earth-threatening orbit by a deflection test mission, although thorough calculations of all possible outcomes should be carried out in each case.

Observations with infrared telescopes, such as NASA’s Wide-Field Infrared Survey Explorer (WISE), have shown that the sunward surfaces of some asteroids appear cooler than others at similar distances from the Sun, a finding that we have modelled on the basis of enhanced surface thermal conductivity. From comparisons of the emitted heat radiation with radar reflectance measurements, NEOShield research has uncovered a potentially very valuable relationship between a thermal model fitting parameter, related to the surface temperature distribution, and the metal content of asteroids. Our results suggest that values of η (the so-called “beaming parameter”), are a useful indicator of asteroids with high metal content (Fig. 4). It is evident that the peak in the mean-η plot of Fig. 4a coincides with the region occupied by M-type (primarily metallic) asteroids. Comparisons with radar data (Fig. 4b) support the conclusion that η traces metal content. The fact that the peak persists after removal of the currently identified or suspected M types implies that many more asteroids with high metal content are present in the main belt, and therefore probably in the NEO population too. We have provided a list of 18 NEOs, 9 of which are potentially hazardous, for which unusually large η values are suggestive of high metal content (Table 3).

A threatening NEO containing a large amount of metal would presumably be relatively robust and massive, depending on its internal structure, factors that would require careful consideration by deflection-mission planners and/or those mandated to manage mitigation, e.g. evacuation, activities on the ground in advance of a possible impact. Moreover, the identification of NEOs with high metal content is an important task for recently-announced endeavours in the field of planetary resources. The NEOShield results imply that next-generation asteroid surveys, provided they are equipped with sensors operating at multiple thermally-dominated infrared wavelengths, could provide a valuable indication as to which discoveries warrant further investigation regarding possible high metal content.

A first main-journal peer-reviewed paper on NEOShield results in this field has been published:

Harris, A. W. and Drube, L. (2014) How to find metal-rich asteroids. Astrophysical Journal Letters, 785:L4.

The results of NEOShield investigations into the mitigation-relevant properties in the NEO population are described in Deliverable 2.1.

The successful outcome of a deflection attempt would depend on the availability of information on relevant physical parameters of the threatening NEO. While knowledge of the mass, shape, and spin vector could be sufficient for the basic design of a gravity tractor, knowledge of further parameters, such as porosity, mechanical properties, mineralogy, would be a prerequisite for the planning of an effective mitigation strategy involving impulsive options. The observational techniques with which this crucial mitigation-relevant information can be derived, considering both Earth-based facilities and dedicated space missions, have been reviewed in NEOShield Deliverables 2.2 and 2.3. Issues examined include the relevance and accuracy of a variety of observational techniques and data types, how to combine observational techniques to optimise acquisition of the necessary information, the relative merits of rendezvous and flyby missions, and how to optimise the reconnaissance strategy depending on the time available before mitigation actions become necessary. The appropriate instrumentation payloads for a reconnaissance precursor mission in an emergency scenario, and a realistic deflection demonstration mission, were also studied.

An overview of NEOShield results focusing on mitigation-related science, with a brief discussion of technology development and deflection demonstration-mission designs, has been published:

Drube, L., et al. (2015) NEOShield - A global approach to near-Earth object impact threat mitigation. In Handbook of Cosmic Hazards and Planetary Defence, Pelton, J. N. and Allahdadi, F (eds.), Springer-Verlag, Berlin, Heidelberg.

Target selection for deflection demonstration missions:

Physical properties
While the properties of the next seriously hazardous object may turn out to be completely different to those we would expect on the basis of our statistical knowledge of the NEO population, we can attempt to narrow the range of the “expected properties” to provide a rational basis for the choice of objects to serve as targets in deflection demonstration missions. In order for a mitigation demonstration to be convincing, the target object should be as realistic as possible, i.e. typical of the size and type of NEO we are likely to be faced with in the first space-borne deflection campaign.

