Outline

Observations:

· Scales assume a dominant role in communicating natural hazards to the media and society at large.

· Confusion, misunderstanding, and sensationalism are often associated with Asteroid and Comet Impact Hazard Predictions.

· The scales pertinent to Asteroid and Comet Impact Hazard Predictions are partially ineffective.

Explanations:

· The Torino Scale is flawed. · It provides no indication as to how the public should respond to an impact hazard prediction.

· Competing Scales are reducing the effectiveness of the message. · References to the Torino Scale and the Palermo Scale are problematic, in that they are ignoring the two scales incompatibilities: they were devised for wholly different purposes and they yield results that are essentially incommensurable.

Recommendations:

· Couple the Torino Scale’s assessments with clear descriptions or recommended actions designed specifically for the general public.

· Devise an IAU policy to encourage accuracy, consistency, and clarity in reports of impact predictions. Reviewing the Torino Scale; Critiquing the Critiques

Communicating the risk that near-Earth objects (NEOs) pose to the general public presents a serious challenge to the astronomical community. The challenge stems, in part, from the unique character of the NEO hazard. The concept of asteroids and comets colliding with the Earth represents the most extreme example of a low probability, high consequence event - outside the domain of our common lived experience. The NEO hazard is highly prone to sensationalism, especially in the wake of two Hollywood blockbusters (Armageddon and Deep Impact) depicting havoc wreaked by earth- approaching celestial bodies. Finally, cosmic impact events differ from other hazards in that, in theory, impact dates that lie far in the future can be forecasted with great accuracy. This risk communication challenge was addressed by the Torino Scale, a simple hazard scale system devised to encourage “careful and responsible communication between astronomers and the public” (Binzel, 2000). Since its creation, the Torino Scale has since been criticized by numerous parties, many of whom have suggested modifications to the existing scale or new scales altogether. However, the Torino Scale’s detractors have all but ignored the scale’s raison d'être: “to place into context the level of public concern that is warranted for any close encounter event within the next century” (Binzel, 2000). To examine the success or merit of the Scale, we must do so in the context of its justification for existence. To deride the Torino Scale on the basis of supposed goals - projected by the critic(s) - is a non sequitur. This is not to say that the Torino Scale effectively realizes its purpose; it does not. Before we evaluate the Scale according to its stated goals and recommend improvements, let us review past criticisms and suggested alternatives to see where they have gone awry. William J. Cooke, suggests that the Torino Scale is “too coarse” to establish the risk threshold required to trigger an asteroid defense. This assertion is supported by the claim that the Torino Scale neglects the “attention span” of the average person (100 years). He proposes a risk number (RN) scheme and challenges others to establish

RNdefense: a value that dictates when society should defend against an asteroid/comet, without which “an adequate mitigation means cannot be designed and developed” (Cooke). Cooke’s evaluation ignores the fact that the Torino Scale does include a time element: it is defined as being relevant only for the next century (100 years - precisely his quantification of a human “attention span”). His implicit acknowledgement that a hazard scale must motivate some form of human response is laudable, but his critique of the

