Last updated: 20 August 1999

This Case Study was substantially revised and updated in July 1999. GO TO REVISED, UPDATED CASE STUDY which includes analysis of the AN10 and OX4 cases. Meanwhile, the penultimate draft remains below. The section on history of the impact hazard has been separated from this obsolete Case Study.

CASE STUDY: THE 1997 XF11 AFFAIR (obsolete)

Clark R. Chapman

Revised: 7 October 1998

Abstract. An impact by an asteroid or comet larger than 1 km in diameter (30,000 Megaton energy) occurs about every 100,000 years, or 1 chance in a thousand per century. Objects of this size could cause serious regional disasters (e.g. tsunami) and objects only slightly larger would have global environmental consequences (e.g. severe ozone loss, injections of water and dust into the stratosphere, wildfires) that might threaten the future of civilization as we know it. (Smaller impacts can create damage similar to other major natural disasters, like earthquakes and floods, but they probably account for <0.1% of such disasters.) This "impact threat" was virtually unknown until the past two decades.

The first part of this Case Study reviews the history of how this "new" hazard came to be "discovered" by the scientific community and about how knowledge of the hazard has spread to the general public in the last few years. I also review the modest efforts by the Congress, other U.S. governmental agencies, and international groups (national, international, military, and private) to deal with the impact hazard. The unusual nature of the hazard (enormous consequence, extremely low probability of occurring in our lifetimes) has presented difficulties in getting it considered along with other natural hazards, especially by agencies with responsibility for mitigation.

It now appears likely that the discovery rate of Earth-approaching objects will increase dramatically in the next decade, raising a practical issue that has not been adequately addressed: how scientists should communicate with officials and the public concerning discoveries of objects that may impact the Earth in the near future. The second part of this Case Study reviews a recent (March 1998) event in which the world was alerted to an erroneous calculation by a supposedly reliable scientist implying a >0.1% chance of an impact by a mile-wide asteroid 30 years from now. The event dominated headlines for a day before a "correction" was issued, which undercut public confidence in astronomical calculations.

Public and political awareness of the impact hazard seems to be reaching a threshold so that it can no longer be ignored. Yet profound questions -- psychological, social, legal, political, scientific, and technological -- will not be easily dealt with. As an end-member in the spectrum of natural hazards, cosmic impacts are more likely to provide us with a mirror for assessing other rare but important hazards than to become real-life disasters like those depicted in the blockbuster movies of the summer of 1998.


The Earth is subjected to a continuing "celestial rain" of cosmic debris. Nearly all of it is fine dust or larger objects that mostly burn up in the atmosphere and are of little consequence to our environment. Occasional larger (or stronger) objects survive to be found on the ground and are known as meteorites. Every couple years a meteorite causes damage, penetrates a roof, strikes a car, or is otherwise a nuisance, but such events are newsworthy only because of their rarity and unusual nature. Yet the power-law that describes the size distribution of cosmic debris has no large-size cut-off, unlike other terrestrial natural hazards. Very rarely (every few 100,000 years or so, or 1 chance in several thousand during a human lifetime) a comet or asteroid more than a mile in diameter strikes the earth with serious global environmental consequences that could well threaten the future of civilization as we know it. Even more rarely (every 100 million years or so, but it could happen this decade) a cosmic projectile 5 to 10 miles across strikes, with consequences so terrible that most species (like the dinosaurs 65 million years ago) are threatened with extinction. It is plausible that an even larger object, perhaps 25 miles across or larger will strike the Earth during our Sun's lifetime with the possibility of virtually sterilizing the surface of our planet.

This "impact hazard" by near-Earth objects (NEOs) is strikingly different from most natural hazards in two ways: (1) the potential consequences of a major impact exceed any other known natural or man-made hazard (including nuclear war), and (2) the probability of a major impact occurring in a politically relevant timescale (say, during our lifetimes) is extremely low. It is interesting, however, that multiplying the probability of a major impact by its consequences yields an annualized death rate similar to that of some of the traditional natural hazards being discussed in this Workshop (hundreds of deaths per year worldwide, see Morrison et al., 1994). The impact hazard is, therefore, the ultimate high-consequence, low-probability hazard.

Another interesting feature of this hazard, in the context of "prediction", is that this hazard is evolving from one that was almost unknown two decades ago (hence a serious impact would certainly not have been predicted) to one in which 75% of the risk could, within two more decades, be predicted so exactly that mitigation measures within our technological capability could be applied with a high degree of effectiveness. For the remaining 25%, the prediction would either be too late to mount more than partial countermeasures or there might (if a comet were suddenly to appear from the direction of the Sun) be a wholly unforecast "act of God".

In this Case Study, I outline the history of how this hazard was "discovered" by scientists. Then I delve more deeply into the recent history of how the world's political entities became aware of the impact hazard and what is, and is not, being done about the problem. (Conceivably, there are other important but as-yet-unrealized hazards awaiting discovery; perhaps this history will be instructive about recognizing them and understanding the difficulty of gaining full public awareness of them.) Then I analyze a recent "impact scare" in which an erroneous prediction of a potentially civilization-ending impact within our lifetimes made headlines worldwide before being withdrawn the next day. Finally, I address some of the policy issues that are emerging and how this unusual hazard fits into the framework of the "Prediction in the Earth Sciences" project.

For an immature hazard, which has never happened in recorded history and whose potential for happening was not widely realized until the past decade, the policy goals are different from all others addressed in this report. Sheer recognition by policy-makers and federal and international governmental agencies that a problem exists has yet to occur in a serious way. I consider that the goal is to establish an effective national and international dialog -- including accompanying research -- to meaningfully assess the appropriate level of response to a threat that is very real and very serious, if also very unlikely to occur in our lifetimes. Should the outcome of such dialog be a decision that something should be done, it would then be necessary to develop from scratch ways of evaluating the hazard and preparing for mitigation.

I should note briefly my own background and point-of-view about this issue. I am an astronomer and geoscientist by training and have long had a professional interest in both small planetary bodies (asteroids and comets) and in impact cratering of planetary surfaces. I am not, however, among the small group of astronomers who are engaged in constructing telescopes to search the heavens for threatening objects and have received no funding to study this subject. Although, as a scientist, I have served on policy advisory committees, my own participation in the policy-making process has been rudimentary (a decade ago, I was a member of the Pima County, Arizona, Planning and Zoning Commission and chaired the committee that drafted the county grading ordinance.) My view of how society should deal with the impact hazard has been shaped by the literature of risk perception in which it is argued that there is no single objective way for society to assess the relative importance of the hazards we face.

However, as a scientist, I believe that an important input to political evaluation of this hazard is the concrete scientific information we have learned about the hazard (about the frequency of impacts, about the consequences of those impacts, about the technological issues of how we find potential impactors and what we might do about them) so that policy makers are made aware of and have access to factual knowledge (and associated error bars) about the hazard. As a citizen, my personal belief is that the impact hazard is important enough to merit further study and analysis and that it is worth the cost to augment the search for potential impactors (although certainly not worth advance preparation of on-the-launch-pad defensive measures). But I also believe that some other natural hazards -- and many other hazards, mostly medical, accidental, and related to war -- deserve more urgent attention than the impact hazard.

At the present time, there is minimal funding of work on this hazard, although there are fledgling attempts in several countries, and by several international organizations (including the U.N.) to take the impact hazard into account. Neither FEMA nor the "hazards community" (i.e. the subscribers to The Natural Hazards Observer) has taken serious interest in this particular hazard, despite widespread media attention given to it in recent years, especially since the spring of 1997. Indeed, the previous Associate Administrator of Space Science in NASA deliberately chose to downplay the importance of this topic. However, a forthcoming "near-miss" of Earth by a mile-wide asteroid (predicted for the year 2028, with initial reports suggesting a serious possibility of actual impact) generated front-page news media coverage in March 1998. And two major blockbuster films released in the summer of 1998 are beginning to force more serious consideration of the impact hazard.

In an interesting way, a theme of this project -- prediction -- would seem to be a relatively minor player insofar as the impact hazard is concerned. For asteroids and comets that have been discovered, current technology can usually provide exceptionally precise predictions about when and where an impact might occur. (Astronomers have been applauded for centuries for accurate predictions of sunrise, sunset, and eclipses.) Approaches to mitigation, while preliminary in development and expensive, are inherently simpler than for many hazards, at least from the technical standpoint. And for those potential impactors that haven't been discovered (so far the great majority, though we could change that in the next couple of decades), the probability of impact is a simple cosmic lottery: our inherent chances-of-losing are readily calculated and well known. There are, indeed, uncertainties in the environmental consequences of impact -- akin to those of other meteorologically-based hazards, like storms, global warming, the ozone hole, and El Ni¤o -- but the sudden drama and exceptional character of an impact event itself would minimize the perceived importance of such uncertainties. In this case, the major predictive uncertainties involve questions of how individuals, the society, and political institutions may prepare for and conceivably respond to predictions of such a horrific but unlikely disaster.

There are other issues of prediction, however, which are relevant. They are analogous to those affecting other Case Studies addressed in this project, although I will continue to argue that they are relatively less important. The erroneous prediction of a possible impact issued on March 11, 1998, which made banner headlines around the world, illustrates the problem. The error -- public assertions by supposedly credible astronomers that there was a "small" (interpreted by others as >0.1%) chance that the Earth would be hit in the year 2028 by a mile-wide asteroid (1997 XF11) -- was corrected a day later; revised predictions were reported by a press somewhat disenchanted with the astronomers' credibility. On the other hand, such temporary errors, occasioned by an urgent press looking over the shoulders of naive astronomers (who should not have alerted the press until they knew what they were doing), was hardly as serious as forecasts of earthquakes that never happened (after counties had spent millions in preparation) or failures to forecast hurricanes that did happen. The mis- forecast impact only involved changing the odds from "very unlikely" to "zero", and no public funds had begun to be spent toward unnecessary mitigation of an event 30 years in the future. Nevertheless, astronomers dare not appear to be Chicken Little and lose credibility in an arena in which they conceivably might someday have to forecast an event that would deserve to be taken seriously at the highest public and governmental levels.

There are important policy issues that must be addressed. One is how the scientific community should advise officials and the public about the chances of an impact occurring shortly after discovery of a potentially threatening body, but before it can be proved that the object cannot hit. (This problem, articulated in 1997 drafts of this Case Study, played out dramatically during the early days of the March 1998 media frenzy over 1997 XF11.) Another is to clarify how a plethora of national and international laws (e.g. the National Environmental Protection Act and the Outer Space Treaty) might be interpreted in terms of mounting a defensive infrastructure to deal with asteroids or, in the unlikely event that an object were found to be on an imminent collision course with Earth, to mount emergency countermeasures. Will national and international responses be driven primarily by legitimate public concerns? Or will they be torqued by the self-interested motives of those entities that would stand most to benefit from mounting large scientific or military responses to the threat? Finally, how does a society responsibly decide on the relative allocation of funds to address, on the one hand, mundane, everyday problems that kill millions (e.g. diseases and accidents) or instead to deal with rare but horrifying disasters (biological/chemical warfare or terrorism, exceptionally destructive earthquakes, or even asteroid impacts)?