We used statistical means to investigate the most probable frequency of mitigation-relevant physical properties of objects in the population of NEOs, such as size, albedo, composition, structure, etc. We have combined data from different published catalogues of dynamical, optical, infrared, and radar results, such as DLR EARN, the NASA Planetary Data System Small Bodies Node, the Minor Planet Center, the JPL Small-Body Database, and the new NEOWISE and Spitzer Space Telescope survey data.

Only a small number of the NEOs in the size range relevant for the selection of deflection demonstration mission targets have been physically characterised. Further observing campaigns and surveys of NEOs in the size range 50 - 300 m are essential to improve our knowledge of the mitigation-relevant small NEOs and increase the population from which targets for demonstration missions can be selected. Our results indicate that establishing a realistic probable albedo range for a forthcoming threatening NEO is difficult on the basis of current knowledge. The range observed for NEOs in the size range of interest covers all values observed for asteroids in general (0.03 - 0.7). Our results suggest that a small asteroid may have an albedo significantly higher than the average of all NEOs. However, the analysis presented in this work allows the probable albedo range of a NEO to be constrained if the taxonomic type is available.

NEOs in the size range of interest for mitigation planning may have monolithic structures or be rubble piles with some degree of cohesion provided by dust grains and van der Waals forces. Observational data and modelling results are consistent with rubble-pile structures being common among NEOs. While it would appear likely, it is not yet clear if there is a transition to monolithic structures for diameters below a few hundred metres. Further observations and modelling are required to test the suggestion that very small, elongated, fast-rotating objects are monolithic. In any case the fact that small asteroids can have very high rotation rates (up to around 1 revolution per minute) is a very important consideration for a mitigation mission. Since deflecting a fast-rotating target is likely to be technically challenging, a demo mission targeting a representative relatively fast rotator would provide a revealing test of current NEO deflection capabilities.

While extremely important for mitigation considerations, knowledge of NEO internal characteristics such as structure, density and porosity is still very primitive. A crucial question for future work remains the variety of internal structures possible in the case of very small NEOs and how they influence different deflection options. Another relevant line of investigation would be the exploration of dependencies of structure and porosity on mineralogy. For instance, how might high metal content influence the response of an object to a kinetic impactor or a stand-off blast?

The above work is described in NEOShield Deliverable 5.1.

Dynamical considerations
Simulating the orbit evolution of a target NEO after a potential deflection attempt is not only important from a mission validation perspective. Accidentally turning a previously harmless NEO into a potentially hazardous object (PHO), or increasing the impact risk of a known PHO, should obviously be avoided! Analytic estimates of the changes in an asteroid's minimum orbit intersection distance (MOID, relative to the Earth’s orbit) as a result of the deflection attempt can serve as an indicator of increased short-term collision risk with the Earth. However, given the dynamically active environment, minimum encounter distance (MED) and impact probability predictions have to be performed with tools that are capable of accounting for non-linear changes in orbital elements. The fact that knowledge of any NEO orbit has a limited accuracy has to be taken into account, in addition to uncertainties associated with the outcome of the deflection attempt itself. Thus deflection circumstances that may lead to a future collision of the target asteroid with Earth, directly or via a keyhole passage, have to be identified so that they can be avoided in the test mission design. (An asteroid may closely approach the Earth so that the perturbation by the Earth’s gravitational field is just the right amount to cause its orbit to enter a resonance condition with the Earth's orbit and impact the Earth on a later approach; the small region of space through which the NEO has to pass to enter a resonance is called a “keyhole”). Naturally, realistic simulations of impact probability changes arising from a planned mitigation can only be performed once the mission details are known.

NEOShield results show that for the kinetic impactor, target NEOs for test-mission purposes should have diameters larger than 100 m in order to minimise GNC targeting issues (it should be possible to reduce this limit depending on progress made with impactor targeting accuracy). An upper diameter limit of around 350 m is set by the need to be able to measure the very small change in the target NEO’s orbit as a result of the deflection attempt.