Torino Scale is unfounded. Joaquin Perez proposes that the Torino Scale “fails to satisfy minimal validity and simplicity criteria” (Perez, e-mail correspondence). He suggests that the Scale has technically flaws; most notably that a “very likely” (0.80 < P < 0.95) collision with an energy of one hundred million megatons is considered less threatening than a certain (P > 0.99) one megaton collision. Perez then submits modifications to reduce the Scale’s technical/mathematical weaknesses. His effort are valiant, but his suggestions fail to increase the simplicity and hence the public intelligibility of the Scale. The Torino Scale has “natural limitations” (Binzel, 2000) that, arguably, could be slightly improved with continual tweaking. However, this would only serve to cause confusion and undermine public confidence - counter to the scale’s intended purpose. Jonathan Tate recommends a refinement of the Torino Scale on the bases that the current scale has been little used and that “a single number carries no advantage if lengthy explanation is required”. These claims are false; the scale has been used in several appropriate instances since its development and as it continues to be used, public confidence in it will increase and the scale’s integer values will begin to be associated with their explanations. This is a well-established characteristic of risk communication; “the number of times that a risk communication is repeated has a positive effect on public belief in the information” (Mileti et al., 1991). Californians have learned to readily associate Richter Scale values of earthquake magnitudes with the “lengthy explanations” that support them. Tate further criticizes the Torino Scale because its hazard ratings are liable to change - asserting that this feature makes the Scale “of little value to the media or general public”. This feature is a part of other hazard indices such as the Saffir- Simpson Hurricane Scale that have been embraced by journalists and civilians alike. Tate also opposes any mention of “background hazard”, an element of the problem he deems irrelevant. How else is society to relate to an impact prediction and decide whether it warrants action or not? Society’s inaction as pertains to the background hazard represents an implicit judgement that we can all live with the background threat without taking exceptional measures. Informing the public that a potential impact rises significantly above the background is thus of great utility. Tate’s remedy to the aforementioned ills is a two-parameter matrix that labels each “event” with a “threat level” composed of an appropriate letter and number describing its probability and its consequence respectively. Noting that it is “advisable to have pre-agreed and templated (sic) actions ready for use”, Tate then offers a second 2-D matrix (the “action matrix”) that dictates an appropriate response based on an event’s “threat level” and the time to impact. Tate’s suggestions may have some merit, but in no way do they address the problems of underuse and inadequacy that he emphasizes. A pair of two-dimensional matrices are far too complex to serve as a tool of public communication and would certainly be of less utility than the Torino Scale. Another attempt to quantify the cosmic impact hazard, the Palermo Scale, was developed by an international team of researchers to “facilitate communication among astronomers”(Chesley et al., 2001). The authors carry out their task because the simplicity of the Torino Scale renders it “unsuited for use by specialists in characterizing large numbers of events and in prioritizing objects for observation and analysis” but admit that it is “a clear and very simple measure of the hazard posed by a potential collision” (Chesley et al., 2001). The authors note that the Palermo Scale is “not intended for public communication of impact risks” yet it has been cited by the media. A recent article posted on CNN.com referred to asteroid 2002 NT7 (detected on July 9 by the Lincoln Near Earth Asteroid Research Project) as “the first object to be given a positive value, of 0.06, on the Palermo scale of potential threat posed by asteroids” and makes no mention of the Torino Scale. A follow-up article stated that “NASA's Near Earth Object program gives the asteroid a rating of "0" on the Torino impact hazard scale — within a range of "events meriting careful monitoring," but not concern”, this time refraining from referencing the Palermo Scale. Reuters.com reported Tim Spahr (an astronomer at the Minor Planets Center of the Harvard-Smithsonian Center for Astrophysics) as saying, “We have a scale called the Palermo scale that takes into account size and possible impact velocity and comes up with a rating for an object." The article noted that “2002 NT7 is the first "positive" object on the scale system, meant to predict how much damage an asteroid would do if it just happened to hit”. The purpose of the Palermo Scale has been utterly misinterpreted by this report. Though the intentions of the creators of the Palermo Scale were clearly not to confuse or misinform the public, the presence of a second impact hazard scale has done just that. What is a lay audience to make of an asteroid that registers a zero on one scale and a positive, non-zero value on another?

