Michael J.S. Belton Belton Space Exploration Initiatives, LLC


I consider issues associated with the establishment of a national program in the United States to prevent asteroidal collisions with the Earth. I take the position that costs associated with future damage to social infrastructure rather than potential loss of life will stimulate public representatives to begin work on a system to mitigate the possibility of an asteroidal collision. With some uncertainty, there is a 0.3 percent chance of a 50-meter or larger, sized asteroid impacting United States territory in the lifetime of its current population (~100 years). I show how a probable lack of concern for this small probability might be offset by the cost of the damage that could be caused by the large energy release (>10 Megatons of TNT) on impact. I outline four conditions, focused on the interests of United States citizens, that I believe will need to be met before the start of a national mitigation program is viable. These reflect issues of public concern, feasibility, cost, timing, and security. Establishment of a public consensus on how well these conditions have been met and some modestly detailed preplanning are probably prerequisites for the initiation of a national program. I outline a planning roadmap that indicates what a national program might look like up to the point where work on a practical mitigation project directed at a specific target could begin. I also indicate how responsibilities for the task might be divided up between different government agencies. Rough estimates of the time to complete these preliminary activities (~25 yr), and a rough estimate of the cost (~$5B) are given. This paper will be a chapter in the book “Mitigation of Hazardous Asteroids and Comets” to be published by Cambridge University Press in 2004. It is reprinted with permission from Cambridge University Press.

INTRODUCTION. It is a demonstrable fact that asteroids of all sizes and less frequently cometary nuclei suffer collisions with the Earth’s surface. The impact hazard, which is defined in Morrison et al. (2002) as “…the probability for an individual of premature death as a consequence of impact,” has undergone considerable analysis with the conclusion that the greatest risk is from the very rare collisions of relatively large asteroids that can create a global scale catastrophe in the biosphere (Chapman and Morrison, 1994). In the last decade, the question of how to deal with the hazard has lead to considerable activity and advocacy on the part of the interested scientific community, and activity at government level has been stimulated in the United States, Europe and Japan (a detailed overview is given by Morrison et al., 2002). There are now survey programs to search for objects that could be potentially hazardous; there are high-level calls for increased observational efforts to characterize the physical and compositional nature of near Earth objects (e.g., The UK NEO Task Force report, Atkinson, 2000); an impact hazard scale has been invented to provide the public with an assessment of the magnitude of the hazard from a particular object; there have been considerable advances in the accuracy of orbit determination and impact probability. Nevertheless, it seems that the question of how governments should go about preparing to mitigate the hazard needs some further attention. It has been advocated, as reflected in the review of Morrison et al. (2002), that because of long warning times (decades to hundreds of years have been suggested) we should simply wait until an actual impactor is identified to develop a mitigation system for asteroidal collisions. In the mean time, or so it is presumed, surveys to reach ever-smaller objects, scientific research and exploration characterizing these objects, basic research, etc, would continue to be supported by government agencies much as they are today. Such presumptions are, in my opinion, dangerous and, unfortunately, a high priority for these activities relative to other future scientific endeavors cannot always be guaranteed. Productive programs that enjoy adequate support today may face dwindling support in the future simply because of changing national priorities and interests. In addition, waiting an indeterminate amount of time for an impactor to be found invites, at least in my opinion, neglect; particularly at the level of government. To resolve these problems in the United States an affordable and justifiable national plan is needed, which incorporates the above scientific research and exploration and that is focused on the technical goal of mitigating the most probable kind of the impact that can cause serious damage to the social infrastructure in the lifetime of the current population. Such an approach requires redefining the hazard in terms of cost rather than deaths together with a demonstration that the 2 expected cost of the plan is commensurate with the losses that would most likely be incurred in the impact. This approach also builds into a mitigation program the notion of scientific requirements. An operational mitigation system or device can still wait until an impactor is identified, but meeting the scientific requirements for that system is something that ought and, I believe, should proceed now. There are other benefits to this approach: 1) by defining this program as a technical imperative rather than a scientific one the element of direct competition with established science goals is removed – even while significant elements of the program remain scientifically productive. 2) By focusing on the most probable impacts, i.e., smaller asteroids, a process of learning and gaining experience is implied that might, unless fate and statistics defeat us, allow us to more effectively deal with the larger and less probable objects further into the future. GOALS The probability of impact appears to be random and the average impact rates of the dominant component – the near-Earth asteroids - are reasonably well known. In this paper I will consistently use impact rates estimated by a power law distribution in Morrison et al., (2002). In other recent, but unpublished, work it is pointed out that the observed rates for objects near 50 m in size may be even less by a factor as large as 2 (Harris, 2002). If these new rates are substantiated it should be a straightforward task to adjust the relevant numbers given in this paper with little change to the argument. Asteroids larger than 50 meters across, roughly the minimum size that could cause calamitous effects at the surface, collide with the Earth on average once every 600 years. This is equivalent to roughly a 0.3 percent chance that United States territory could be hit in the lifetime of its population (~100 years). With a typical relative velocity near 20 km/sec (Morrison et al., 2002) the impact will almost instantaneously release an energy of 1016 -1017 Joules into the local environment, i.e., roughly the equivalent of a 10 Megaton bomb or about half the energy that the United States Geological Survey estimates was released in the Mount St. Helens volcanic event. I have chosen to deal with objects of this size because they are the most likely impactors that present day American public officials may have to deal with. Also the effects of such natural disasters are close to the realm of contemporary public experience, e.g., the effects of the 1908 Tunguska meteor explosion over the Siberian wilderness where the blast severely affected an area of 2000 km2 of forestland are widely known (Vasilyev, 1998). Impacts by much larger objects, i.e., larger than about 1 km that can cause global scale catastrophes, will, by definition, also affect US territories whatever the location of the impact (Chapman, 2001). But these less frequent collisions occur at a global rate of about 1 per 500,000 yrs, which translates into a 0.02 percent chance during the lifetime of the current population of the United States. While I include these kinds of impacts in the argument below, it does not depend upon them. At the present time no government agency in the United States has been given the responsibility to deal with these potentially hazardous collisions. NASA exercises a mandate from the US Congress to locate 90 percent of the objects greater than 1 km that exist in near-Earth space by 2008 but has no existing authority to act if an object on a collision trajectory is found (Weiler, 2002). Given the above collection of facts, it would seem that the primary issues that confront society with respect to mitigation are: When is the best time to invest in the research and development that would make it practical to mitigate the effects of such hazardous collisions in the future? Who should be responsible? And, what is the best way to go about it? One can anticipate that achieving resolution on such issues will be a controversial task and each of the above questions could stimulate wide discussion. In this paper I will simply assume that, if the justifications outlined below hold up, most United States citizens will want their government representatives to support the development of a system that could prevent the impact of a dangerous asteroid (i.e., one greater than 50 meters in size) found on a collision course with United States territory, or a ~1 km asteroid found on a collision course with the planet at large, particularly if it were to occur during their lives. The prevention of such collisions I take to be the goal of the national mitigation program.