Given the physical and dynamical property constraints, five potential targets for deflection demonstration missions have been identified with the help of the list of realistic targets compiled as described in NEOShield Deliverable 5.2 (see Fig. 5). The list gives known physical properties relevant to deflection test missions and indicates important uncertainties in the data presented. During the course of the project new discoveries were monitored and new results incorporated into the list. The potential target NEOs are: 2000 FJ10, 2001 QC34, 2002 DU3, 2001 JV1, and 1998 VO.

Extensive post-mitigation threat assessments for the proposed targets and deflection scenarios were carried out in order to address potential planetary safety concerns. The results of our work (NEOShield Deliverable 5.3a) demonstrate clearly that risk analysis has to be performed on an individual basis, as the outcome of a deflection attempt can depend strongly on knowledge of target-specific deflection-relevant parameters (the influence of parameter uncertainties on the outcome of deflection attempts is explored further in D9.6c in which the uncertainty in the target NEO mass is shown to play a vital role).

For the test-mission scenarios investigated the target asteroids 1998 VO, 2000 FJ10 and 2001 QC34 currently appear to be the most suitable candidates for a deflection demonstration mission, where the latter constitutes the “safest” option for which all the test-mission scenarios led to an increase in minimum encounter distance between the NEO and the Earth. However, our results indicate that none of the five proposed mission/target combinations can lead to a significant probability of the target impacting the Earth, compared to the background risk, regardless of the outcome of the deflection attempt.

The work described here is a very good example of how a complex iterative process involving scientific and technical aspects, in this case target NEO selection and space-mission design, can benefit from the close collaboration between scientists and engineers afforded by an international project such as NEOShield, with its diverse scientific and industrial partners.

Kinetic impactor (KI) deflection concept:
A complete demonstration of a KI mission under representative conditions has never been accomplished, although some technological building blocks required to implement a KI mission are already available and well matured through various commercial and scientific satellite projects of the past few decades. However, there are still some technical issues that need to be resolved before we can be confident of being able to successfully implement a real NEO deflection mission on a timescale dictated by nature. Further technology development is necessary, particularly in the field of the impactor GNC and the capabilities (e.g. 3D target reconstruction) of the reconnaissance (orbiter) spacecraft, as well as the high level of autonomy of both spacecraft. A full set of relevant open issues has been identified and analysed for a fully-fledged impactor mission composed of impactor and orbiter spacecraft (NEOShield Deliverable 6.1).

Several kinetic impactor demonstration mission architecture concepts have been identified, evaluated, and traded against each other, and a final mission architecture baseline selected. The final mission design targets the NEO 2001 QC34. Driving criteria for this selection were:

1. Avoid any increase in terrestrial impact threat, i.e. avoid any reduction in the NEO's Minimum Earth Encounter Distance (MED) due to the deflection action, taking account of uncertainties.

2. Allow for deflection validation with adequate signal-to-noise ratio (SNR ≥ 10).

Both criteria are fulfilled for the selected target which is an Apollo-type asteroid (Earth crosser) that has a diameter of about 240 m. The mission consists of two spacecraft, an impactor and a reconnaissance (“explorer”) spacecraft to characterise the target NEO prior to the impact in terms of ephemeris data, rotational state, surface geometry and composition. The impact itself and the ejecta produced are also observed by the explorer. Finally after the impact the explorer determines the change in ephemeris data of the NEO and thus allows a quantitative determination of the momentum transfer and the deflection resulting from the impact. Both spacecraft are launched together as a stack on a single Soyuz-Fregat vehicle from Kourou. In order to increase the momentum transferred to the selected target NEO, the launcher upper stage (Fregat) remains connected to the impactor throughout the mission. This means that the impactor (mass 340 kg) and Fregat (mass 902 kg) crash into the NEO as a composite with a total mass of 1242 kg. The impact velocity is 9.6 km/s. The impact accuracy in terms of center-of-mass offset achievable with the proposed GNC system is about 12 m, which is an excellent result considering the solar phase angle of the target NEO at impact is rather unfavourable, so that most of the NEO is in shade as seen from the approaching impactor (Fig. 6). The explorer uses three swing-bys and solar-electric propulsion for the main orbit manoeuvres and reaches the target NEO 5.3 years after launch. The impactor uses three swing-bys and chemical propulsion and reaches the NEO more than one year later than the explorer, thus leaving sufficient time for detailed characterisation of the NEO prior to the impact. The two spacecraft remain mated until shortly prior to their first Earth swing-by manoeuvre, which occurs roughly one year after launch. A detailed mission design is presented in NEOShield Deliverable 8.2.