The impact hazard made unprecedented inroads in public consciousness worldwide during the last half of the second week of March 1998. The first apparently "official" prediction of a significant probability of impact during our lifetimes by a dangerous asteroid gained headlines, often banner headlines, around the globe. A day later, the headlines expressed relief as newly found observations were reported to yield a miss-distance twenty times farther away and an absolutely zero chance of impact. While news media stories of the affair involving the asteroid numbered 1997 XF11 were generally non-critical toward the astronomers who made the predictions, there was an undercurrent of cynicism about the initial hasty and erroneous warning. As NPR commentator Daniel Schorr remarked on March 22, discussing the asteroid event as well as a recent premature story about supposed terrorists in Nevada carrying anthrax, "couldn't they have waited a few days before scaring us half to death?" Behind the scenes, I witnessed (and participated in) a complex and sometimes bitter exchange of e-mail between asteroid researchers trying to assess the data concerning 1997 XF11 and trying to deal with the associated media frenzy. As it turns out, the newly found data merely provided an excuse to retract the warning: proper analysis of the original data, accomplished by several independent researchers within hours of the announcement, demonstrated that the chance of Earth impact was essentially zero and that the original report was premature and erroneous. By late April, out of the public limelight, it appeared that all parties involved in the affair were moving toward agreement that the original report was erroneous and supported steps to avoid having a similar event happen in the future. Unfortunately, final closure remains elusive and more impact scares are likely to occur, further undermining public confidence in astronomers' predictions.

In this section, I will outline how the 1997 XF11 affair unfolded and discuss some lessons learned in its immediate aftermath. The events are only six months old, so the eventual influence of the affair on how society chooses to deal (or not) with the impact threat remains to be assessed. One conclusion is clear: space scientists, unfamiliar with societal impacts of their research, are far less well organized than other disciplines (e.g. weather prediction, earthquake prediction, and evaluation of nuclear waste dumps) to evaluate predictions in the context of how they may be perceived by users in the society. Steps toward making improvements appear to be nearly paralyzed as of this writing (October 1998).

Daily Log of 1997 XF11's Brief Fame

Discovery of 1997 XF11. 1997 XF11 (hereafter XF11) was discovered on December 6, 1997, by Jim Scotti, observer with the Spacewatch program in Arizona. It was the brightest NEO discovered by Spacewatch -- a program focussed on smaller and fainter NEO's -- in more than a year. Follow-up observations by Japanese amateur astronomers permitted calculation of an approximate orbit for the asteroid with a MOID (minimum orbital intersection distance with Earth) of close to zero, making it the 108th known PHA (potentially hazardous asteroid: those asteroids for which MOID <0.05 Astronomical Units, or 7.5 million km [one AU = 149,600,000 km, the mean distance of the Earth from the Sun]); the MOID parameter and PHA's were defined by Ted Bowell, of Lowell Observatory, and adopted by Brian Marsden, who directs the International Astronomical Union Minor Planet Center (MPC) in Cambridge, Massachusetts (Marsden, 1997). That classification placed XF11 higher on observers' priority lists, but astronomers remained unaware of any near-term close approach to Earth for several months, for reasons discussed below. (The MOID parameter is a generalized measure of the potential for an object to approach the Earth at some time in the future. MOID's can change by a few hundredths of an AU over time, which provides the rationale for the definition of a PHA.) Marsden has since said that, by February, his private orbital calculations and extrapolations indicated that XF11 would pass by the Earth very roughly twice as far away as the Moon in October 2028.

Observations during the first week of March by Peter Shelus of the Univ. of Texas provided the lever arm that Marsden used for a somewhat better orbit determination and for better orbital extrapolations into future decades. Although Marsden had been previously criticized (for example, by reviewers of his NASA funding proposal) for not releasing such observations promptly and despite requirements by NASA funding officials that he do so, Marsden continued his usual procedure of embargoing data like those of Shelus until his monthly publication of aggregate data. Therefore, while Marsden's group at the MPC worked on the data themselves, the full March dataset was not available -- in any practical way -- to other researchers interested in calculating orbits for NEO's. (Shelus was unaware of Marsden's policy of embargoing data until JPL researchers Don Yeomans and Paul Chodas approached him for his data after the March 11th announcement of XF11's potential close approach; at the meeting of NEO researchers on March 17th, Shelus expressed incredulity about Marsden's failure to serve as a prompt world clearing house for such data.)

Announcement of impact possibility. Without consulting or checking with any outside scientists, Marsden published an International Astronomical Union Circular (IAUC) (#6837) during the early afternoon of March 11th, announcing that XF11 would pass just 0.00031 AU (Astronomical Units) from the Earth on Oct. 26, 2028, thirty years hence. The miss distance from the center of the Earth would be just 46,000 km (29,000 miles, or 25,000 miles above the Earth's surface). As an indication of the error in the estimate, Marsden wrote that an approach within 0.002 AU was "virtually certain"; that distance is significantly less than the distance to the Moon. As an indication of the significance of the prediction, the usually restrained Marsden included an exclamation mark after the 0.00031 AU prediction. (Five weeks later, on April 18th in IAU Circular 6879, Marsden explained that the "virtually certain" statement was in error and that the miss distance could be, in fact, 10 to 15 times larger.)

The IAU Circulars (distributed by e-mail) are the chief way that the world's astronomers learn about rapidly changing events in the skies, requiring urgent observational follow-up which can't wait for more leisurely publication in scientific journals. Supernovae, X-ray bursters, comets, and asteroids are among the subjects of these publications, which are available via the Internet and mailed (in postcard-like format) to observatories around the world. The Circulars are the latest form of the official announcement service of the International Astronomical Union which began as telegrams many decades ago. Marsden's stated reason for including a future close-approach prediction on the Circular was to motivate observers to make new observations of XF11 (and, implicitly, to search for any unreported past observations) in order to refine the orbit and, hence, the prediction.

Marsden was well aware that his words would be read by many people other than the targeted astronomers. The Circulars are normally available by subscription, and subscribers include science journalists and interested amateur scientists, among others. Marsden had previously learned that unusual predictions (similar to XF11's close approach to Earth) would cause his telephone to ring with inquiries from reporters. So, in this case, Marsden prepared a kind of press release (he has called it a Press Information Sheet, or PIS) prior to release of the Circular. In the PIS, Marsden reported that "the chance of an actual collision is small, but one is not entirely out of the question." He described the miss distances as ranging from "scarcely closer than the Moon" on the one hand, to "significantly closer than 30,000 miles" on the other. He went on to describe the "splendid sight" the object would make, as viewed from dark skies in Europe, on the evening of closest approach in 2028. [2]

[2] Many of these statements later proved to be grossly incorrect. The actual miss-distance is more than 3 times the erroneous "virtually certain" maximum distance originally mentioned by Marsden. Far from being a "splendid sight," 1997 XF11 it will require a telescope to be seen in 2028. Marsden subsequently admitted that his alarming words about chances of impact were written hastily and are "perhaps...somewhat unfortunate."

He made explicit what was only implicit in the Circular, asking for prediscovery observations during several particular years, which he listed. Marsden did not release his statement simultaneously with the Circular. But within half-an-hour of the posting of the Circular, astronomer Stephen Maran, press officer of the American Astronomical Society (AAS), telephoned Marsden, and Marsden gave him the PIS. Maran immediately, and without further investigation, e-mailed the PIS to a list of science journalists maintained by Maran for the AAS.[3]

[3] A distinction of subsequent importance to Marsden's relationship with the International Astronomical Union (whose leadership was highly critical of him during the days following the original announcement), under whose mandate his Minor Planet Center operates, is whether the PIS was released as part of MPC business or not. Marsden says that he did not indicate his IAU affiliation on the PIS but that Maran added that affiliation in the AAS press release. In a practical sense, the distinction makes no difference since Marsden's credibility in such matters derives from his IAU mandate, whether he is issuing a Circular, a press statement, or giving an interview based on a Circular or a press statement.

Much subsequent comment has questioned the advisability of Marsden issuing the PIS. But the logic for the PIS is sound, once an IAUC has been released. According to Maran, both the Dallas Morning News and the New York Times began writing stories based on the IAUC before receiving the associated PIS. Maran points out that many other news media learn about important stories from a story list put out on the New York Times news wire long before publication of the first edition of the Times, so the story was destined for prominence. And the Washington Post was informed, prior to the PIS, by an amateur astronomer "news- tipster" prior to receiving the PIS. The problem was clearly the release of the incorrect IAUC in the first place.

The Swift-Tuttle precursor. Being on travel away from my office, I was not in ready contact by telephone or e-mail on the afternoon of March 11th. I first learned of the prediction of possible impact during late afternoon while watching the ABC evening newscast from my hotel room. Many of my colleagues who are experts about asteroids, as well as NASA officials, first learned of the prediction when they were telephoned by reporters and asked to comment. Their predicament of being "on-the-spot" but with no first-hand information reminds me of a disturbingly analogous situation -- again instigated by Marsden -- in autumn 1992. Astronomers had been searching for a long-period comet, named Swift-Tuttle, and Marsden had been working on observers' positional observations, just as for XF11. While talking with Boston Globe science reporter David Chandler, Marsden made a back-of-the-envelope calculation that the comet would have a 1-in-10,000 chance of colliding with Earth during a close approach early in the 22nd century. Soon thereafter, Marsden repeated his estimate to the New York Times.

Unaware of any predictions about Swift-Tuttle, I was caught by surprise when the Times telephoned me at my office. Assuming that Marsden's calculations, as relayed to me by the reporter, were correct, I commented on the possible implications of an impact. My remarks were included in the prominent story published in the Times announcing Marsden's prediction. I felt embarrassed when Don Yeomans, of JPL, subsequently showed that Marsden's calculations were in error and that the Earth was in no danger of being hit by the comet. Marsden was slow to admit that Yeomans' interpretation was correct, and the continuing argument between Marsden and Yeomans was highlighted in a subsequent cover article in Newsweek entitled "Doomsday Science." Most experts believe that Marsden's calculations regarding Swift-Tuttle were simply wrong (in essence, he made a one-or-two-dimensional calculation of what is, in reality, a three-dimensional situation), but Marsden has never fully admitted to his error. The episode is recounted by Steel (1995).

Follow-up on XF11. I attribute my gullibility in 1992 to my belief that Marsden is one of the world's experts in orbital computations of small bodies combined with my perception that he is a careful and conservative scientist. People, like the press, have short memories, so I expect that few of my colleagues remembered the Swift-Tuttle episode as they responded to reporters' questions about XF11 on March 11/12. I do not know what my colleagues said to the press during the first hours of inquiries, but their concurrent e-mail illustrates what they were thinking. Taking Marsden's quoted miss- distance and estimated error (many times the miss-distance) at face value and assuming (falsely, as it turns out) that the Earth would therefore be within the targeting error ellipse, a simple calculation suggests that the probability of impact is at least 0.1% (this estimate was distributed via e-mail to nine colleagues by Alan Harris, of JPL, within 2 hours of the posting of the IAU Circular); many other scientists undoubtedly did the same elementary calculation for themselves (Goldman, 1998) -- it is the straightforward implication of Marsden's announcement, even though he never stated a quantitative impact probability himself.