There is a set of conditions that I expect would have to be satisfied in order to justify the expenditure of US national treasure on an asteroid mitigation system. These conditions reflect the kinds of questions that I believe any reasonable citizen might ask before agreeing to proceed, e.g., why are such a low probability events worth worrying about? Is today’s technology up to the job? Will the result of this effort be useful to us even in the absence of a collision in our lifetimes? Will this effort to protect our lives and property create collateral problems we don’t need? I have tried to capture the essence of these questions in the following statements: 1. The public would need to view the prospect of an impact by a 50 m asteroid within the territorial boundaries of the United States, or 1 km object impacting anywhere on Earth, as a serious concern. 2. Our technical ability to create a reliable mitigation system would need to be reasonably assured, and it should be possible to build it in time to give a fair chance that the next hazardous object to threaten 3 the territories of the United States could be dealt with. 3. The net cost of creating a reliable mitigation system should be no more than typical losses that might be incurred if an impact of a 50 m object were to happen within the territorial boundaries of the United States. 4. The implementation of a mitigation system must not create more dangers than already exist. It seems self evident that the first step towards a national program would be a high-level, governmentsponsored, study of such issues. This would be followed, if warranted, by the assignment of responsibility and the establishment of a funded program perhaps along the lines of existing community recommendations. (e.g., those in the report of Belton et al., 2003). The first condition involves the perception and assessment of risk by the public. This is apparently a topic with few experts (cf. Chapman 2001) and maybe impossible to quantify. In my view, it is essentially a political issue and any assessment is almost certainly made best by politicians currently in office, e.g., by relevant congressional committees or in the administration itself. I have already noted that the impact rate for 50-meter and larger objects give about a 0.3 percent chance of an asteroid collision on US territory during the lifetime of the population. The chances that any particular location in the US would be directly affected are approximately 5000 times less. These chances have to be modified for coastal cities (where much of the population resides) since they could be seriously inundated by a tidal wave, say 5m high or greater, caused by asteroids that impact in the ocean. Ward and Asphaug (2000) have considered such impacts, but their impact rates for the most efficient impactors for this process are about six times too high relative to those in Morrison et al. (2002). Correcting for this I find the respective chances of this happening are about 0.07, 0.03, and 0.1 percent for San Francisco, New York City, and Hilo in a 100-year period. To make it clear that these are small probabilities, I note that the chance that the population will not experience the effects from a collision in its lifetime is about 99.6 percent. Such small chances are, I believe, unlikely to raise much public concern even though the threat is real. It is only when palpable knowledge of the level of destruction that a random 10 megaton explosion could cause on a particular area, e.g., the combined energy released by more than 770 Hiroshima bombs, or roughly half the energy of the Mt St Helens disaster, or roughly 10 times the energy radiated by the largest earthquake ever recorded in the US, is pointed out to the public that notice might be taken. When knowledge of this level of destruction is combined with an awareness that a reliable defense could be built for a relatively modest cost, and that some significant fraction of the costs could themselves be mitigated through productive applications to science and space exploration, then I believe there is a chance that the need for a mitigation effort now could become justified in the public mind. It is interesting to speculate on how typical individuals in the population might view these risks and trades. I would imagine that such persons would quickly conclude that an impact would be very unlikely to have any direct affect on them, their family, or their livelihood. I would expect that they would quickly lose interest and presume that if something should be done about such rare and terrible events then “someone” in government would be taking care of it. They might be surprised to learn that the “someone” in government they assumed to be taking care of things doesn’t exist and that, in fact, no one in government presently has any responsibility to do anything about it. Certainly, in the aftermath of a random 10 Megaton explosion somewhere in the United States, or a 5-meter tsunami wave inundating a coastal city, they would be both pleased at the performance of disaster relief and tsunami warning organizations but sorely perplexed by the lack of preparedness in government organizations that might have prevented the disaster. The second condition addresses whether the construction of a reliable mitigation system can be assured and whether it would be timely. There appear to be four essential elements in such a system. First, there must be an assured ability to locate and determine the orbit of the impactor with sufficient accuracy and warning time; second, it must be possible to reliably deduce the general physical properties of the impactor so that planning for a mitigation system can achieve a reliable result; third, we must have the ability to intercept it before the collision takes place; and fourth, we must have the ability to deflect or disrupt the impactor. Most objects hazardous to the earth are on near- Earth orbits (Chesley & Spahr, 2004). To reach most of the 50 m sized objects in 10 years, telescopic surveys would have to operate at around V = 25 magnitude (this is based on an extrapolation of data in Morrison et al., 2002). By comparison, the surveys that are operating today have a limiting magnitude near 19.7 mag, i.e., more than a factor of 100 brighter. These rough figures simply mean that at present telescopic technology is very far from what would be required to meet the goal of the national mitigation program. However, plans are already afoot that will push the present survey capability to a limiting magnitude of V=24 where most 200 meter objects could be found in a 10 year period. 