An alternative low-cost demonstration mission - changing the spin rate of Itokawa:
A kinetic-impactor mission to change the rotation rate of a well-studied asteroid, such as (25143) Itokawa, could be an alternative low-cost approach to obtaining information on momentum transfer. One of the main goals of a kinetic-impactor demonstration mission is to investigate the efficiency of momentum transfer and its dependence on the properties of the target NEO. An option for a low-cost test mission is to target a well-studied NEO with the aim of modifying the spin rate of the asteroid (Fig. 7). A potential target for such a test would be Itokawa, studied by the Japanese Hayabusa rendezvous mission 2005 - 2007. The advantage of this approach is that only one spacecraft, the impactor, is required. A simple calculation (NEOShield Deliverable 5.1) has shown that the change in rotation rate would be measurable by means of lightcurve observations from groundbased telescopes. An accompanying reconnaissance spacecraft would be very desirable to enhance the science return, but may not be necessary if the primary goal were to measure the momentum enhancement due to ejecta. The development of a corresponding mission design ("NEOTWIST") was the subject of a supplementary “mini-project” carried out during the last 3 months of NEOShield (Deliverable 9.6b).

Gravity tractor (GT) concept:
The GT simultaneously serves to make small adjustments to the asteroid orbit and to provide the information needed for very precise tracking of the asteroid from the Earth. In many ways this capability overlaps that of the KI reconnaissance spacecraft, which is required to determine the precise initial orbit of the asteroid, to witness the impact, and then to stay with the asteroid long enough to determine its new, post-deflection orbit. In practice, a GT may be used for fine-tuning the orbit of an asteroid that has already been deflected by other means (this scenario is discussed in NEOShield Deliverable 8.5). However, a standalone GT mission may be the option of choice, depending on the circumstances, e.g. in a scenario in which keyhole avoidance is the goal and the corresponding deflection required relatively small. The mission architecture outlined here refers to a GT standalone mission:

A launch in 2026 on a Falcon-9 to the NEO 2000 FJ10 (diameter 120 - 210 m) is assumed (Deliverable 8.3). The target selection criteria were: maximise the deflection while ensuring minimum threat to Earth due to the deflection taking into account all relevant uncertainties, and deflection validation must be possible with high signal-to-noise. The spacecraft mass is ~1160 kg at launch and ~1100 kg at the beginning of the tractoring phase. Sufficient propellant is available for the GT to operate for 12 years. The approach and rendezvous phase lasts about 2 months and accounts for acquisition and manoeuvre planning in order to zero out the relative velocity between the spacecraft and 2000 FJ10. Trajectory knowledge and successive rendezvous manoeuvres are based on DSN tracking of the spacecraft, space-based tracking of the NEO, and on-board images of the NEO taken by the spacecraft optical detection camera. The spacecraft approaches the sunward side of the asteroid. The GT hovers at a distance of about 125 m from the asteroid’s surface (Fig. 8), assuming the adopted asteroid physical parameters are nominal and a fixed thruster mounting is used with a canting angle of 45°. To maximize the deflection the gravity tractor would be placed such that the NEO is accelerated either along or against its instantaneous velocity vector. It should be possible to detect the deflection on the basis of radio science after 1 - 2 years. The nominal mission duration would be between 4.5 and 6 years.