Since the size of XF11, based on its observed brightness, is about 1 or 2 km (roughly the size of a civilization-threatening impactor) [4],

[4] Until detailed telescopic measurements are made of the object, the diameter could be anywhere between 0.5 and 3 km, but 1.5 km is most likely.

a 1-in-a-thousand chance of the end-of-civilization occurring just 30 years from now surely seemed to be an awesome possibility, deserving of being taken very seriously. Richard Binzel, of M.I.T., considered the threat to measure 3.5 on his proposed 5-point impact hazard index (Binzel, 1997) -- higher than he had ever imagined would be registered on the scale. On the other hand, the chance of one of the ~1000 so-far- undiscovered asteroids as large as XF11 colliding with Earth in the next 30 years, is about 0.01%, only ten times less than the (incorrect) apparent collision probability of XF11. So by that measure, the predicted impact was not all that unlikely...what was remarkable was that we had found the threatening object before conducting the Spaceguard Survey!

Don Yeomans and his colleague Paul Chodas at JPL had not forgotten their previous conflict with Marsden over Comet Swift-Tuttle, and they approached Marsden's announcement with skepticism, immediately seeking the original data so that they could make an independent assessment. They requested that Marsden release the data, which he did after about two hours. (In the meantime, they had calculated an effectively zero probability of impact from just the available data through early February.) Within 15 minutes of receiving the March data from Marsden, Chodas made his own orbital calculations from the full data set and continued to conclude, unlike Marsden, that the chances for collision were zero ("that's zero folks," Yeomans emphasized in e-mail sent to numerous colleagues that evening, 5 hours after Marsden's original circular). While the Yeomans/Chodas calculation placed the close approach about twice as far away from Earth as Marsden's calculation, the nominal miss- distance was still unusually close. More significantly, their quoted error was 4 times larger than Marsden's. (The actual miss distance that was eventually calculated is consistent with the JPL error bars but not with Marsden's.[5])

[5] It was unclear just what Marsden's quoted error meant. Usually, scientists in this specialty quote "3-sigma" errors, meaning that there is a 99.7% chance that the correct answer will fall within the error bar. Marsden used the words "virtually certain" to refer to his original estimate that the asteroid would come within 80% of the distance of the Moon, but he later suggested that those confident-sounding words referred to a "1-sigma" error, meaning only 68% chance. Even as a 1-sigma error, it appears inconsistent with the eventual miss-distance of the object at 250% of the lunar distance. Five weeks after the event, Marsden agreed that calculations allowed miss-distances up to 10 to 15 times as great as the 80% of lunar distance.

Yeomans and Chodas, however, immediately emphasized a point overlooked (or at least not mentioned) by Marsden: the error ellipse is extremely skinny, more nearly a line than an ellipse (more than 1000 times longer than it is wide), and it entirely misses the Earth. Indeed, when Chodas attempted to calculate the probability of impact, the number underflowed his computer, meaning that it was less than 1 chance in 10 to the 300th power (as close to zero as you can get)! As it turned out, Chodas and Yeomans were somewhat confused by a plotting-program error when they tried to interpret their calculations: the program plotted the line-like error ellipse perpendicular to its actual direction. It was, therefore, very confusing when later calculations (made after inclusion of the prediscovery data of Helin, see below) fell far away, in a perpendicular direction, from the original error ellipse. The error had the effect of somewhat exaggerating XF11's apparent minimum possible miss-distance from Earth, but it did not affect the "zero" estimate of impact probability.

When reporters began telephoning the following morning, March 12th, few space scientists were armed with the information that I by that time had at hand, thanks to my being included among the 11 recipients of Yeomans' message. (Two of Marsden's colleagues called on him late on the 11th to distribute, via a new IAU Circular, the Yeomans/Chodas probability = zero calculation so as to help astronomers deal with the media frenzy, but Marsden rejected the suggestion.[6])

[6] Marsden has since justified his refusal by pointing out that the IAU Circular never explicitly mentioned the possibility of collision (which was in the associated PIS), so it would have been inappropriate to place the correction in a Circular. That seems to be splitting hairs.

Reporters calling me early on the 12th, to ask what we might do about the threatening asteroid, were unaware of the JPL result so I suggested that they talk with Yeomans. Later in the morning, reporters told me and others that Marsden was sticking by his original estimate that the asteroid might impact, even after he was informed of what Yeomans had told them. I personally spoke with Marsden mid-day on the 12th and implored him to telephone Yeomans in order to compare notes; he refused.

By early afternoon, Eleanor Helin (of JPL's NEAT program) reported that she had located prediscovery observations of XF11 on films taken at Mt. Palomar in 1990. The much-longer time baseline would permit a much more accurate calculation of the 2028 encounter circumstances. The observations were received and used by both Yeomans/Chodas and Marsden/Williams to fold into their calculations. Late in the day, Marsden issued a new IAU Circular and the JPL group reported to the press essentially the same thing [7]:

[7] Marsden has complained bitterly (e.g. in his March 29th Boston Globe op-ed essay) about the J.P.L. press release that was issued late on the 12th. On the one hand, he criticizes J.P.L. for trying to "jump the queue" and announce the correction before Marsden could make his own announcement. On the other hand, he complains that J.P.L. failed to give Marsden's assistant Gareth Williams credit for being the first to calculate the new miss-distance based on the 1990 data. The latter complaint is based on the fact that the J.P.L. press release and a Yeomans e-mail were issued about two hours after his IAU Circular, notwithstanding the fact that the Chodas/Yeomans calculations had been made hours earlier, and reported by them to the press during constant interviews since that time. (It is unseemly, some feel, for Marsden to be making claims of scientific priority for results based on observations he receives in his IAU role as disseminator of international astronomical data.) Clearly the larger, and more important, problem was Marsden's failure to acknowledge, accept, and report the conclusions of his colleagues (like Chodas and Muinonen), which were available the previous evening or during the morning of the 12th, to the effect that there was virtually no chance of impact (so that the original announcement was incorrect).

the asteroid will miss the Earth by nearly a million km, about 2.5 times the distance to the Moon.

"No Impact" aftermath. The following morning's newspaper headlines told the story of the Earth's escape from cosmic doom. The New York Post's treatment was emphatic: "Kiss Your Asteroid Goodbye!" Another Post story headlined "NASA Needs a 'Crash' Course in Math," and continued that the "Doomsday figures were way off [the] mark." Domestic and international news reports tended to emphasize that "NASA scientists" (meaning JPL researchers Yeomans and Chodas, using data from JPL observer Helin) had corrected the work of "the International Astronomical Union" (meaning Marsden), notwithstanding the fact that all are funded by NASA.

Headlines in the March 23rd issues of Time and Newsweek illustrate the general tone of media reaction to what one labelled a "Cosmic False Alarm". "Oops, Never Mind!" "For a day, it looked like we could all be toast as an asteroid hurled through space. Then astronomers double-checked their figures." Many media outlets noted the fortuitous promotion of the Hollywood movies "Deep Impact" and "Armaged- don", due for premier within the ensuing few months. A few conspiracy-minded reporters persistently attempted to uncover illicit connections between Marsden and the Dreamworks producers of "Deep Impact." A St. Louis radio talk-show host felt he had been unfairly used by Spacewatch astronomer Tom Gehrels, who had appealed for NEO funding during a March 12th live interview. While most astronomers felt that the media had generally handled the coverage accurately and benignly, there is little doubt that astronomers lost some of their vaunted reputation for unassailable predictions as a result of the XF11 affair. Let's look more closely at some elements of the impact hazard illuminated by this episode.

Despite Marsden's IAU Circular 6879, issued in mid-April which can be parsed into an obscure acknowledgement that the March prediction was mistaken, Marsden has continued a face-saving campaign of attempting to rationalize his involvement in the March scare. In a series of essays contributed to the Internet newsletter CCNet Digest, Marsden has vehemently objected to assertions that there never was a chance of impact by XF11 based on the data he used for making his prediction. Indeed, he has identified actual impact trajectories, in 2037 and 2040, that are within the observational error bars of all data from 1997 - March 1998; he has even asserted that a 2028 impact remained possible after inclusion of the 1990 data. These scenarios are regarded by experts as contingencies, within an inherently chaotic dynamical context, of wholly negligible probability -- rather like threading a succession of microscopic needles. Marsden has continued to discount the relevance of impact probability calculations and to assert that his pursuit of prediscovery observations is what saved the day. Adopting this view of history, several analyses of the interactions between science and science journalists have cited the XF11 affair as a case where science essentially "worked," rather than a case of a wholly erroneous prediction that never should have been made in the first place (cf. Gladstone 1998). The July/August 1998 Skeptical Inquirer (Gardner 1998) and the July 1998 Astronomy (Gordon 1998) are just two magazines that continued to promote the conclusion that "science worked" in the XF11 case and that "old photos" of XF11 saved the day. As late as September, Brian Marsden was still providing sound bites on national radio, blaming the Internet for the XF11 affair.

Papers presented to the annual AAS Division of Planetary Sciences meeting (Madison, WI, Oct. 1998) have confirmed the results of near-zero impact probability calculations initially made within a few hours to a few days of the March IAU announcement. For instance, according to Muinonen (1998a, 1998b), data available as early as December 22, 1997 (within 2 weeks of XF11's discovery) were sufficient to calculate a 2028 impact probability <10^(-42). Much smaller probabilities are derived when data through February are properly analysed. Of course, no probability of a physical event can be exactly zero, so that Marsden's contingent impact trajectories are theoretically possible. But they are so close to having zero probability that they cannot rationally be regarded as relevant, and certainly not newsworthy. The bald fact is that the March prediction was erroneous and never should have been made. What is equally disturbing is that nobody correctly calculated the essentially zero impact probability that could have been calculated with available software, from data Marsden published on his website, at anytime during the two-and-a-half months from December 22nd until Marsden's faulty announcement. The only person to examine these data before March was Marsden, himself. According to one expert who has discussed the XF11 prediction with Marsden, Marsden "could not rule out an impact using the techniques he had at his disposal." Those who did have the appropriate software apparently never examined the publicly available data. Why not?

Reconsideration of the XF11 Affair

Why the Urgency? Within hours of his original announcement of 1997 XF11's predicted encounter, Brian Marsden began receiving what grew into a storm of criticism from his colleagues for "crying 'wolf'". As in the Swift-Tuttle affair earlier in the decade, Marsden had made a sensational public announcement of a possible impact -- soon proved to be wrong -- without cross-checking his work with his colleagues. Scientific tradition usually demands that publication of results (a predicted impact is surely such a result) be subjected to peer-review, although the Circulars have traditionally been peer-reviewed only in the sense that Marsden's office evaluates data before publishing them in the Circulars. One could readily forgive Marsden for his haste if the prediction demanded an immediate response. But, with 30 years to go, no urgency was obvious. Especially after the "learning experience" of Swift-Tuttle, Marsden's colleagues were unforgiving (e.g. in e-mail and in comments directed to Marsden at the March 17th meeting at the Lunar and Planetary Institute) of his haste in going public. A few of Marsden's fellow orbit-calculators were especially angry that Marsden had censored the data he received as an international data center and used them in a private research effort without allowing them an opportunity to make independent calculations (words like "stealing the intellectual property" of the observers were used). A NASA official who oversaw the funding request from the Minor Planet Center was irate that Marsden had continued to ignore NASA's mandate that he make positional data available immediately to the astronomical community.