4 The proposed Large-aperture Synoptic Survey Telescope (LSST) facility could do this if the requirement is built into the design. The implementation of such a telescope, which is at the edge of present engineering technology, has already been advocated in the reports of two independent committees backed by the National Research Council (Space Studies Board 2001, 2000a). To reach 90% completeness at V=25 in a reasonable amount of time new technological limits would need to be achieved on the ground or space based systems will be required (e.g., Jedicke et al., 2002; Leipold et al., 2002) As put succinctly by Jewitt (2000) if these, or similar, facilities are not made available: “…we will have to face the asteroidal impact hazard with our eyes wide shut.” Detection of near-Earth objects is only a part of the equation. Also essential is the capability for rapid determination of accurate orbits to yield long warning times and accurate calculation of impact location and probability. These are not minor requirements and demand extended post discovery follow-up observations (Chesley and Spahr, 2004), advances in astronomical radar systems (Ostro and Giorgini, 2004), and in computing technology (Milani et al., 2003). While the above discussion indicates that a large increase above today’s capability is called for and a considerable amount of telescope building and observational and interpretive work over an extended period of time are implied, there appear, at least in my opinion, to be no fundamental showstoppers to this aspect of a mitigation system. Time and money are the limiting factors. Detailed knowledge of the general physical properties (mass, spin state, shape, moments of inertia, state of fracture, and a range of surface properties) will be needed for any hazardous asteroid that becomes a target (Gritzner and Kahle, 2004). Just the choice of a particular mitigation technology and its operating parameters will obviously be sensitive to the physical and compositional nature of the target. Experience shows that only a few of these parameters can be deduced with any precision from Earth based observations and in situ space missions will need to be flown to determine these parameters. Since this would at least take the time needed to build, launch and to intercept a hazardous target, typically 4 or 5 years, it is possible that there will not be enough warning time to accomplish this. In such a case the mitigation system itself may have to determine some of the critical properties (e.g., shape, mass, moments of inertia, internal state of fracture…) when it arrives at the target while other properties would have to be inferred from a database of properties that has been built up as part of a more general exploration and research program. The latter will also play a crucial role in developing several new and essential measurement techniques, e.g., radio tomography (Kofman and Safaeinili, 2004) and seismic assessment (Walker and Huebner, 2004; also Ball et al., 2004) of the interior structure of small asteroids, and new ways to measure the composition and porosity of surface materials. It seems clear that an aggressive near-Earth asteroid space exploration program will need to be integrated within the mitigation program. The requirement for robotic spacecraft to intercept and to land on a small asteroid is easily within current capability and has already been demonstrated by the NEAR mission at the asteroid Eros (Veverka et al., 2001). Mitigation techniques may require more advanced capability for operations around these small, very low mass, objects as discussed by Scheeres (2004), but, again no serious impediments that could derail a future mitigation project are anticipated. Our ability to disrupt, or adequately deflect, a rogue asteroid of a particular size headed towards Earth is completely hypothetical at the present time. There are many ideas (for a summary see Gritzner and Kahle, 2004) on what should be done and there are clearly many serious uncertainties in the application of nuclear devices (Holsapple, 2004). Similar uncertainties are also latent in the application of a solar concentrator (Gritzner and Kahle, 2004). From a purely theoretical point of view it should be possible to find technical solutions these problems. However, it is clear that early in situ interaction experiments need to be done on small objects before we can be sure where the problems are and which techniques are viable. The B612 Foundation ( has been formed to address the challenge of demonstrating that significant alterations to the orbit of an asteroid can be made in a controlled manner by 2015. Success with this endeavor would also be a major landmark in any mitigation program. In summary, it would seem that we already have experience with many of the elements needed for mitigation, but that significant development, new capability, and time will be required for success. The lack of a demonstrated technique for deflection or disruption is a particular cause for concern. There are also other serious uncertainties, the chief being whether or not human activities in space (e.g., for the assembly of parts of the system in low Earth orbit, or at the target asteroid) would need to be included. This could strongly affect the ultimate cost of a practical mitigation system and therefore its viability. But overall, though there are many technical areas that need considerable investment in time and money to achieve success, there appear to be no fundamental reasons why a mitigation system could not succeed. The third condition has to do with the cost of a mitigation system. For costs to be acceptable the mitigation program costs should be comparable (hopefully less) than estimates of the cost of the 5 damage caused by the most probable kind collision, i.e., that of a 50 m asteroid, on the territory of the United States in the lifetime of the current population. The advantage of estimating costs this way is that we can deal with real examples of costs incurred as a result of damage to infrastructure that are provided by historical events. The United States is a well-developed country and has many large metropolitan areas and valuable, if modestly populated, rural areas. Even its under populated desert areas often have valuable resources embedded in them. The economic losses, mainly timber, civil works and agricultural losses associated with the 1980 Mt. St Helens event in rural Washington State (approximate energy release: 24 megatons) were estimated at $1.1 billion in a congressionally supported study by the International Trade Commission. In a metropolitan area near Los Angeles, the 1994 Northridge earthquake caused economic loss that was officially estimated at $15 billion with most of the damage within 16 km of the epicentral area, and here the energy release was far less than that which could be released by the kind of impact that we are considering. I believe that these two examples are near the extremes of the economic losses that might be incurred as a result of a localized 10-megaton event occurring at a random place within the United States. On this basis I would argue that a $10 billion cost cap to a mitigation program would not be out of line. In the planning roadmap developed below an investment of approximately $5 billion should cover the costs of the initial preparatory phase of a mitigation program with the expenditures extending over 25 years, i.e., an average funding level of $200 million /year. This is not far from the typical levels invested in major program lines at NASA today, and so the amount is not unusually large. This leaves a further $5 billion that would be available for the implementation of mitigation mission to a specific target. Providing human spaceflight participation is not needed, this is within the expected costs of other extremely large robotic missions that have been flown or proposed. My conclusion is that condition on cost can be met and that the annual budget for a mitigation program will not be too different from costs experienced in existing robotic space programs. If human spaceflight is shown to be an essential element in a mitigation system, then the cost argument made here will need to be substantially modified, The fourth and final condition has to do with environmental and civil security. Mitigation concepts that depend on even a modest proliferation of explosive nuclear devices in space or on the ground will, in my opinion, be non-starters if this condition is to be met.