The GT demonstration mission design work included a detailed post-mitigation risk analysis. Using a state-of-the-art mission design and impact monitoring tools it was shown that deflection actions and their corresponding uncertainties have to be considered on a case by case basis to ensure that the target NEO's threat potential to the Earth is not increased by the mitigation demonstration attempt. It was found that the GT demonstration mission proposed would reduce the risk to the Earth from 2000 FJ10 provided the GT operates in a trailing position with respect to the asteroid. The advantage of the GT lies in the fact that leading or trailing configurations can be chosen in situ. This can ensure planetary safety without changing the general layout of deflection demonstration mission designs. Kinetic impactor concepts are much more rigid in this respect.

Work within the NEOShield project on the GT concept has also investigated how the performance of a GT could be improved by incorporating knowledge of the NEO shape into an algorithm that governs the tilting of the tractor ion engines to direct the ion beams away from the surface of the NEO, while maintaining optimum thrust. The concept uses a realistic asteroid polyhedron model to determine optimum engine tilt angles, instead of assuming a spherical shape. In a modelled scenario based on a fictitious potential impact of the NEO 2011 AG5 in 2040, use of the newly developed algorithm gave 35.2% more deflection of the NEO than that obtained under the assumption of a spherical object (for details see NEOShield Deliverable 7.2 and Ummen, N. and Lappas, V., 2014, Polyhedron tracking and gravity tractor asteroid deflection. Acta Astronautica, 104, 106-124).

Blast deflection concept:
Several interrelated tasks regarding the problem of using nuclear or chemical blasts for the deflection of hazardous asteroids were undertaken, as described in NEOShield Deliverable 7.3. In order to facilitate calculation of the mechanical impulse transmitted to an asteroid by a nuclear or chemical explosion, numerical codes were developed for the calculation of the thermodynamic characteristics of silicates, for solution of the transport problem of x-ray radiation, and for the estimation of the momentum change imparted by stand-off nuclear explosions. The results of past ground explosions were re-analysed to allow estimation of the momentum transfer potential of nuclear and chemical blasts. A complete thermodynamic equation of state is proposed that describes the vaporisation of asteroid analogue material. Example mechanical impulses were calculated for buried and near-surface nuclear explosions of 10 kt and 100 kt of TNT. Results were also calculated for the case of a buried TNT explosion of 0.5 t. The effects of the impulses on an Apophis-type target were calculated. An analysis of the main physical processes associated with stand-off and buried nuclear explosions, and calculations of the mechanical momentum change of an NEO resulting from such explosions, are given by Meshcheryakov et al. (2015) Estimated efficiency of the deflection of a dangerous space object. Technical Physics, 60, 26-30.

A test mission design for the blast deflection concept was carried out to provide insight into the special issues associated with this technique. The objectives of such a mission would be to demonstrate the principle of blast deflection by refining our understanding of the underlying physical processes taking place in a blast near an asteroid, and assessing the efficiency of the technique by accurately measuring the deflection produced by the blast. Asteroid 2001 JV1 was selected as the target asteroid; an analysis of its orbital characteristics and approaches to the Earth in the period 2020 - 2042 was performed and the results used to develop a mission plan and approach trajectory to the asteroid. It was found that the only practical launch window within the coming 20 years occurs around 2021. The optimum mission design, considering fuel requirements and visibility of the asteroid during approach, has a flight duration of 296 days. The scheme of guidance and deceleration on approach to the asteroid is such as to allow a programme of onboard navigation observations and a final stage of precision guidance to the target. Possible schemes of orbiting or hovering near the asteroid were studied, with preference given to the latter. After separation of the explosive charge, the spacecraft performs small manoeuvers in order to achieve a safe distance from the asteroid (~ 100 - 200 km). Analytical solutions were derived for changes to the target’s dynamical parameters depending on the direction of the velocity impulse produced by the blast; the solutions were validated by numerical computations. It was found that in most cases the largest deflection is produced by directing the velocity impulse along the direction of the asteroid orbital motion. A post-deflection period of navigation observations (angular and ranging) is included in the mission design to measure, in the shortest possible time, the change in the target’s orbit resulting from the blast. A detailed test mission design for the blast deflection concept is given in NEOShield deliverable 8.4.