Brian Marsden's defense, expounded in spirited e-mail to which I was copied and at the March 17th meeting, is as follows. The purpose of the IAU Circulars is, generally, to announce astronomical events and phenomena that require rapid follow-up. Sudden events among the stars are rare and urgent so that follow-up observations within hours to weeks can be vital. In the case of near-Earth asteroids, newly discovered objects may become much fainter within hours or days, so urgency in informing observers is important. (The problem is compounded by the tradition of assigning observing time on telescopes typically half-a-year in advance.) The urgency in the case of XF11 was not so extreme because any new observations during spring of 1998 would not substantially reduce errors, but it was fading and within three months would be approaching the sun, and hence "dusk", making observations increasingly difficult.

Since XF11 seemed to be important (Marsden described it as being the most interesting asteroid he had seen in over 40 years of calculating orbits), Marsden wanted to encourage observers to make follow-up observations. His usual tool for informing observers is via the IAU Circulars which he regularly issues. One might ask why he felt it necessary to mention the eye-catching 0.00031 AU miss distance on the Circular. Marsden's answer is that there are many potentially interesting asteroids for observers to check and he needed to highlight the reason for raising XF11's priority. In hindsight, it is obvious to most of his colleagues that a more appropriate procedure for an inherently sensational situation bound to be reported publicly would either (a) to have been less explicit about his reasons for urgency (just saying that is was an "very interesting" asteroid would have bought him hours or possibly days) or (b) to have telephoned the half-dozen most likely observers who could make (or might have already made) additional observations. But Marsden was following his normal habit, and he issued the IAU Circular.

Why did he also write a press release (and why did the American Astronomical Society distribute it, without checking its validity )? This question is largely irrelevant, since reporters subscribing to the IAU Circulars got the story anyway. But the logic follows immediately from what I've just said: if reporters would get the story anyway, Marsden reasons, then it is best that he prepare a document (Press Information Sheet) that tells his expert version in clear English rather than allowing them to mis-report the technical data in the IAU Circular.

Marsden's defense for not immediately releasing the positional data to the astronomical community is that it takes more work, hence more manpower and more funding than he is receiving. His competitive colleagues respond that the funding he receives is sufficient (especially in the current technological environment) for him to electronically distribute the data he receives instantly, or daily, but that he is spending his efforts, instead, on private research that is inappropriate for an international data- dissemination center. Marsden, who takes an accountant-like approach to his subject, is adamant that the diverse data from around the world require sifting and evaluation so that only "reliable" data are distributed to the community. Marsden's opponents retort that he should report all data for their independent analysis, and -- if he wants -- report the edited and quality-assured data in separate files [8].

[8] An example of what is wrong with Marsden's secretive policy was provided by Andrea Milani, of Pisa. Early on March 12th (Italian time -- still middle-of-the-night in America), he found the IAU Circular about 1997 XF11 in his e-mail and immediately tried to do his own calculations. But the data available on Marsden's Web site included none after February 4th. Based on the partial data, Milani found possible orbits in 2028 that passed far beyond the Moon, but also some passing close to the Earth. Milani, who had been asking Marsden to make the international observers' data available for some time, was thus unable to fully check Marsden's result. And news media were phoning him, as well.

Indeed, if the data need to be quality-assured, it would seem best to have independent people doing it, as well.

Several NEO researchers, primarily observers who provide Marsden with positions, have defended Marsden on the grounds that the IAU Circulars were intended to alert observers and Marsden has done just what he is mandated by the IAU to do. In the context of the themes of the Prediction Workshop Case Studies (e.g. "who becomes empowered when [a] prediction is made?"), it is worth noting that these observers are highly dependent on Marsden continuing to perform his traditional role. It is through him that their observations are integrated into something important (asteroid and comet orbits). Indeed, one of the chief rewards for such observers is having a comet named after its discoverers, and it is Marsden who makes the judgements about whose names are to be given to the comets (lately Marsden has been de-emphasizing such rewards). I also think it is noteworthy that several of these observers have been the most prominent advocates, in Internet chat, of the view that Marsden should be praised, rather than criticized, for the XF11 announcement because the higher public visibility will result in increased funding of the observers' programs. (Despite the obvious self-interest of many of these observers, I believe that they are sincere in their belief that the impact hazard is a serious threat and that it is in the public interest to increase the survey efforts.)

An element of the view that Marsden's actions were traditional and appropriate is the common ivory-tower view that it is not the responsibility of scientists if the media listen in on our traditionally-open technical discussions and blow them out-of-proportion. (This view would have a purer basis if Marsden had released only the IAU Circular and not the accompanying PIS.) This is not, however, the accepted view of ethicists who study the interaction of scientists with the media and the public. Scientists are obliged to be aware of the social context in which their work is done, and it is our responsibility to ensure that we minimize the chances of sensational misreporting of our work.

My own view is that Marsden's un-checked public announcement (the original IAUC) represents a serious failure of judgement. While he may have acted in normal and habitual mode, his own admission that XF11 seemed to him to be the most interesting asteroid in decades is sufficient reason for him to have treated it as a special case. Given his previous error involving Swift-Tuttle, Marsden cannot be forgiven for being naive; XF11 is not his first "learning experience." He was negligent in not waiting for a day or two to verify his calculation. He was aware that several of his colleagues had different, independent, computer programs for calculating orbits and impact probabilities. As exemplified by Chodas' immediate calculations upon receiving the data, analytic calculations can be concluded within an hour and the more reliable Monte Carlo simulations can be done within a matter of hours to a day-or-so. Therefore, if Marsden had disseminated the data to his colleagues, contrary conclusions would have been reported back to him within hours, as indeed they were after the data finally were made available. Moreover, Marsden should have realized the high probability that an asteroid as bright as 1997 XF11 would have been recorded (even though unnoticed at the time) in earlier photographs, which he could have sought in a less public way; such prediscovery observations were, in fact, found and made available for inclusion in the calculations, within less than a day [9].

[9] Still other prediscovery observations were located in the films from the Shoemaker program within hours of Helin's discovery of her own 1990 images of XF11.

So a day's delay would have surely negated any need for a public announcement and XF11 would have been relegated -- outside of the glare of public scrutiny -- to its proper place as a relatively large, but otherwise unremarkable member of the 100-plus other known "potentially hazardous objects".

Technical confusion. Specialists in calculating the orbits of asteroids and comets normally have weeks, months, or even years to do their work. The urgency of the XF11 case, exacerbated by Marsden's public announcement, resulted in the necessity for unusually hasty work. Since it was done under the spotlight of the media, the conditions were all-the-more prone to result in temporarily erroneous conclusions. I was party to a voluminous e-mail correspondence (involving hundreds of messages) during the first four days following Marsden's announcement, which sheds light on how the conclusions unfolded. No one was immune from making is human nature to err and scientists operating outside the realm of normal methodology and traditions are bound to err as well, and they did. I believe that scientists facing similar situations should be especially restrained in their public pronouncements, as I counselled (via e-mail) during the early hours of the affair. In particular, they should be aware that potential errors of interpretation or judgement might be much larger than what are known as formal estimates of uncertainty. (This is a case of the "meta-error bars" discussed in this Prediction Workshop, in September 1998.) They should also be aware that hasty avoidance of the traditional requirements of "checking one's math" and peer-review demands much restraint in reporting of one's preliminary results. Banner headlines in the New York Times should be reserved for more dispassionate and well-checked conclusions, barring true urgency.

Marsden's original report ("virtually certain" to be closer than the Moon) was seriously in error; in addition, the words he used poorly represented his results. Marsden at first claimed that his calculations were technically correct but admitted that his words may have been "unfortunate". Later he confessed that he had initially misunderstood the implications of the technical calculations. Estimates of miss-distances by others, using the same data once they were made available, were comparable to Marsden's, but they correctly quoted larger error bars and much smaller (near-zero) probabilities of impact. Apparently the early calculations yielded closer miss distances by simple random chance (the new Shelus data could have resulted in the nominal pass anywhere along the skinny error ellipse, but just happened to move the solution close to the closest tangent of the ellipse to Earth). Perhaps the particular calculations were also influenced by the non- uniform nature of the data-sets available for analysis. Observations of asteroid positions are subject to a variety of systematic errors (e.g. in the positions of the reference stars), and they are distributed peculiarly in time (due, for example, to the difficulties of making observations when the Moon is bright). Moreover, due to the informal international network of observers, which includes individual amateur astronomers as well as professional enterprises like NEAT and Spacewatch, it can be misleading to apply formal statistics in order to evaluate uncertainties in the orbits calculated from such disparate data. Yet scientists routinely report "formal" estimates of errors, occasionally multiplied by an arbitrary factor to take account of unforeseen errors. In the case of XF11, it appears that the errors calculated by calculators (other than Marsden) were sometimes a bit too small [10].

[10] There was original suspicion that Marsden's calculations of 1997 XF11 were in error, for example because of the incompatibility between his statement that it was "virtually certain" to pass within 80% of the lunar distance and the final reality. Marsden later admitted that he may have used ill-chosen words to describe his results, but continued to say that there were no errors in his calculations or in his interpretations of them. The IAU General Secretary, Johannes Andersen, also stated that the accuracy of Marsden's calculations had not been challenged. But, in fact, one of Marsden's international colleagues complained that the "MPC [has] never published its algorithms, and never had its software inspected and certified from external experts." So doubts remained. On April 18, Marsden issued a new IAU Circular acknowledging that he had misinterpreted his calculations and that there never was any chance that 1997 XF11 would impact Earth. Later he seemed to back away again from this admission.

The real disagreement, however, was over the possibilities for Earth impact [11].

[11] The approach of 1997 XF11 was news only because of the chance of impact, not because the asteroid might (although we now know it will not) become a bright object in the sky 30 years from now. The media naturally think (as do scientists who aren't thinking clearly) that "how close" the asteroid might come to the Earth is a proxy for chance of impact, which it is not. This is another reason why the 1990 data did nothing to change the meaningful facts. The 1990 data changed the miss-distance a lot but they provided no significant additional assurance that it would not impact.

Had Marsden checked with others making similar calculations, he would have at least been aware of different opinions, in particular the "that's zero folks" conclusion of Yeomans. Lack of consensus among his colleagues (had such conclusions been available) would have given Marsden a powerful excuse to delay public announcement.

I immediately objected (in e-mail messages sent to colleagues, including Yeomans and Marsden) that the "that's zero folks" conclusion of Yeomans was as misleading as Marsden's over-dramatization of the impact probability. I felt that it was too soon for Yeomans to be sure that he and Chodas hadn't made a mistake and that the word "zero" when used by a mathematical scientist has a degree of precision that was premature, no matter what the formal errors were calculated to be. And, indeed, Yeomans was laboring under somewhat false impressions at the time. While his colleague Chodas's initial calculation turned out to be much closer to the truth than Marsden's, his interpretation of it was clouded by the error in the plotting program, which caused an erroneous visual representation of the situation. Prior to Chodas' new calculation based on Helin's 1990 data, other members of the very small community of asteroid orbit experts had made their own estimates of collision probabilities. Karri Muinonen, of Helsinki, reported a maximum probability of impact in the range 0.000004 to 0.00002. The fact that none of the other experts had yet settled on a probability as small as exactly "zero" was evidence to me that it was premature to announce "zero" to the media. NASA official Carl Pilcher used more appropriate common English words in stating simply to the media that the asteroid was "not going to hit."