Mounting a defense against a sizable incoming object from space will be a complex task. There are national and international issues that need to be resolved; there are issues involving the delegation of responsibility between civil and military authorities; there are science issues; there are political issues involving goal setting, mission scope, and cost containment; and, finally, there are environmental and civil security issues. Here I advocate a three-phase process to establish a mitigation capability that roughly separates out strategic, preparatory, and implementation functions. It is probably prudent if these are accomplished sequentially since changes in one can be expected to have large consequences for the phases that follow. The purpose of the first, or strategic, phase is to clarify the overall goal of the program, set up its scope, identify funding, and the assign responsibilities. Because of the significance of the mitigation program to the entire population, It should be initiated by a responsible entity within the federal government, either in the administration or the congress, with, presumably, expert advice from individuals and grass roots organizations. The second, or preparatory, phase includes all that needs to be done to achieve the scientific and engineering requirements on which the design of a reliable and effective mitigation system will depend. This phase begins once an assignment of responsibility is made and funds are available to proceed. It should ideally be completed before a target on a collision course is identified, but in case we are not this fortunate, it should also include an “amelioration” element that takes care of what to do if an unexpected collision occurs. The last, or implementation, phase can only be pursued efficiently after the preparatory phase is completed and a hazardous target has been identified. In this phase all of the specific requirements of a particular target are addressed and the construction, test and implementation of an actual mitigation device is carried out. To my knowledge no one has advocated beginning work on this phase at this time. It is probably the most expensive part of the work and may involve elements of human spaceflight.


I have already advocated that the goal of a national program would be to design and implement a system to negate the most probable collision threat to United States territories in the next 100 years: a 50-meter or larger near-Earth asteroid. The prime task in the strategic phase, which might take 3 - 5 years to accomplish, would be to assess this goal in competition with alternative program concepts and make a definitive selection. Identification of an approximate timeline, suitable programmatic arrangements, and an 6 adequate budget profile, i.e., a roadmap, would follow. Institutional responsibility would need to be assigned. Expert preliminary technical evaluation in the strategic phase is necessary to ensure that the goal is achievable and to obtain a better basis for cost estimation. There are many sources of advice including existing expertise within government agencies, their advisory committees, and committees of the National Academies. I have placed considerable stress on the idea that the program should start out as a national program rather than one that is international in scope. This is a matter of pragmatism rather than xenophobia. Fostering program growth from existing expertise within the national space program should be more effective and less costly than initiating a brand new top-down international effort. The program may also involve discussion and use of military assets that could be a sensitive issue if placed in an international context. Finally, it is well known that national policies and priorities change on short timescales tied to political cycles, while stable funding and a sustained effort over two or three decades is needed for a mitigation program. I believe that such stability is best obtained in the context of a national program. Cost can also be expected to be an issue in an international program. While it would be beneficial to share development costs, I would expect the total program costs to be enlarged over that of a national program in order to immediately encompass a mitigation system capable of addressing the more difficult goal of combating large near-Earth asteroids that can do global damage. With this said, it is important to recognize that the collision threat is worldwide and much expertise lies beyond national boundaries. International cooperative projects that contribute to a national program are obviously to be encouraged. For an indication of the level of international interest and direction the reader is referred to the conclusions reached in the Final Report of the Workshop on Near Earth Objects: Risks, Policies and Actions sponsored by the Global Science Forum (OECD 2003) that suggest actions that could be taken at governmental level. It should also be understood at the outset that the mitigation program advocated here is aimed at a specific technical goal and is not a scientific or space exploration program. To be sure, the program will have remarkable scientific and exploratory spin-offs, but these are not in any sense the primary goal. This is important because closely allied scientific and exploratory endeavors already have well thought out priorities and widely supported goals that should not be perturbed by the establishment of a mitigation program. This is particularly so in astronomy and astrophysics, in solar system exploration, and in space physics where goals are focused on understanding origins – particularly of life, physical and chemical evolution, and the processes that explain what we experience in space (Space Studies Board, 2001, 2002a, 2002b). It would, in my opinion, be disruptive to try and embed a national mitigation program within one of these scientific endeavors. For mitigation, a separate program with a clear technical goal is required.