Comparison of NEO deflection concepts:
The feasibility and effectiveness of a number of proposed deflection methods (alternatives to the kinetic impactor) were compared, including the gravity tractor, the ion beam shepherd, laser ablation, and electrostatic deflection. Results show that the gravity tractor, ion beam shepherd, and laser ablation all have potentially useful capability for asteroid deflection (Fig. 9). Multiple gravity tractors improve the performance of the traction and provide redundancy in case of failure of a spacecraft. Analyses of potential manned deflection missions to enable reliable positioning of an explosive device were carried out. The work indicates that such a mission could be conducted with liquid oxygen/liquid hydrogen propulsion. However, nuclear thermal propulsion could reduce the launch mass significantly. Open issues include life support systems and consumables, and radiation protection, for long duration spaceflight. The deflection concept trade-off study is presented in NEOShield Deliverable 7.5.

Global response campaign roadmap:
NEOShield personnel at DLR, CNRS, and OU have become members of the International Asteroid Warning Network (IAWN) and/or the German, French, and UK delegations, respectively, to the Space Mission Planning Advisory Group (SMPAG); both groups were recently established under the auspices of the United Nations Action Team 14 (COPUOS). We have thereby established an interface between the NEOShield project and the most prominent current international efforts addressing the impact hazard. The SMPAG will help to coordinate the technical know-how of national space agencies and other competent bodies by recommending activities in the field of the impact hazard and mitigation measures in general. Meetings of these groups present an opportunity to highlight and discuss NEOShield work and results, and suggest internationally coordinated efforts based on NEOShield experience. The interface of our work with the activities of the United Nations secures our participation in a truly international coordinated strategy for NEO impact mitigation.

The establishment of the IAWN and of the SMPAG by the United Nations represents an important step forward in the definition of a global response roadmap. The networking of emergency management agencies worldwide seems already active and well developed. However, we identified the following issues requiring further attention (NEOShield Deliverable 9.1):

- An internationally-agreed “decision making protocol” has to be implemented in order to address the asteroid impact risk while minimising the potential for misunderstandings amongst countries involved, either as active partners in a deflection attempt or as possible victims of an impact. Plans, roles and responsibilities have to be defined well in advance of a specific impact threat arising.

- The development of new technologies and the realisation of deflection demonstration missions are fundamental to reducing our reaction time, to enable an effective response even in the case of short warning times.

- NEO research and physical characterisation efforts have to be strengthened in order to fully understand how relevant parameters (e.g. composition, porosity, inner structure, etc.) influence the outcome of deflection attempts and the consequences of impacts on the Earth.

- The asteroid impact hazard should be specifically included in the prevention and preparedness programmes of emergency management agencies.

A suite of three software tools has been developed for NEO impact-risk mitigation within the NEOShield project. These tools serve as an aid in the selection of the most suitable deflection mission given the circumstances of the potential impact scenario. The software tools are:

1. NEO Impact Risk Assessment Tool (NIRAT).
2. NEO Deflection Evaluation Tool (NEODET).
3. Risk Mitigation Strategies Evaluation Tool (RIMISET).

NIRAT, the first tool enables b-plane dispersion ellipses on the date of a possible impact to be evaluated, and the presence of keyholes that could lead to future impacts to be identified. Given knowledge of the relevant NEO physical parameters, NIRAT allows the impact risk in terms of the Palermo Scale and the Torino Scale to be evaluated. The results from NIRAT are required by the following tools.

NEODET calculates the required optimal change in NEO velocity (magnitude and direction) at any given instant prior to the possible impact epoch that would shift the dispersion ellipse out of contact with the Earth. The change in NEO velocity could be achieved by means of an impulsive deflection method (one or several impacts) or by means of a slow-push/pull technique (e.g. the gravity tractor).