There were several reasons for technical quibbling about exactly how close to zero the impact probability might be. Analytical calculations, for instance, use a "linear approximation" that can go awry under some circumstances. Monte Carlo simulations, which take longer to do, soon showed slight departures from the early analytical work. Because of the particular circumstances of XF11, the errors due to the linear approximation were totally negligible, but (until closely studied) it might have been otherwise and it wasn't known with certainty that the linear approximation applied exactly until those results could be compared with the Monte Carlo results. In addition, Marsden emphasized the possibility that several circumstances not included in the formal orbital models (e.g. passage of XF11 near another large asteroid before 2028, or comet-like "non-gravitational" accelerations) made the "exactly zero," 10^(-300) impact calculations suspect. These were quibbles, however; in the informed judgement of all experts, these technical details were of such small effect that they couldn't affect the validity of an "almost zero" chance of impact (i.e. far less than background impact probability, hence not "newsworthy").

Unfortunately, the arguments over whether "exactly zero" was a correct description of the probability of impact obscured the more relevant question of the probability relative to the background probability of an unknown asteroid impacting during the next 30 years. Although discussion continues about exactly what probability of impact by a known object should be considered newsworthy, probabilities computed even prior to inclusion of the 1990 data were so much smaller than background that the "news" (if any) should have been that one of the ~1800 undiscovered Earth-approaching asteroids larger than 1 km diameter was discovered and found to be in an orbit that will not hit the Earth in the foreseeable future. (Marsden has quoted an early March 12th e-mail from Muinonen as justification for ignoring the "that's zero folks" conclusion of Yeomans and continuing to tell the press that the asteroid might hit. But Muinonen had only agreed with Marsden that the impact probability was not exactly zero; in fact, he told Marsden that the probability of an impact was extremely small, which Marsden should have taken as an independent reason to withdraw the spiralling news story.)

The technical confusion became truly extreme once the data from Helin yielded the new result that XF11 would pass about 2.5 times the distance of the Moon. Not only was this result seriously at odds with Marsden's conclusion that it was "virtually certain" to pass well within the Moon's distance, but it was dramatically (indeed impossibly) outside of the mis-oriented error ellipse plotted by Chodas. It wasn't until Sunday, four days after Marsden's initial circular, that Chodas finally deduced, and reported to his colleagues, that there was an error in his plotting routine; when it was plotted with the correct orientation, the new miss-distance, and its associated error ellipse, fell well within the original, larger ellipse. Although Chodas and his colleagues are very confident that their programs that calculate collision probability are reliable, one must always be cautious of a small possibility that "bugs" might exist in those programs, too. Chodas and Yeomans tested their analytic solution with Monte Carlo simulations that confirmed their result. Muinonen's independent Monte Carlo calculations (based on algorithms described by Muinonen & Bowell [1993]), as well as the later calculations by Marsden's assistant Gareth Williams and others, proves that the more distant miss-distance is very nearly correct. But there was some confusion about that for several days due to imperfections in communications. For example, Muinonen, being located in Finland, was half-a-day off-schedule compared with his American colleagues, impeding timely communication despite use of the Internet.

Uncertain Orbits and the Hazard Scale. Several asteroid experts have been concerned about the process that evolves from first discovery of an NEO to its eventually certain prediction that it will not (or unluckily will) hit the Earth. Bowell and Muinonen (1992) were the first to present a might-be scenario. This was amplified by Chodas & Yeomans (1997) as well as by myself in the first draft (July 1997) of this Case Study. The basic problem is that initial observations of an NEO permit only crude estimates to be made of its orbit. Forward extrapolation of an orbit might sometimes lead to a prediction of an encounter with the Earth; an early prediction will necessarily have a huge uncertainty, because of the incomplete dataset obtained over a short time interval, but an impact with the Earth can't be ruled out. Subsequent observations will inevitably reduce the size of the error ellipse and the Earth will, almost always, move outside of the shrinking ellipse. In a highly unlikely case, the ellipse could shrink and yet include the Earth, which would raise the unexpected scenario of a very close approach or even an actual Earth impact.

1997 XF11 presents an excellent example, although more insights can be obtained by looking at other "what if" scenarios based on variations of the XF11 case. As it actually played out, there were five stages in the process of improving its orbit and predictions of impact: (1) In late December, two weeks after, discovery observations were sufficient for Marsden to list 1997 XF11 as a PHA (potentially hazardous asteroid); (2) in February, Marsden noted (to himself) the fairly close approach in 2028; (3) just prior to his March 11th announcement, based on early March data, Marsden concluded that the 2028 approach might be very close; (4) a few hours after Marsden's Circular early in the afternoon of March 11th, others (using the same data) demonstrated that the probability of impact was essentially zero, and (5) the following afternoon, additional data were found that Marsden took to nail the lid on the coffin of any possibility of impact.

Until Marsden's work in early March, nobody (including Marsden) took 1997 XF11 very seriously, despite its comparatively large size and its December listing as a PHA. Indeed, the paucity of reported observations during February was one of Marsden's motivations for his March 11th Circular, designed to raise XF11 in observers' priorities. One can ask what the results of the computational algorithms would have been if they had been applied in December, January, and February? (And one could ask why such computations weren't done: the answers, presumably, are the lack of manpower, lack of practical access to the data, and distraction of researchers -- not well funded, if at all, to follow each PHA closely -- by other obligations.) Was the Earth ever within the error ellipse in 2028, and if so, how did the probability of impact compare with the background? If computations had been repeated as new observations were reported, how did the error ellipse shrink and what were the associated impact probabilities? One can do thought experiments about how Marsden and the scientific community might have acted, or should have acted, had they been paying such close attention to XF11.

The answers to some of these questions are now available. Paul Chodas reported to the April meeting of the Division for Dynamical Astronomy of the American Astronomical Society that his re-analysis of the 26 observations of XF11 available as early as December 24th yielded a near-zero impact probability. The highest impact probability that ever would have been calculated by his algorithm subsequent to December 24th, as data trickled in, was a 2x10^(-30) probability, based on the 65 observations available as of January 10th. Karri Muinonen (1998a, 1998b) has shown that more comprehensive calculations than those of Chodas, using the early data only, lead to collision probabilities even closer to zero! In another hypothestical, analogous, but highly improbable case, such early data -- if properly analyzed at the time -- might yield a significantly non- zero chance of impact so that a call for more observations (or prediscovery data) should be issued then, rather than weeks later.

Marsden and some of the NEO observers take the position that the answer to any uncertainty in the status of a potential close approach by an NEO is acquisition of new data (and, surely, new data never hurts). So some of them argue that all the debates are moot about whether the three months of data available about XF11 in March were sufficient to rule out an impact: the correct thing to do, they say, was to get more data and make sure. Unfortunately, as we have seen, there are unintended but inevitable consequences of calling for more data (at least doing it Marsden's way): scaring the world half to death, to paraphrase Dan Schorr. In future cases, circumstances may make it difficult to respond with data in a timely way -- there may be no prediscovery observations and the vagaries of weather and the position of a rapidly moving NEA may put off successful acquisition of new data for months or longer. So it is vital to understand the uncertainties associated with calculations based on partial datasets as the orbit of a new PHA is gradually improved. A lesson that should be learned from the XF11 affair -- a lesson obscured by the common media story that the prediscovery observations were responsible for the corrections -- is that if the choice is to pay observers to make follow-up observations or to pay for better orbital calculations from existing data, the latter would be much more effective.

A major problem confronting astronomers is how to report to the public discovery of a new NEO that can't yet be ruled out as potentially impacting the Earth in the foreseeable future. While it may usually happen, as in the case of XF11, that the errors can be reduced and the impact ruled out quickly, it is possible that there will be some cases where the ambiguities may persist for months or even years, depending on asteroid observability. The probabilities of impact may be low during such a period of ambiguity, but so long as the probability isn't essentially zero (as it actually was for XF11), the situation would have a different tone from the purely statistical argument that an as-yet-undiscovered object might impact next year with a one-in-three-hundred-thousand probability. We would now have a case of a real, known body that might strike -- with some probability -- at an exact date in the foreseeable future (Oct. 26, 2028 in the case of XF11).

Richard Binzel, of M.I.T., thought it would be useful to abstract the detailed complexities facing astronomers (e.g. the inability to observe the object to improve its orbit for many months, if it happened to be near the Sun in the sky) in terms of a simple numerical scale, like the Richter scale used to measure earthquakes or other scales used to label the severity of such natural hazards as hurricanes. He proposed (Binzel 1997; see also Verschuur, 1998) a 5- point scale, and associated evaluation, that is based on two vital measures of a specific impact threat: probability of impact and size (and hence potential energy) of the impactor. Clearly, there is a much higher level of concern if the threatening object is a few km across than if it is a small meteoroid that will explode harmlessly in the upper atmosphere. Another attribute of a potential impact, not included in the scale proposed by Binzel, is also surely important: when the predicted impact will occur. Obviously, the seriousness and urgency of responding to the prediction will depend on whether the impact will occur in a year, a decade, or a century in the future.

Despite the obvious merits of adopting such a Hazard Scale as a uniform way for the scientific community to communicate the seriousness of a prediction to the public, it has not yet been adopted -- either officially by NASA, the IAU, the UN, or by any other entity, nor even just pragmatically by those serving as interfaces between the astronomical community and the public (Brian Marsden, the AAS press officer, the NASA press office, etc.). Probably the reason is because the idea is relatively new and there hasn't been time to resolve differences of opinion about just how the hazard scale should be defined and implemented. For example, Brian Marsden seems to oppose a Hazard Scale. Others quibble about its definition. For instance, should the scale involve the integers 0 through 5, or should it be a decimal number (e.g. allowing 2.3)? Some have proposed that the scale be normalized to the mean background probability of impact, so that the scale records a value above nominal only if the chances for a particular impact exceed the chances for a random impact by an unknown body. Others worry that a given asteroid will confuse the public by having an ever-changing value on the scale, as it evolves toward 0 or (conceivably) 5. In that sense, it would be unlike the Richter scale, which records the actual magnitude of an earthquake and is a fixed number for a given event. Of course, predictions of hurricanes and other evolving natural disasters (like probabilities of heavy rains in an already flooded region) change as circumstances change and predictions become more certain.

In the early hours of the XF11 affair, Binzel opined (in e-mail directed to a few colleagues, not in public, so far as I know) that the event was a 3.5 on his scale, an unexpectedly high value, translating between "unlikely" and "possible" and falling between a category of "events which should be carefully monitored" and "potentially threatening events." As it turned out, the probability of impact that he fed into his calculation of the hazard index (0.1%) was itself faulty, as explained above. This example calls into question, if not the hazard scale itself, at least the basis on which it should be computed and reported to the public. (Binzel's original assumption was that the probability of impact would be calculated correctly, as it was not in the case of XF11.)