The Torino Scale’s Real Weakness

The Partnership for Public Warning (PPW) recently published a report outlining several unanimous recommendations devised to improve the Homeland Security Advisory System (HSAS). One such recommendation stressed that the HSAS “be clear about the risks and the actions required to reduce the risks” (Partnership for Public Warning, 2002). Indeed, the natural hazards community has long acknowledged the relationship between warning systems and human response: “warnings are effective only if they are accurate and result in appropriate action” (Working Group on Natural Disaster Information Systems, Subcommittee on Natural Disaster Reduction, 2000). Joanne Nigg’s article, Risk Communication and Warning Systems cites three key components of a public warning message, one of which is that “people must be told what they can personally do to reduce their exposure to danger” (Nigg, 1995). To evaluate the effectiveness of the Torino Scale in serving its intended purpose, we must examine whether or not it moves the average person to appropriate action. What actions are to be taken based on where an event lies on the Scale? The Torino Scale is limited in the advice it offers the public as to how they should respond to an event. Each coloured zone on the Scale is coupled with a qualitative descriptor: white events have no likely consequences, green events merit careful monitoring, yellow events merit concern, orange events are threatening, and red events are certain collisions. Beyond this information, the Torino Scale provides no indication as to how the public should respond to an impact hazard prediction. Since the Scale is a tool of public communication, one can assume that these descriptors are designed to be read be the public. But are they? Green and yellow Torino “events” appear especially problematic: Should the average person carefully monitor a green “event” (potentially a 20 m sized body with a less than 10% chance of collision)? Should the media? Or is this ‘advice’ aimed at the astronomer? Yellow “events” merit concern - but who should be concerned and what is meant by “concern” remains unclear. These aspects of the Torino Scale must be considered and improved upon if the Scale is to provide to the public a helpful assessment of NEO close encounter predictions. The Homeland Security Advisory System was designed to convey the risk of terrorist acts in America by means of a threat condition. “The assignment of a Threat Condition shall prompt the implementation of an appropriate set of Protective Measures. Protective Measures are the specific steps an organization shall take to reduce its vulnerability or increase its ability to respond during a period of heightened alert” (as stated in the Federal Register, March 18 2002, Volume 67, Number 52, Page 12047- 12049). These protective measures are directed at federal departments and agencies; they provide little advice to the average citizen as to how to respond to a given threat condition. The American Red Cross (ARC) recognized this and developed five sets of recommended actions tailored specifically for individuals, families, neighbourhoods, schools, and businesses. According to the Partnership for Public Warning, these actions will be included in the HSAS implementation report issued to the President. The ARC recognized a weakness in the HSAS - it failed to advise the public as to how they should act based on the current threat condition (it lacked one of Nigg’s three key components of a public warning message). The same weakness can be found in the Torino Scale: though a tool of public communication, the limited “action-oriented” information it provides is (mis)directed at the astronomical community. After identifying this weakness, the ARC moved to amend the System - by coupling each threat condition with a set of recommended actions, directed at a certain element of society. For example, if the threat condition is blue (guarded), individuals are told to “review stored disaster supplies and replace items that are outdated” (American Red Cross, 2002). Perhaps the Torino Scale could be amended in an analogous fashion. Instead of suggesting that the average citizen “carefully monitor” a green Torino “event”, the Scale could recommend that that the public refrain from concern, indicating that astronomers are carefully monitoring the body, refining orbital calculations, and likely to soon announce with certainty that no impact will occur.

Why the 2002 NT7 flap occurred...

A great deal of the confusion, misunderstanding, and sensationalism that accompanied announcements of the discovery and tracking of Asteroid 2002 NT7 was caused by references to two scales that exist to convey the cosmic impact risk: the Torino Scale and the Palermo Scale. Although each assumes the lyrical name of an Italian city, the two scales were created for greatly different purposes. The Torino Scale is a simple hazard scale system devised “to place into context the level of public concern that is warranted for any close encounter event within the next century” (Binzel, 2000). It is commonly accepted that the impact hazard is a ‘sticky subject’, presenting astronomers with an enormous risk communication challenge. The Torino Scale was designed to meet this challenge, encouraging “careful and responsible communication between astronomers and the public” (Binzel, 2000). On the other hand, the Palermo Scale was developed by an international team of researchers to “facilitate communication among astronomers” (Chesley et al., 2001). The authors carry out their task because the simplicity of the Torino Scale renders it “unsuited for use by specialists in characterizing large numbers of events and in prioritizing objects for observation and analysis” but admit that it is “a clear and very simple measure of the hazard posed by a potential collision” (Chesley et al., 2001). It is emphasized that the Palermo Scale is “not intended for public communication of impact risks” (Chesley et al., 2001).

INSERT HERE - a paragraph that evidences my claim that confusion, misunderstanding, and sensationalism were associated with Asteroid 2002 NT7 (newspaper headlines, etc), especially dues to references to both scales.