RIMISET evaluates how each of the possible impulsive and slow-push techniques could produce the required change in NEO velocity, and the requirements that doing so would impose on the design of the deflection mission. Each solution can be simulated to allow an assessment of its efficiency in achieving the required deflection by any of the proposed methods (kinetic impactor, blast deflection, gravity tractor, and possible combinations of these). Ultimately, it allows a quantitative evaluation of the technical requirements of the chosen deflection space mission.

The software suite has been validated with scenarios based on the NEOs 2011 AG5 and 2007 VK184 that were potential Earth impactors for some time. Both NEOs have recently been removed from the risk list as a consequence of additional observational data. The scenarios were constructed as if these objects still posed a risk for Earth, thus allowing a useful assessment of the three tools in the risk mitigation chain. The three software tools are described in NEOShield Deliverable 9.3.

A detailed NEO threat response campaign based on the 2011 AG5 scenario is developed and discussed in NEOShield Deliverable 9.5.

With regard to practical prerequisites for reconnaissance observations of hazardous objects, current capacities and the shortcomings of ground- and space-based astrometric and radar facilities have been studied. While large ground-based surveys are the most prolific contributors to NEO discovery and asteroid astrometry at the moment, smaller astrometric and radar programmes dominate high precision astrometry. ESA's Gaia mission will have a profound impact on this picture by providing an improved astrometric catalogue and high precision astrometric measurements for about 15% of the known NEO population. In order to quantify the performance of current astrometric and radar facilities in terms of impact risk assessment, we have derived analytic and semi-analytic tools to simulate the achievable orbit quality as a function of data arc and number of observations. Our results confirm that accurate predictions of impact probabilities require either astrometric data arcs of at least half a year, radar observations, space based astrometry (e.g. by Gaia), or a reconnaissance spacecraft. A significant percentage of potential discoveries from large-scale surveys are lost due to the lack of follow-up capabilities. Accurate estimates of orbit uncertainty reduction are essential for successful detection planning and validation. The detection signal-to-noise ratio (DSNR) introduced in Deliverable 5.3a offers a simple framework to quantify the necessary precision for orbit determination before and after a detection attempt. We have provided simple estimates for the observation time requirements to achieve a certain DSNR. In the future it would be desirable to have groundbased observatories located near the Earth's equator in order to increase sky coverage. Since only about 100 NEOs can be observed per year using current radar facilities, new dedicated radar stations would also be valuable assets. The considerable personnel and power requirements to reach out to distances beyond several tenths of an AU, however, would make those extremely costly. (For details see Deliverable 9.2).

Note: The software tools described in the deliverables from work packages 3, 7, and 9 were developed for project internal purposes and are not available at present for open distribution.

Potential Impact:
We consider the main results from the NEOShield project to be:

- Greater insight into the mitigation-relevant physical characteristics of NEOs and how threatening objects may respond to impulsive deflection attempts. Our work has provided an improved understanding of the ranges of relevant physical parameters, and the possible structures and compositions, of objects most representative of those likely to threaten the Earth. The results provide a basis for the selection of targets for realistic, technically and financially feasible, deflection test missions.

- A greater understanding of the importance of post-deflection trajectory analysis. The fact that knowledge of any NEO orbit has a limited accuracy has to be taken into account, in addition to uncertainties associated with the outcome of the deflection attempt itself. Thus deflection circumstances that may, even slightly, increase the probability of a future collision of the target asteroid with Earth, directly or via a keyhole passage, have to be identified so that they can be avoided in the test-mission design.

- Detailed strategies, including the most appropriate instrumentation, for the provision of vital astrometric and physical deflection precursor data from ground- and space-based reconnaissance observations.

- The identification and characterization of suitable target NEOs for deflection test missions.