This phase should include at least the following five elements: hazard identification, amelioration, basic research, physical characterization of targets, and, what I call, interaction system technology. Hazard Identification. The operational goal of this element would be to locate and determine the orbit of the next 50-meter, or larger, near-Earth object that will, if mitigation measures are not taken, collide with the Earth. This goal must be accomplished with sufficient accuracy to determine if the object will also collide on United States territory. It should also provide a sufficiently long warning time. Initially I propose to set the goal for this warning time as at least 10 years, which is the minimum time that I expect it would take to implement a robotic mitigation system that might be capable of deflecting a 50 meter object. Astronomical survey systems are expected to yield much longer warning times (~100 yr) for collisions with the Earth itself. But these warning times shrink when the impact error ellipse must fit within the area of United States territories (D. Yeomans, private communication). This is a distinctly different kind of goal from that associated with the Spaceguard survey and clearly goes far beyond it. Yet it is, in my opinion, a necessary goal if a national mitigation program is to be justified to the public. To pursue this goal, this element should contain the following components: 1) Completion of the Spaceguard survey. 2) Implementation of the Largeaperture Synoptic Survey Telescope project, along the lines recommended in the recent Solar System Exploration Survey (Space Studies Board, 2003), and a parallel development of the USAF/Hawaii PanStarrs telescope system ( to pursue a modified Spaceguard goal which will lead to the detection and orbital properties of 90 percent of near-Earth objects down to a size of 200-meters within about 10 years from the start of the survey. 3) Design and implementation of a technologically advanced survey system, or possibly a satellite project to take the Spaceguard goal down to the 50-meter size range. 4) A ground-based radar component developed from the capabilities that already exist at Goldstone and Arecibo in conjunction with other facilities (Ostro and Giorgini, 2004) to provide improved orbits for potentially hazardous objects and to lengthen collision-warning times. 5) The final component is a suitably fast computing, data reduction, orbit determination, and archival capability. This capability could be part of the 7 arrangements of one or more of the above telescope projects. To scope the size of the problem there are an estimated one million near-Earth space objects down to 50-meters in size and, using the results in Bottke et al. (2004), only about 250 of these may be hazardous to the Earth at the present time. However, there are some 210,000 objects in this population that, while not currently Earth impactors, could, through the effects of planetary perturbations, become hazardous to Earth in the relatively short term future (D. Yeomans, Private communication). In the roadmap (Figure 1) I show these projects with some overlap stretched out over a period of 25 years. It is envisioned that these telescope systems (and others available to the astronomical community) would provide follow-up observations for each other and, where possible, make physical observations. The goal of the Amelioration element is to mitigate the effects of unavoidable impacts. There are many community organizations that could fulfill this function throughout the United States and on a national level the new Department of Homeland Security would obviously be involved However, none of these organizations have, to my knowledge, been tasked on how to respond to an unanticipated impact. As the mitigation program progresses accurate warnings and alerts should become available and the newly invented Torino scale (Binzel 2000) will be used to communicate the level of danger to the public. Resources in the event of an actual disaster would presumably be allocated as is done today to provide relief from the effects of tsunamis, earthquakes, fires, and other natural disasters and not charged to the mitigation program itself. Basic Research. There is a need for a small basic research program within the umbrella of the mitigation effort that is unfettered from well-focused goals of the other components. Here a research scientist or engineer would be able to obtain funds to support the investigation of novel theoretical ideas or laboratory investigations that are related, but not necessarily tied, to established mitigation goals. Examples are investigations into the causes of the low bulk densities that are being found for many asteroids (Merline et al., 2002; Britt et al., 2002; Hilton, 2002), or the details of how shocks propagate in macroscopically porous materials are a couple of areas of current interest. There are already a number of individuals, many at academic institutions or private research facilities, undertaking such investigations in the United States who could form the core of this effort. Target Characterization. The goals of this element are two fold: 1) To obtain the information needed so that observations of a hazardous target can be confidently interpreted in terms of the surface and interior properties that are of most interest to mitigation; 2) To develop and gain experience with measurement techniques that allow characterization of the state of the interior of a small asteroid and the materials within a few tens of meters of its surface to the level of detail required for mitigation. To meet these goals the program should provide opportunities to try out novel types of instrumentation and perform detailed characterizations of the physical, compositional and dynamical properties of a wide sample of the primary asteroidal types with the purpose of creating an archive of such properties. This kind of research, of course, already has a substantial history with considerable advances in understanding spin properties (Pravec et al., 2002), multiplicity and bulk density (Merline et al., 2002; Britt et al., 2002; Hilton, 2002) for asteroids as a group and the distribution of taxonomic groups within the NEOs (e.g., Dandy et al., 2003). Nevertheless, studies of the physical and compositional properties of these NEOs are being outstripped by their discovery rate. There are three elements that should run in parallel: 1) an Earth-based observational program focused on physical and compositional characterization, including radar studies that can reach large numbers of objects and sample their diversity. Diagnostic spectral features over a broad frequency range should be sought to better characterize the nature of each object. 2) A reconnaissance program of low-cost multiple fly-by missions, similar to that advocated by the UK NEO Task force (Atkinson, 2000), to sample a wide diversity of objects and to respond quickly to particular hazardous objects so that a first order characterization of their properties can be accomplished. 3) A program of medium sized rendezvous missions that can sample their interiors, and get down onto their surfaces to do seismic investigations. I have included four of these relatively costly missions that would include ion drive propulsion and visit at least two targets each. The final component is a strong, coherent, data analysis and interpretation program. This should cut across all missions and include Earth based work. Participation beyond the membership of the scientific flight teams would be strongly encouraged. The goal here is to integrate the net experience of the entire suite of investigations and produce the most complete database available on the properties of near-Earth asteroids, a database that can be confidently used to diagnose the properties of a potential Earth impactor. Interaction System technology. This element is the most technically oriented part of the preparation phase. Here the goal is to learn how to operate spacecraft and instruments in the close vicinity of the surfaces of very small asteroids, emplace and attach devices to their surfaces, learn their response to the application of 8 various forms of energy and momentum, etc. All of these techniques must be learned (see, for example, the advice of Naka et al., 1997). Experience must be gained over the full range of surface environments that the various types of asteroids present. Experiments to test the ability and efficiency of candidate techniques to deflect and, possibly, disrupt very small, i.e., otherwise harmless, near-Earth asteroids should be done as part of this element. The history of space flight tells us that when the time comes to implement a particular mitigation device we should not trust the first time application to deliver on its promise. Much can go awry and practice will be needed. It is in this element of the plan that the necessary practice should be acquired. It is also in this element where it will become clear what, if any, role human spaceflight might play in a mitigation system. A completely robotic approach would presumably be much cheaper if, in fact, such an approach were feasible. But it is possible that human participation may be essential for the effectiveness and reliability of a mitigation system.