NASA and Scientific Community Reactions. Thanks to the efforts of Carl Pilcher (acting director of NASA's Solar System Exploration program), NASA was better prepared than it might have been to deal with the XF11 emergency. Pilcher had already recently raised his own consciousness, and that of several officials who worked for him, about the NEO impact hazard. And he had arranged for the scheduling of a program review of the NEO search activities to be held on March 17th at the Lunar and Planetary Institute, in association with the annual Lunar and Planetary Science Conference. Nevertheless, NASA officials were caught unaware by the media calling to inquire about NASA's response to XF11. For Brian Marsden, a NASA fundee, his failure to inform NASA prior to releasing the March 11th Circular was a major mistake.

Pilcher took advantage of the already scheduled program review to augment the list of invited panel members and he expanded the agenda to include an evaluation of the XF11 affair. Pilcher attended parts of the March 17th meeting and stated that "this will not happen again." He mandated that the group (about 20 panelists and presenters, mainly comprising NASA-funded asteroid researchers, including Marsden and Yeomans, but also including the Italian president of the Spaceguard Foundation, Andrea Carusi) report within a few days interim recommendations about how future cases should be dealt with.

During a lunch break, food was brought in and the group worked up a series of recommendations for Pilcher to take to Wes Huntress (NASA Associate Administrator for Space Science) and perhaps to NASA Administrator Dan Goldin and the White House. The outspoken and unrepentant Marsden made a take-it-or-leave-it offer to make certain observations available immediately to the astronomical community; when others present objected that he instead should be required to make all observations available immediately, Marsden refused to compromise and the group finally agreed to procedures on an interim-only basis, pending an early workshop to delve more deeply into the issues. (Six months later, no such workshop has been scheduled; there are preliminary plans for it to take place in Italy in May 1999.) It was also agreed that public announcement would follow (rather than precede) cross-checking with other experts on orbital calculations and that NASA officials would be informed 24 hours before any future public release (also, numerous other scientists and officials would be personally alerted to the release simultaneously with the release, so that they would be less likely to be caught uninformed by the media). Some attendees remained skeptical that Marsden would abide by them, especially after his subsequent (March 29th) op-ed essay in the Boston Globe (and a longer version distributed in an Internet newsletter), describing some of the guidelines as "smacking of censorship".

In early April, NASA issued a slightly revised set of the guidelines, despite some international apprehensions about a free-flow of information. NASA officials expressed willingness to coordinate with international efforts, but wanted preliminary guidelines implemented immediately.

Secrecy versus Peer Review. In extensive Internet chats in the aftermath of the XF11 affair, several commentators worried about possible "secrecy" and urged that scientists immediately make impact calculations public. I find this position to be plainly specious and wrong, but it deserves to be taken seriously for it is likely to resonate with conspiracy-oriented members of the public who are unfamiliar with the scientific process. And traditional methods of ensuring credibility of scientific results are undergoing revision in the face of the realities of the Information Age.

The world has long been exposed to the uncensored "babble" of anyone who wants to say something and has access to a printing press. Yet society, and the scientific community, have developed ways to deal with it, to sort out the wheat from the chaff. With the advent of the Internet, the babble has increased to a roar and new methods may need to be developed. The traditional way that the scientific community has "certified" reputable results has been by having them checked, peer-reviewed, and published in technical journals. Also, by self-regulation (sometimes imposed by editorial policies of journals), scientists don't go to the public media until after the date of publication of the peer-reviewed technical article.

Of course, such procedures don't prohibit lay people, amateurs, truthsayers, fortune-tellers, pseudo-scientists, and professional scientists who don't care about their reputations from publishing their hasty results immediately -- on the Internet, or anywhere -- and such stories are the grist for supermarket tabloids. But serious people, opinion leaders, policy makers, and the like rely on those media that attempt to abide by society's self-regulating procedures in order to report stories with a higher degree of reliability. Just a few days before the 1997 XF11 affair, there was Internet chat of a claim by some unknown foreign scientists that the asteroid Icarus might hit the Earth in the year 2006, but the false story did not create banner headlines around the world presumably because the source was suspect and because such "official" sources as the MPC did not endorse the prediction.

The Minor Planet Center represents itself to be, should be, and was taken (by the media) to be a reliable source of information, representing the astronomical community. So astronomers as a whole lost some credibility when failures in the peer-review process at the MPC led to the XF11 scare. It is appropriate and necessary that serious astronomers, and the serious media, adapt to the modern realities of the Internet and establish effective procedures -- which are functionally like peer-review, whether mandatory or voluntary -- to reduce the chances that mistakes like 1997 XF11 will happen again. Entities like the IAU, NASA, the Spaceguard Foundation, amateur and professional astronomical societies, funding agencies, etc. (national and international) should adopt procedures to ensure that centuries of traditional peer-review procedures are maintained in the current Information Age. Conspiracy theorists -- like those who believe that NASA is suppressing information about the "Face on Mars" -- will complain. But that is a small price to pay for ensuring that scientists continue to retain the amazingly high level of credibility that they currently hold in the public's eye. There can be no possible public benefit from releasing incomplete, premature, unchecked scientific results prior to cross-checking and review.

More serious concerns, no doubt, will be generated by additional requirements -- like the interim guidelines of NASA -- that select people or organizations be informed before (say 24 hours) the rest of the public is informed. There is nothing new about scientific organizations requiring peer-review before publishing research results [12].

[12] It does not seem right that an internationally mandated organization, like the MPC, should be subject to NASA reporting requirements. Unfortunately, the IAU withdrew its modest funding of the MPC and NASA is the only agency to step in. Other nations should step up to their responsibilities.

And one can readily understand not only the institutional desire but also the potential public benefit of NASA officials being given a heads-up about a public announcement before the press calls on them unawares. Yet one can also appreciate that such procedures may not be trusted by the wary public or by non-Americans, so the competing responsibilities and opinions need to be balanced. It would be far better to work these issues out in advance, rather than to rely on arbitrary, ad hoc responses to an evolving crisis. Other hazards evaluated at this Prediction Workshop have evolved effective means for analyzing predictions prior to having the predictions officially promulgated. Such approaches should be evaluated by the space science community; examples include the work of the Nuclear Waste Technical Review Board and the California Earthquake Prediction Evaluation Council.


Conclusions Regarding the Impact Hazard

The impact hazard raises various issues involving the interface between scientists and public policy. Some of them are similar to issues presented by other natural hazards, although the specifics are often different because of the unusual character of the impact hazard. The first part of this Case Study has described the 18-year history during which the impact hazard went from being a science fiction tale to a documented element of our reality, to which the public has been widely exposed. However, very little has been accomplished in integrating this topic into the existing private and governmental structures that deal with natural hazards and their mitigation.

Although I am a principal in this topic, I can count on the fingers of one hand the number of inquiries that I have dealt with from entities involved in hazard mitigation and analysis. In the early 1990's, I did receive an inquiry about the impact hazard from a German insurance consortium (Munchener Ruckversigherungs-Ges.) that deals with analysis of natural hazards. I also had one telephone call from a FEMA official. In late 1996, I learned that the American Geophysical Union had a committee making recommendations about hazards and I was able, at the last minute, to get language into the adopted resolution that included the impact hazard.

There are significant NEO issues that the hazard community must face. One concerns the elements of predictability, and how predictions can be communicated effectively to relevant officials and to the public. As of mid-1998, 122 of the estimated 2000 Earth-approaching asteroids larger than 1 km diameter have been discovered plus 10 Earth-approaching short-period comets. That means that more than 90% of asteroids large enough to potentially threaten civilization have not been discovered; in addition, virtually all comets that might strike the Earth have not been discovered. Should the Spaceguard Survey be undertaken, that situation will change over the next two decades to one in which most of the potential impactors will have been surveyed and it will have been reliably determined whether or not (almost certainly not, but if so, then when and where) any of them will impact Earth during the next century. Even if the Spaceguard Survey is not formally undertaken, the rapid advance in astronomical instrumentation -- including that available to amateur astronomers -- combined with the augmented interest in this topic is likely to accomplish the Survey's objectives in the next half century.

During this time of augmented discovery, there could be many "false alarms". This could result from the fact that the process that discovers a new Earth-crossing object could take a significant period of time before the orbit would be reliably determined to be on a collision course or not. Even though the odds are strongly against the possibility that even one such large object will be found to be in an orbit that will actually strike Earth within the next century (less than one chance in a thousand), one could be found that will appear to have, for a period of days (conceivably years), a non-zero (even if very low) probability of collision with the Earth. Although such a scenario is unlikely for truly dangerous objects, the Spaceguard Survey will find many smaller objects for every >1 km object that is found. Indeed, the size of such objects may be uncertain for some time, making it likely that possible impacts might be reported in the media before orbits or sizes of the objects are refined.

In addition to the possibility of reliable error ellipses temporarily including the Earth, the Swift-Tuttle and 1997 XF11 case illustrate that well-intentioned experts may, through errors, announce a potential danger even when it does not actually exist. Less reliable people, who nevertheless are -- or represent themselves to be -- genuine scientists, frequently make erroneous predictions. There was a case several years ago of a well known French astronomer being reported in normally reliable news media as predicting that the asteroid Toutatis might collide with the Earth. I've already noted that, early in the same week as Marsden's predictions concerning 1997 XF11, there were faulty reports that the asteroid Icarus might collide with the Earth during the next decade.

The higher rates of detection certainly have the potential to yield many more "near misses" and potential future impactors than we have seen in the past. Because of competition among different observing programs and among different orbit-calculators, it is plausible based on past experience to expect announcements to be made to the press that are poorly qualified. And then, of course, the media often mis-report even accurately qualified scientific announcements (a recent example was a report on CNN in the last week of July 1997 that an asteroid 3/5ths of a mile across would "destroy the Earth" -- I checked into it, and it resulted from oversimplification of a longer, more accurate CNN story based on a scientifically factual press release by E. Helin).

I know of no efforts, even at a preliminary level, to establish formal communica- tions links between potential discoverers of NEO's (professional and amateur astronomers in the U.S., Japan, France, Russia, and other countries, as well as Air Force and other military detection programs) and public and military officials who might be called on to respond to potentially threatening impacts. Fortunately, it is much more likely that a threatening object will be found decades or centuries before impact (as exemplified by XF11's 30-year advance warning) than that it will be found hours or weeks before impact. But, especially if the size is poorly constrained or if exaggerated attention is given to small objects, imminent impacts could well find themselves thrust into the laps of unprepared high government officials. The fall of Skylab and the more recent (1996) crash of a Russian would-be Mars probe in the southern hemisphere are reminders that even minuscule debris from space can create a sensation. It is reliably known that the White House was alerted in 1994 when a rather modest sized atmospheric impact (10's to 100's of kilotons) was detected by downward-looking surveillance satellites (this is the event described technically by McCord et al., 1995). Exploding meteors, with yields approaching a megaton, enter the upper atmosphere from time to time, and are very spectacular even if they do little or no damage on the ground. Indeed, with widening media interest in the topic, even modest impacts -- of rocks that yield meteorites on the ground -- are sometimes generating major media reports (these common fireball or bolide events are spectacular, but widely misinterpreted by non-professional observers).

Yet another kind of uncertainty relates to how government officials and the public would react to a genuinely near miss. For instance, there is roughly a 5% chance that an object at least 1 km in diameter really will, sometime in the next century, pass as close to the Earth as Marsden originally reported for 1997 XF11. And a couple times a decade, an object large enough to cause a Tunguska-like event (were it to hit) passes within the same distance. While astronomers would consider these to be comfortably safe encounters, it is not certain that political leaders and the military would take such a dispassionate view. Efforts might be mounted to "do something" (expensively or even dangerously) about such close approaches, despite astronomers' assurances.