- Detailed designs of deflection test missions to demonstrate our ability to deflect a threatening NEO with current technology. The results obtained from the detailed mission design studies demonstrate that such missions are feasible and suggest that with current technology the deflection methods investigated should be adequate for the most probable emergency scenarios. In addition, the feasibility and effectiveness of a number of proposed deflection methods (alternatives to the kinetic impactor) were compared, including the gravity tractor, the ion-beam shepherd, laser ablation, and electrostatic deflection. Results showed that the gravity tractor, ion-beam shepherd, and laser ablation all have potentially useful capability for asteroid deflection. Gaining experience with deflection techniques is crucial in order to maximise the probability of success of a space-borne response to a threatening object that may have to be executed at short notice. While the NEOShield project did not have sufficient funding to launch a test mission, we expect that such a mission will be carried out in the framework of a subsequent international initiative with European participation (a current example is AIDA – the “Asteroid Impact and Deflection Assessment” concept, which is under study by NASA and ESA). With the experience gained from the project, NEOShield partners are well-placed for participation in such an initiative.

- A novel low-cost concept and detailed mission study for a kinetic-impactor test mission, based on changing the spin rate of the NEO Itokawa. In its cheapest form, the concept requires only one spacecraft, the impactor, since the change in spin rate of the asteroid, and therefore the momentum transfer efficiency, can be measured via groundbased lightcurve observations.

- The demonstration that a large international team of scientists and engineers, brought together by the European Commission’s research funding programme, can work closely and effectively together to make significant advances in the complex and diverse fields relevant to NEO impact threat mitigation. The efficiency with which the team has tackled the complex issues inherent to this field has increased with time as the partners developed greater mutual understanding and respect. Resources should be made available beyond the horizons of short-term project funding to ensure the momentum built up during the course of NEOShield (and NEOShield-2) does not go to waste, but rather the work of the NEOShield partners can be continued on a long-term basis. The NEO impact hazard is a permanent problem, which can only be tackled by permanent effort.

The events of 15 February, 2013, when a superbolide exploded over Chelyabinsk (Fig. 1) just hours before the predicted close approach of the ~30-m diameter NEO 2012 DA14, have sharpened public awareness of the dangers of NEOs and led to an avalanche of press interest in the work of NEOShield.

A major impact on the project has been the enormous media interest, which has led to extra unforeseen demands on the time of a number of project personnel, in particular the Coordinator. Correspondence and phone calls with journalists, and hosting radio and TV news and documentary teams, have been a seemingly daily occurrence since the Kick-off Meeting (not to mention countless phone calls from interested members of the public). A broad selection of TV, radio, press, and internet items on NEOShield from the beginning of the project is appended to this report. In addition to NEOShield-related queries resulting in press items such as those in the appendix, NEOShield personnel are often approached by the media on related topics (e.g. the origin and nature of asteroids and comets); press items resulting from such related queries are not included in the appendix.

NEOShield’s social media activities are followed with interest by users. The project continues to have a presence on Facebook and Twitter and answers are provided to our followers’ questions and private mails. Our interaction with users organically generates more followers. NEOShield has exchanged website links with the NASA JPL NEO program and the B612 Foundation.

In terms of exploitation of NEOShield results, NEOShield has so far generated around 20 peer-reviewed publications in major international journals, in addition to many conference papers. We comply as far as is reasonable/possible, given the restrictions imposed by many major journals, with the policy of open access. In many cases major refereed journals in our field accept the parallel posting of papers to open online archiving repositories, such as; use has been made of such opportunities, depending on the policies of the partner organisations. The participation of several NEOShield partners in the international UN-sanctioned SMPAG group (see above) is already leading to discussions between SMPAG participants on the use of results from NEOShield deliverables for SMPAG tasks, such as consideration of mitigation mission types and technologies, reference mission design studies for different NEO threat scenarios, instruments and mission requirements for the characterisation of a threatening NEO, and the development of a coordinated strategy for future work on planetary defence.

Finally, the socio-economic impact and the wider societal implications of NEOShield lie in easing public concern over the impact hazard, and demonstrating that the scientific and space-engineering communities are abreast of the problem and have a good chance of successfully deflecting a dangerous NEO should one threaten the Earth in the near future.
List of Websites:

DLR (NEOShield project Coordinator): Alan Harris []
Airbus D&S, Germany (website host, supervisory interface to technical work packages): Albert Falke []