The goal of this phase is to safely deviate, disrupt, or otherwise render harmless a 50 meter or larger object found to be on a collision course with United States territory in the most reliable manner and at the lowest cost. This goal can be extended to the entire Earth if the hazardous object is found to be above the size that can cause global scale havoc. If the object is smaller than this critical size and not threatening US territory, the United States may still be involved in the implementation of a mitigation device, but jointly with those nations whose territory is threatened. While this goal is clearly stated, addressing it will have some subtle difficulties due to errors latent in locating the precise impact point. Locating the latter within United States territory is much more difficult than determining that the Earth will undergo a collision. It may be that the implementation phase may have to start before it is determined for sure that United States territory is at risk (I thank D. Yeomans for this insight). It will not be possible to outline a detailed plan for this phase until the preparation phase is largely complete. Nevertheless, a few essential attributes seem self-evident: 1) It would only begin when a collision threat is confidently identified. 2) It would normally, i.e., if there were enough warning time, involve many of the same components found in the preparatory phase, but with their focus entirely oriented towards the target object itself. 3) It would include the design, construction, and application of the chosen mitigation system.


Figure 1 lays out a crude timeline for the preparatory phase that shows how the different activities that have been described interlace with one another. Estimated dollar costs, without allowance for inflation, are simply based on personal experience in NASA flight programs. The timeline for the preparatory phase is presented over a 25-year period. This time span is somewhat arbitrary and could have been made shorter by increasing the parallelism of the components. However, there are practical limits to such parallelism. These include the availability of facilities and qualified manpower, as well as acceptable limits on average and peak annual dollar costs. In my experience, average costs of $200 – 250M/yr with a peak of $300 - 400 in any one year are not untypical. The profile for this plan gives an average cost of $200M/year with a peak of $610M in year fifteen. This, relatively large peak is due to the confluence of work on six flight missions in a single year. Expert consideration of this plan with more focus on costs could presumably relieve the magnitude of this peak. Hazard identification includes the remainder of the Spaceguard program, half of the LSST, and PanStarrs programs, and, towards the end of the phase, a space based asteroid survey mission (SBAS) for the smaller objects and objects in orbits that are difficult to observe from the ground. In the case of the Spaceguard program, which is underway at the present time, I have assumed that this program would continue until the LSST and PanStarrs survey are well underway. The National Science Foundation (NSF) would presumably support the LSST and part of the PanStarrs program. Also included in this component are provisions for an underlying and continuing research and analysis program. One provision (HIR&A, or Hazard Identification Research and Analysis) is focused on providing search software, archiving, orbital analysis, and related tasks; the other is support for an ongoing program of radar observations related to high precision orbital determination. I have assumed that the SBAS (Space Based Asteroid Survey) mission would be pursued on the scale of a NASA Discovery program. For the Amelioration component I have assumed that elements of the Department of Homeland Security would undertake this task for a modest cost of $1.5M per year. This includes approximately $1M/yr for research into such issues as risk control, management, disaster preparation, etc. In the unlikely event that a collision occurs during the preparation period, special disaster relief funds would need to be appropriated as is usually done for unanticipated natural disasters on a case-by-case basis. The Basic Research component is shown as equally divided between theoretical and laboratory investigations. The correct balance between these lines would have to be judged on the basis of proposal pressure. The program scope is at the modest level of $2M/y, which should adequately support some 20 independent investigations. 9 Target Characterization is broken down into four groupings: 1) A Reconnaissance mission line, which is conceived of a series of low-cost multiple flyby, impact, or multiple rendezvous missions similar to those recommended by the UK NEO Task Force (Atkinson, 2000). Its purpose is to provide basic physical and compositional data on the wide variety of NEOs that are known to exist. Based on experience with planning proposals, three targets per mission seems feasible with a new start every four years, i.e., six missions seems plausible. To lower costs, I also assume that the basic fight system will be similar in each mission with an average cost of $175M per mission. 2) An Interiors mission line consisting of three moderately complex missions with the goal of making a detailed survey of the state of the interior and subsurface of six different types of asteroids including, if possible, a candidate cometary nucleus. These multiple rendezvous mission missions are conceived of as focusing on either radio tomography or seismic investigations and would address at least two targets each. They are expected to fall near the low end of the cost range of the NASA New Frontiers mission line. 3) A data analysis line. Here the object is to encourage the larger science community (i.e., beyond the scientific flight teams) to get involved in the interpretation of the return from these missions and ensure that the data from all of the missions are looked at in an integrated way. 4) A Characterization (R&A) line which is to primarily to support Earth-based telescopic investigations, including radar, of NEOs and potentially hazardous objects from the point of view of understanding their global physical and compositional properties. The Interaction system technology component is, at present, the most poorly defined part of the preparation phase. The necessity and scope of this component is based on the discussion of Naka et al. (1997) and in the roadmap I have broken the tasks down into two broad elements: 1) Interaction experiments, and 2) Intercept technology. It is clear that this element has goals of significant complexity and will need a considerable amount of detailed pre-planning. The lead responsibility for carrying out these missions should lie with the Department of Defense, although some sharing of responsibility with NASA may be required. I have imagined that the tasks in this element could be carried out within the scope of five relatively complex missions with costs similar to those of the Interior line.