It is quite possible that there already have been significant tsunamis in Earth history caused by impacts into the ocean, but whose causes have been unrecognized. From data given by Toon et al. (1997), it seems plausible that every thousand years, or so, an impact (probably by an iron projectile with kinetic energy of about a megaton) would cause a tsunami equivalent to the biggest tsunamis recorded during the 20th century. It is distinctly possible that warnings of impact-caused tsunamis could be prepared, but the infrastructure for doing so is now lacking. (It must be realized that impact is a truly minor contributor to the rate of normal tidal waves, but the very biggest ones on a time scale of centuries or longer will be disproportionately caused by impacts.)

Despite the fact that we have been in an International Decade of scientific research on natural hazards, I am aware of no integration of the impact hazard into those research programs. Indeed, as an example of how research on the impact hazard has fared, I should note that I myself have never received any funding (other than payment of consulting services for advising several committees) for explicit studies of the impact hazard, including my personal weekend time spent on this document. As far as I can tell, except for the telescopic search efforts, there has been no formal funding by any federal agency for research on the impact hazard (as distinct from allied topics, like analysis of the K/T boundary, research on the Shoemaker-Levy 9 comet crash, or spacecraft studies of asteroids). As I understand it, what little research has been done has either been funded by modest in-house discretionary funds, has been done "on the side," or has resulted from bits and pieces of funding by miscellaneous grants that were awarded for more general research. This is probably why the literature is thin in terms of the numbers of professional papers published on this topic in the refereed literature, as distinct from brief conference reports or white papers.

Therefore, it is mostly for the future to tell us whether NASA, the National Science Foundation, the Department of Energy, elements of the DoD, FEMA, or other entities (including equivalent international entities) will fund research programs on this topic. In the absence of such programs, it is expected that telescopic searches will go on at a modest rate, but relatively little else will be done about the resulting environmental consequences, methods of mitigating impacts, and ways for society's institutions to appropriately deal with news that a particular NEO discovery has been made. In particular, the detailed technical question about how to make impact predictions will remain a topic for occasional cocktail-party speculation by a few astronomers rather than being addressed in a thorough fashion, despite the inevitability that it will sooner or later (most likely sooner) become an issue of major public attention. (The last sentence was written before the 1997 XF11 affair; perhaps that episode of media frenzy qualifies, but newsworthy events will surely happen again.)

Even the future of the NEO search programs is uncertain. There have been views among some officials in NASA, in the Office of Management and Budget, and even in language accompanying the House of Representatives' multi-year NASA authorization bill for FY98 and FY99, that the currently funded programs will substantially accomplish the goals of the Spaceguard Survey. This is wrong. Estimates by Alan Harris, reported at the March 1998 meeting, are that the current discovery rate of potentially hazardous asteroids (about 1 PHA per month, >1 km diameter) is down by a factor of 30 compared with the 350 per year required to meet the goal. Existing surveys currently cover about 3000 sq. degrees of sky per month but need to cover 15,000 sq. degrees ("all sky") to avoid having PHA's escape through the grid. In addition, the NEAT and LINEAR programs are currently reaching magnitude 19 rather than the Spaceguard goal of ~21 (causing a loss in discovery rate of another factor of 6). Finally, current programs would be unable to provide the required astrometric follow-up observations to ensure that newly discovered objects are not lost; a sampling of some discovered objects for physical properties (e.g. spectral observations for composition, or infrared observations to reduce uncertainties in the size of the object) would require even more resources, especially rapidly scheduled time on large telescopes and/or on planetary radars.

Later in 1998, there were several events that spurred more consideration of what is technically required to accomplish Spaceguard goals. First, the U.S. House of Representatives Subcommittee on Space and Aeronautics held a hearing on May 21st entitled "Asteroids: Perils and Opportunities," at which several individuals (including myself, Greg Canavan, Univ. of Arizona professor John Lewis, and NASA's Carl Pilcher) testified. The panel was asked to provide a follow-up two-page action plan, which I provided on 9 June (Chapman, 1998). The first week of June, the COMPLEX committee of the National Research Council convened a meeting to assess Near-Earth Objects and consider interactions between asteroid scientists and the press. In preparation for these events, new analyses were made of the current rate of discoveries and of designs for the Spaceguard Survey. Primarily due to discoveries by LINEAR, Alan Harris concludes that the current rate of discovery has improved so that it is only down by a factor of ~7 (not the factor of ~30 estimated in March) from what is required for Spaceguard.

If telescopes, like the Air Force GEODSS facilities, could be made available and the other augmented search programs brought on line, then attainment of the Spaceguard goals would still require effective coordination, which does not now exist. For example, observers now compete to maximize discoveries by observing near the so-called opposition point (direction opposite the Sun in the sky, where asteroids are typically brightest). But an efficient program will require some observers to search far away from the opposition point; a small Univ. of Hawaii search program is now attempting to do just that. Los Alamos National Laboratory's Greg Canavan has reportedly said that if you want to "put your name on an NEO" [discoverers get to name asteroids], it is best to observe around the opposition point, but if you want to "find the asteroid that has our name on it" then you should observe 90 degrees away from the opposition point, along the Earth's orbit. Estimates of annual operating costs for a bare-bones program are about $4 million per year for about 20 years, not including completion of capital investments in equipment under development nor costs of the GEODSS telescopes. There is considerable uncertainty whether or not NASA and the Air Force will successfully negotiate incorporation of the GEODSS telescopes into a coordinated Spaceguard Survey program.

The dominant reasons why the impact hazard has failed to become incorporated into the natural hazard policy arena are, in summary: the newness with which the hazard has been recognized; the extremely unusual nature of the hazard (extremely high consequence, extremely low probability); the wide gap between the community knowledgeable about the hazard (planetary scientists) and the natural hazards community and associated public officials; the fact that in an era of budget-cutting, federal agencies are loathe to begin new research programs; the controversial issues related to mitigation (i.e. diverting an incoming object with nuclear explosives); and self-interested motives on the part of both astronomers and ex-Cold Warriors that have distorted the projects that have been proposed for funding.

The larger issue raised by the impact hazard is the level at which governments should, or should not, do anything about it. Several years ago, Fortune magazine published a cost/benefit analysis demonstrating that it was worth spending hundreds of millions of dollars a year on planetary defense. Some advocates of planetary defense within the DoE community have argued, in effect, that billions should be spent annually. Gerrard and Barber (1997) show how one could even defend an annual expenditure of $16 to $32 billion, thousands of times more than is now being spent. However, based on nothing more than the low probability (often incorrectly translated into being equivalent to a long waiting time until the next event happens: see below) of a major impact, an argument can be made that few public funds should be expended on this unlikely hazard and that such funds should instead be spent on other hazards that certainly will happen (even if with more modest consequences) during the next years.

The nature of the impact hazard makes it unusually difficult to develop a rational, objective perspective on the priorities of dealing with it. A common mistake is for people to think that there is no reason to deal with it during our lifetimes (or for a politician to think it is irrelevant to his or her "watch") because there is a long waiting time until the next impact. The chances of a civilization-threatening impact are extremely low during a politician's term, but the chances are the same that it will occur on "her watch" as on her successor's...or during a few-year period several hundred thousand years from now. People will build in a floodplain saying, "the hundred year flood occurred here a few years ago, so I don't have to worry." It is a faulty argument. The only valid argument is that the chances of a flood (or an impact) are so low that we are willing to do nothing (or little) to avoid the consequences and, instead, spend funds to avoid lesser but more certain consequences (e.g. saving lives by investing in automobile safety).

Another choice that can be made is to let future generations, with their no-doubt superior knowledge and technology, deal with the problem. But that choice, it must be realized, avoids responsibility for consequences that could happen on our watch. Future generations cannot deal with an impact that happens during the next 30 years. If we have failed to search for the impactors that may hit during our lifetimes, then not only we, but future generations will suffer if civilization is knocked back into the Dark Ages (or, heaven forbid, our species goes the way of the dinosaurs) because we failed to discover the object that actually will hit in 2028. To be sure, the odds are so extremely small that anything of such consequence will happen during our lifetimes, that it may well be justifiable to do little or nothing about the impact hazard. The choice necessarily involves subjective values, and the issues need to be engaged.

Currently, what public debate has ensued concerning these questions has been conducted either by technical people primarily involved in other activities or in the media. Since there has been an inadequate base of research on many of the fundamental questions, it is understandable that there has been no formal analysis by an advisory body (like the National Academy of Sciences) or by governmental policy-makers about how to handle this threat. (An NAS/NRC report [COMPLEX 1998] on Near Earth Objects, which was recently released, explicitly failed to address measures motivated by the impact hazard, as distinct from scientific research on the origin and nature of NEO's.) It seems to me that, at the very least, a conscious decision should be made by society about whether to deal with the impact hazard or not.

Implications for Predictive Science and Policy

Astronomy has, traditionally, been the quintessential predictive science of unexcelled precision. For millennia, astronomers have been famously successful in predicting eclipses, the Sun and Moon unerringly rise and set on time, and planetary spacecraft reach their distant targets within seconds of the times predicted. Of course, other astronomical predictions are much less secure (e.g. predictions of the date of the next "great" Leonid meteor shower, predictions of solar flares, predictions of how bright a comet may become, or predictions about Jupiter's "weather" some months hence). Still other aspects of astronomy are so uncertain as to preclude prediction. The impact hazard, since it is based on the same kind of physics that successfully gets spacecraft to planetary encounters, would seem to be fairly foolproof, but the 1997 XF11 affair illustrates the kinds of complexities and errors that can make it seem unreliable as well.

Most of astronomy (like broad areas of many other sciences) does not have immediate practical applications, so there are few astronomical predictions which public officials will ever have to deal with. Powerful solar flares might be one of the exceptions. The impact hazard is another. A general lesson to be learned from this Case Study is that whenever the public is introduced to a "new" hazard or a new type of prediction, one must expect a rocky road as scientists, who have been unfamiliar with communicating their predictions in ways that are useful to the public or public officials, learn how to do it. On the other side of the coin, the particular journalists and public officials called on to deal with a new arena may be those (in this case, astronomy journalists and NASA officials) who have not previously been engaged in dealing with such issues. They need to learn, as well.

Because of its unusual nature, in having extraordinarily high potential consequences balanced by remote chances of occurring, the impact hazard exaggerates public responses to predictions in ways that probably affect, but less prominently, responses to other predictions. Regarding all kinds of predictive science, there is a spectrum of responses on the part of the public and policy officials ranging from great concern to minimal concern. The impact hazard generates especially strongly held but opposite views ranging from "this is the most important issue affecting the human species because it is the only one that could actually result in the near-term eradication of our species" to the view that "it is ridiculous to waste a moment thinking about such an unlikely catastrophe." As with mad- cow disease, the danger of nuclear plant melt-downs, or alleged terrorist sabotage of pills or grapes, the impact hazard Case Study demonstrates that public perceptions and media treatment of technical issues tend to overwhelm the "objective" technical facts and details. Policy officials must simultaneously keep grounded in the essential technical realities (even while the scientists argue) and yet also respond to public opinion, no matter how "irrational" it may seem.