In programs of this size it is helpful to identify major accomplishments towards the underlying goal through a series of milestones. In Table 1 I list some candidate milestones showing the relative year in which they might be accomplished and the agency that would presumably be responsible. Milestone Responsibility Year Start of strategic phase Congress or Administration 1 Assignment of authority and responsibility Administration 2 Congressional approval for a new program line Congress 4 Start of preparatory phase NASA, DOD, DHS 5 Start of reconnaissance line missions NASA 5 Beginning of LSST survey (objects down to 200 m) NSF 8 Start of Interiors line missions NASA 9 Beginning of SBAS survey (objects down to 50 m) NASA 20 First demonstration of a deflection technique DOD 21 Determination of need for human participation in space DOD 21 Conclusion of preparatory phase NASA, NSF, DHS, DOD 30


I have presented what I believe is a practical approach to a national program to mitigate the threat from asteroidal collisions. It is based on a goal that addresses the most probable threat from an extraterrestrial object to the United States during the lifetime of the current population, i.e., the impact at of a 50-meter or larger near-Earth object within the territorial boundaries of the United States during the next hundred years. I propose four conditions that would need to be met before the start of a program could proceed. In essence these conditions try to balance a presumed public disinterest due to the low probability of an impact and the relatively large cost of a program to deal with it, against the typical cost of damage to the social infrastructure that might occur and the bonus in scientific knowledge that the program would produce. The program itself is constructed from three components that would be pursued sequentially. A strategic phase, which lays the political and programmatic basis; a preparatory phase, which creates the necessary scientific and technical knowledge that is needed to provide a secure foundation for the design and implementation of a mitigation system; and an 10 implementation phase, in which a mitigation system is built and flown with the goal of preventing a collision. A plan is outlined that accomplishes the strategic and preparatory phases within three decades at a modest annual budgetary level for a total cost of approximately $5 billion dollars. The final implementation phase needs to be accomplished within a cost cap of $5 billion in order for the above argument to hold. It is expected that this can be achieved with a purely a robotic system. If, however, it is determined during the preparatory phase that human presence in space is needed as part of the system, the implementation costs can be expected to be larger than are allowed by the above arguments. In developing this program, I largely downplay three important issues often associated with mitigation: an impact by comet nucleus, an asteroidal collision by an object that is sufficiently large to cause a civilization-wrecking global catastrophe, and the large number of deaths that could caused by such events. This is done simply because of the rarity of such events, and the lack of any palpable public experience of the destructive force of such an incredible events on the Earth and, finally, what I perceive as a necessity: we must learn how to deal with small asteroids before we can expect much success in mitigating a collision involving a large one. Asteroidal collisions will continue to happen and, as our society grows, will have increasingly costly consequences. I would hope that the program that I have sketched out here might be considered as a first step towards the realization of an operational mitigation system in the United States.


 I would like to thank D. Yeomans, D. Morrison, C.R. Chapman, and W. Huntress for critical reviews of an early draft of this paper.


Atkinson, H., Tickell, C., and Williams, D. 2000. Report of the Task Force on Potentially Hazardous Near Earth Objects. British National Space Center, London ( Ball, A.J., Lognonné, P., Spohn, T., Seiferlin, K., Green, S.F., and Zarnecki, J.C. 2004 Lander and Penetrator Science for NEO Mitigation Studies. In Mitigation of Hazardous Asteroids and Comets, Ed. Belton, M.J.S. et al. Cambridge Univ. Press. Belton, M.J.S. 2003. Final Report. NASA Workshop on Scientific Requirements for Mitigation of Hazardous Comets and Asteroids. A copy of this report can be found on the internet at Binzel, R.P. 2000. The Torino Impact Hazard Scale. Planet. Space Sci. 48, 297-303. Bottke, W., Morbidelli, A., and Jedicke, R. 2004. The Near-Earth Object Population: Orbits, Sizes, Impact Frequencies, and Survey Simulations. In Mitigation of Hazardous Asteroids and Comets, Ed. Belton, M.J.S. et al. Cambridge Univ. Press. Britt, D.T., Yeomans, D., Housen, K., and Consolmagno, G. 2002 Gravitational Aggregates: Evidence and Evolution. In Asteroids III. (W.F. Bottke, Jr, A. Cellino, P. Paolicchi, and R. Binzel, Eds) Univ. of Arizona Press, Tucson. 485-500. Chapman, C.R. 2001. Impact Lethality and Risks in Today’s World: Lessons for Interpreting Earth History. In Proc. Conf. On Catastrophic Events and mass Extinctions; Impacts and Beyond. (eds. C. Koerberl & K.G. MacLeod; Geological Society of America Special Paper 356), 7-19. Chapman, C.R., and Morrision, M. 1994 Impacts on the Earth by Asteroids and Comets: Assessing the hazard. Nature 367, 33-39. Chesley, S.R., and Spahr, T.B. 2004. Earth Impactors: Orbital Characteristics and Warning Times. In Mitigation of Hazardous Asteroids and Comets, Ed. Belton, M.J.S. et al. Cambridge Univ. Press. Dandy, C.L., Fitzsimmons, A., and Collander-Brown, S.J. 2003. Optical Colors of 56 near-Earth objects: Trends with Size and Orbit. Icarus 163, 363-373 Gritzner, G., and Kahle, R. 2004. Future Mitigation Technologies and their Scientific Requirements. In Mitigation of Hazardous Asteroids and Comets, Ed. Belton, M.J.S. et al. Cambridge Univ. Press. Harris, A.W. 2002. A New Estimate of the Population of Small NEAs. Bull. Amer. Astron. Soc., 34, 835. Holsapple, K., and Housen, K. 2004. The deflection of menacing rubble pile asteroids. In Mitigation of Hazardous Asteroids and Comets, Ed. Belton, M.J.S. et al. Cambridge Univ. Press. Jedicke, R., Morbidelli, A., Spahr, T., Petit, J-L., and Bottke, W. Jnr 2003. Earth and Space Based NEO Survey Simulation: Prospects for Achieving the Spaceguard Goal. Icarus 161, 17-33. Jewitt, D. 2000. Astronomy: Eyes Wide Shut. Planetary Report XX, 4-5. Kofman, K., and Safaeinili, A. 2004. Peering inside NEOs with radio tomography. In Mitigation of Hazardous Asteroids and Comets, Ed. Belton, M.J.S. et al. Cambridge Univ. Press. Leipold, M., von Richter, A., Hahn, G., Harris, A.W., Kührt, E., Michaelis, H., and Mottla, S. 2002. Earthguard-1: A NEO Detection Space Mission. Proc. Conf. Asteroids, Comets, Meteors ACM2002. ESA SP-500. 107-110. Merline, W.J., Weidenschilling, S.J., Durda, D.D., margot, J-L., Pravec, P., and Storrs, A.D. 2002. Asteroids Do Have Satellites. In Asteroids III. (W.F. Bottke, Jr, A. Cellino, P. Paolicchi, and R. Binzel, Eds) University of Arizona Press, Tucson. 289-312. Milani, A., Chesley, S.R., Chodas, P.W., and Valsecchi, G. 2002. Asteroid Close Approaches: Analysis and 11 Potential Impact Detection. In Asteroids III. (W.F. Bottke, Jr, A. Cellino, P. Paolicchi, and R. Binzel, Eds) University of Arizona Press, Tucson. 55-69. Morrison, D., Harris, A.W., Sommer, G., Chapman, C.R., and Carusi, A. 2002 Dealing with the Impact hazard. In Asteroids III. (W.F. Bottke, Jr, A. Cellino, P. Paolicchi, and R. Binzel, Eds) University of Arizona Press, Tucson. 739-754. Naka, F.R., 1997. Report on Space Surveillance, Asteroids and Comets, and Space Debris. Vol. 2 Asteroids and Comets. United States Air Force Scientific Advisory Board. SAB-TR-96-04. OECD 2003. Final Report. Workshop on Near Earth Objects: Risks, Policies and Actions. Organization for Economic Co-operation and Development (OECD), Global Science Forum. Ostro, S.J., and Giorgini,J.D. 2004. Radar reconnaissance of Potentially Hazardous Asteroids and Comets. In Mitigation of Hazardous Asteroids and Comets, Ed. Belton, M.J.S. et al. Cambridge Univ. Press. Pravec, P., Harris, A.W., Michalowski, T., 2002. Asteroid Rotations. In Asteroids III. (W.F. Bottke, Jr, A. Cellino, P. Paolicchi, and R. Binzel, Eds) University of Arizona Press, Tucson. 113-122. D. J. Scheeres, D.J. 2004. Close proximity operations at small bodies: Orbiting, hovering, and hopping. In Mitigation of Hazardous Asteroids and Comets, Ed. Belton, M.J.S. et al. Cambridge Univ. Press. Space Studies Board 2001. Astronomy and Astrophysics in the New Millennium, National Academy Press, Washington, D.C. Space Studies Board 2002a. New Frontiers in the Solar System: An Integrated Exploration Strategy, National Academy Press, Washington, D.C. Space Studies Board 2002. The Sun to Earth – and Beyond: A decadal Research Strategy in Solar and Space Physics. National Academy Press, Washington, D.C. Vasilyev, N.V. 1998. The Tunguska Meteorite Problem Today. Planet. Space Sci. 46, 129-150. Veverka, J.; Farquhar, B.; Robinson, M.; Thomas, P.; Murchie, S.; Harch, A.; Antreasian, P. G.; Chesley, S. R.; Miller, J. K.; Owen, W. M.; Williams, B. G.; Yeomans, D.; Dunham, D.; Heyler, G.; Holdridge, M.; Nelson, R. L.; Whittenburg, K. E.; Ray, J. C.; Carcich, B.; Cheng, A.; Chapman, C.; Bell, J. F.; Bell, M.; Bussey, B.; Clark, B.; Domingue, D.; Gaffey, M. J.; Hawkins, E.; Izenberg, N.; Joseph, J.; Kirk, R.; Lucey, P.; Malin, M.; McFadden, L.; Merline, W. J.; Peterson, C.; Prockter, L.; Warren, J.; Wellnitz, D. 2001. The Landing of the NEAR-Shoemaker Spacecraft on Asteroid 433 Eros. Nature 413, 390-393. Walker,J.D., and Huebner, W. 2004. Seismic investigations of asteroid and comet interiors. In Mitigation of Hazardous Asteroids and Comets, Ed. Belton, M.J.S. et al. Cambridge Univ. Press. Ward, S.N., and E. Asphaug 2000. Asteroid Impact Tsunami: A Probalistic Hazard Assessment. Icarus 145, 64-78. Weiler, E.J. 2002. In invited remarks at the NASA Workshop on Scientific Requirements for Mitigation of hazardous Comets and Asteroids. Held in Arlington, VA. Figure Caption Figure 1. The elements of the preparatory phase of a national mitigation plan are listed in the left hand column and the estimated costs to completion in the right hand columns.