Nearly all of the 24 "lessons learned" from the first workshop on Prediction in the Earth Sciences appear to be applicable to the case of the impact hazard. A prediction itself (as in the case of 1997 XF11) has a dominant effect on public perception of the entire issue. There are issues of multiple ways of calculating the predictions, assessing uncertainties (precision and accuracy), reporting predictions and uncertainties, being aware of societal and international contexts, potential conflicts of interest (and noting what people and agendas are promoted by predictions), and so on. It seems to me that attempts by scientists to educate the public about the objective aspects of hazards should go forward. But the narrowness of scientists' awareness of the societal and political context of their work means that the goal of greater understanding is often not met. The ever more rapid and superficial exchange of information fostered by popular culture, educational institutions, and the Internet is increasingly inimicable to the kind of in-depth understanding of complex technical issues on which society's future depends.

I found it particularly disappointing that even some technical experts in NEO's, and the best science writers in the world, largely failed to understand the essential realities of how the false prediction of 1997 XF11 was made and retracted. For example, six months after the 1997 XF11 media frenzy, the predominant view remains that the matter was resolved by finding prediscovery observations of the asteroid on existing plates. (Maybe it's the first headlines that stick in history: just three days into the fiasco, the New York Times [Browne, 1998] had headlined: "Old Photos Helped Refine Progress of Asteroid".) An official might well conclude from that false perception that more attention should be given to procedures that will increase the efficiency of finding images in plate archives. Actually, the predominant failings were (a) failure for 2.5 months of the NEO community to apply existing software to data that were publicly available, which would have demonstrated that the asteroid presented no hazard, and (b) failure of individuals and institutions to adopt and adhere to procedures that would increase the likelihood that public announcements would be valid instead of erroneous. While efforts are slowly advancing to deal with (b), failure (a) largely resulted, in my view, from inadequate funding of a group of researchers in orbit-calculation, including but hardly limited to the IAU Minor Planet Center; this funding problem will not be automatically alleviated by plans to increase funding of NEO search activities. I wouldn't be surprised if there are many other arenas in which public policy officials, doing the best they can, nevertheless adopt "solutions" that are only tangential to the technical realities of the associated problems.


I thank David Morrison for his direct and indirect assistance. Discussions with Alan Harris, Ted Bowell, Rick Binzel, Paul Chodas, Don Yeomans, Hal Levison, David Morrison, and numerous other colleagues in astronomy, as well as participants in the two Prediction Workshops, have helped me to describe the history of this topic and formulate the issues that face us.


Ahrens, T.J. & A.W. Harris 1992. Deflection and fragmentation of near-Earth asteroids. Nature 360 429-433.

Alvarez, L.W., W. Alvarez, F. Asaro, & H.V. Michel 1980. Extraterrestrial cause of the Cretaceous- Tertiary extinction. Science 208 1095-1108.

Alvarez, W. 1997. T. rex and the Crater of Doom (Princeton Univ. Press).

Baldwin, R.B. 1949. The Face of the Moon (Univ. Chicago Press).

Binzel, R.P. 1997. A Near-Earth Object hazard index. In Near-Earth Objects: The United Nations International Conference (ed. J. Remo, Annals N.Y. Acad. Sci. 822) 545-551.

Bowell, E. & K. Muinonen 1992. The end of the world: An orbital uncertainty analysis of a close asteroid encounter. Bull. Amer. Astron. Assoc. 24, 965.

Browne, M. 1998. Old photos helped refine progress of asteroid. New York Times, 14 March.

Chapman, C.R. 1998. The threat of impact by near-Earth asteroids ( and Action plan statement to: House Subcommittee on Space & Aeronautics (

Chapman, C.R. 1995. What If?... In The Great Comet Crash (ed. J.R. Spencer & J. Mitton, Cambridge Univ. Press) 103-108.

Chapman, C.R. & D. Morrison 1989. Cosmic Catastrophes (Plenum Press).

Chapman, C.R., D. Morrison, E. Bowell 1989. Hazards from Earth-approachers: Implications of 1989 FC's "near miss". Meteoritics 24 258.

Chapman, C.R. and D. Morrison 1994. Impacts on the Earth by asteroids and comets: assessing the hazard. Nature 367 33-40.

Chodas, P.W. & D. Yeomans 1997. Impact warning times for Earth Crossing Asteroids. Bull. Am. Astron. Soc. 29 960.

Chyba, C.F., G.E. van der Wink, C.B. Hennet 1998. Monitoring the Comprehensive Test Ban Treaty: possible ambiguities due to meteorite impacts. Geophys. Res. Lett. 25 191-194.

COMPLEX 1998. Exploration of Near Earth Objects (Committee on Planetary and Lunar Exploration, National Research Council, National Academy Press, Washington D.C.) 44 pp.

Ferris, T. 1997. Annals of science: is this the end? The New Yorker, Jan. 27, 1997, 44-55.

Gerrard, M.B. and A. W. Barber 1997. Asteroids and comets: U.S. and international law and the lowest-probability, highest consequence risk. N.Y.U. Environmental Law Journal 6 (1), 4- 49.

Gladstone, B. 1998. Report on Morning Edition, National Public Radio, 4 June 1998.

Goldman, S.J. 1998. The most dangerous rocks in space. (from Sky & Tel., June 1998, 33).

Gordon, B.B. 1998. The asteroid caper, editorial in Astronomy, July issue, pg. 6.

Harris, A.W., G.H. Canavan, C. Sagan & S. Ostro 1994. The deflection dilemma: use versus misuse of technologies for avoiding interplanetary collision hazards. In Hazards due to Comets & Asteroids (ed. T. Gehrels, Univ. Ariz. Press) 1145-1156.

Helin, E.F. and E.M. Shoemaker 1979. Palomar planet-crossing asteroid survey, 1973-1978. Icarus 40 321-328.

Hildebrand, A.R. & 6 others 1991. Chicxulub Crater: a possible Cretaceous/Tertiary boundary impact crater on the Yucatan Peninsula, Mexico. Geology 19 867-871.

Kleiman, L.A. (ed.) 1968. Project Icarus (M.I.T. Report No. 13, M.I.T. Press).

Kunich, J.C. 1998. Posted on World Wide Web, August 1998:

Marsden, B.G. 1997. Overview of orbits. In Near-Earth Objects: The United Nations International Conference (ed. J.L. Remo, Ann. N. Y. Acad. Sci. Vol. 822) 52-66.

Mascoli, G. (Producer) 1998. Impact: Ground Zero (All Around Us Interactive,, CD-rom.

Masursky, H. (1967). Quoted in "Monitoring of asteroids urged to warn of impending collisions," New York Times, 18 Oct.

McCord, T.B. et al. 1995. Detection of a meteoroid entry into the Earth's atmosphere on February 1, 1994. J. Geophys. Res. 100 3245-3249.

Michel, P., Ch. Froeschl‚ & P. Farinella 1998. Dynamics of Eros. Astron. J., in press.

Morrison, D. (chair) 1992. The Spaceguard Survey: Report of the NASA International Near-Earth- Object Detection Workshop (JPL/NASA, Jan. 25, 1992).

Morrison, D., C.R. Chapman & P. Slovic 1994. The impact hazard. In Hazards due to Comets & Asteroids (ed. T. Gehrels, Univ. Ariz. Press) 59-91.

Morrison, D. and E. Teller 1994. The impact hazard: issues for the future. In Hazards due to Comets & Asteroids (ed. T. Gehrels, Univ. Ariz. Press) 1135-1144.

Muinonen, K. 1998a. Asteroid and comet encounters with the Earth: Impact hazard and collision probability. In Proceedings of The Dynamics of Small Bodies in the Solar System: A Major Key to Solar System Studies (eds. A.E. Roy and B.A.Steves, NATO Advanced Study Institute, Maratea, Italy) in press.

Muinonen, K. 1998b. Upper bounds for the Earth-XF11 collision probability. Bull. Am. Astron. Soc., in press.

Muinonen, K. & E. Bowell 1993. Asteroid orbit determination using Bayesian probabilities. Icarus 104, 255-279.

Opik, E. 1957. In Irish Astron. J.

Park, R.L. 1992. Star warriors on sky patrol. New York Times (March 25, 1992) A19.

Powell, J.L. 1998. Night Comes to the Cretaceous. (W.H. Freeman, N.Y.)

Rather, J.D.G., J. H. Rahe, and G. Canavan (organizers) 1992. Summary Report of the Near-Earth- Object Interception Workshop (JPL/NASA, Aug. 31, 1992).

Remo, J.L. (Ed.) 1997. Near-Earth Objects: The United Nations International Conference (Annals of the N.Y. Acad. of Sci. 822), 632 pp.

SAIC 1994. Proceedings of the Scientific Exchange Meeting on Asteroids (USAF Space Warfare Center, Science Applications Intl. Corp., May 12-13, 1994).

Sagan, C. 1992. Between enemies. Bull. Atomic Sci. 48 24.

Sagan, C. and S. Ostro 1994. Dangers of asteroid deflection. Nature 369 501.

Sharpton, V.L. & P.D. Ward (eds.) 1990. Global catastrophes in Earth history (GSA Special Paper 247).

Silver, L.T. & P. H. Schultz (eds.) 1982. Geological implications of impacts of large asteroids and comets on the Earth (GSA Special Paper 190).

Shoemaker, E.M. 1963. Impact mechanics at Meteor Crater, Arizona. In The Moon Meteorites and Comets (ed. B.M. Middlehurst & G.P. Kuiper, Univ. Chicago Press) 301-336.

Shoemaker, E.M. 1983. Asteroid and comet bombardment of the Earth. Ann. Rev. Earth & Planet. Sci. 11 461-494.

Shoemaker, C.S. & E.M. Shoemaker 1995. A comet like no other. In The Great Comet Crash (ed. J.R. Spencer & J. Mitton, Cambridge Univ. Press) 7-12.

Shoemaker, E.M. et al. 1995. Report of the Near-Earth Object Survey Working Group (NASA).

Steel, D. 1995. Rogue Asteroids and Doomsday Comets. (John Wiley & Sons, N.Y.)

Toon, O.B. et al. 1997. Environmental perturbations caused by the impacts of asteroids and comets. Reviews of Geophys. 35 41-78.

Urey, H.C. 1973. Cometary collisions and geological periods. Nature 242 32-33.

Urias, J.M. et al. 1996. Planetary defense: catastrophic health insurance for planet Earth. Research paper presented to Air force 2025. [This and other recent documents are available on the "Asteroid and Comet Impact Hazards" Web site:]

Verschuur, G.L. 1998. Impact hazards: truth and consequences. (from Sky & Tel., June 1998, 26-34).

Watson, F.G. 1941. Between the Planets (Harvard Univ. Press, Cambridge).

Whipple, F. L. 1951. A comet model II. Physical relations for comets and meteors. Astrophys. J. 113 464-474.

Wood, L., R. Hyde & M. Ishikawa 1990. Cosmic bombardment II: intercepting the bomblets cost- efficiently (Lawrence Livermore Natl. Lab., UCID-103771).

Available Abstracts, Preprints, Articles.

Clark R. Chapman's Publications.

Clark R. Chapman's Home Page.

SwRI Boulder Office (GAP) Home Page.

SwRI Home Page