Perhaps the public and the press - often both more curious and more intelligent than given credit for by scientific specialists - were/are not satisfied by the Torino Scale for one simple reason: it fails to satisfy the experts! Experts found the Torino Scale to be “unsuited” for their purposes; its crudeness barred it from possessing any practical utility. But why should this not also apply to the public? Since the inception of the Torino Scale, no celestial body has merited any value other than zero or one on the Scale. One could argue that this is telling the public something important: hazardous cosmic impacts are very rare, therefore we should not expect to hear of any Torino “events” other than zeros and ones. However, the public seems to be getting another message altogether: the Torino Scale alone is insufficient; we need more information to compare this prediction with previous ones. The Palermo Scale allows this; in fact, it was contrived “to quantify in more detail the level of concern warranted for a future potential impact possibility” (Jet Propulsion Laboratory, 2002 - my emphasis). The perception that scientists are pacifying the public with “dumbed down” information while they internally manage the “real” data is not new. As long as this mentality exists, interested lay people and media members will continue to track down this “real” information - the same stuff the scientists deal with. The Partnership for Public Warning (PPW) notes, among the lessons learned from decades of experience with issues concerning natural hazards, that “withholding information... is counterproductive. If authorities do not provide information, people will seek it from other - usually less reliable - sources” (Partnership for Public Warning, 2002). As far as public communication is concerned, the Palermo Scale may be considered a less reliable source, but it has been sought out because the Torino Scale alone does not suffice. In addition, it would be counterproductive to ‘hide’ the Palermo Scale from prying eyes. The PPW also asserts “that no single source has complete credibility regarding all aspects of the threat and protective actions. Identify procedures by which different sources can ensure that their messages are compatible” ” (Partnership for Public Warning, 2002). The presence of two scales could serve to bolster public perception of the hazard - “Consistency of what is said in a risk communication with what is said in other communications enhances public belief in the information” (Mileti, 1991) - but, at first glance, the messages appear to be incompatible (2002 NT7 initially registered 0 on the Torino Scale and 0.06 on the Palermo Scale). Astronomers must now realize that future impact predictions will now be publicly evaluated in light of both the Torino and Palermo Scales. This could again lead to conflicting reports that readily lend themselves to confusion, misunderstanding, and sensationalism - but all this can be averted. Scientists can amend one of the two scales to demonstrate that they are indeed commensurable. They can increase the resolution of the Torino Scale to aid the public in comparing current and past impact predictions, devise a graphical interface that improves the intelligibility of the Palermo Scale, or they can tweak either of the two Scales’ number systems to increase the correlation between the two. If these options (or other, similar ones not listed here) are ignored, then the astronomical community can expect to read dramatic, apocalyptic headlines about the next discovered near earth object, just as they did about 2002 NT7.

Torino vs. Palermo: Battling Scales

It is important to note that competing scales exist not only in the cosmic hazard domain, but in the realm of natural hazards at large. For example, several earthquake- related scales exist and may serve to offer contradictory advice to the general public. The first and mostly widely known of these is the Richter Magnitude Scale, developed by Charles F. Richter at the California Institute of Technology in 1935. Basically, the scale expresses local earthquake magnitude (ML) as the logarithm of the amplitude of seismic waves recorded by seismographs, adjusted for variations in the distance between the instruments and the earthquake’s epicenter. Richter developed his Scale specifically for Wood-Anderson seismometers used in Southern California and, as such, scale factors are required for other instruments used in other locales. In 1977, Hiroo Kanamori created a scale that is independent of instrument type: the Moment Magnitude Scale. This scale defines seismic moment (M0) as the product of rock rigidity, fault area, and slip distance. The seismic moment can then be converted to the moment magnitude (MW). Kanamori’s scale is attractive for several reasons: it accounts for the total seismic energy released by an earthquake and can thus measure earthquakes large or small, near or far; it can operate based on geologic or instrumental input, thus allowing researchers to compare earthquakes that predate the instrumental record with more recent events; it is inherently more reliable than the Richter Magnitude Scale, allowing earthquake similarities and differences to be evaluated with increased confidence. Although it is a common public misconception that the Richter Magnitude Scale ranges from one to ten, it is in fact an unbounded scale. However, ML tends to be used only for earthquakes smaller than approximately magnitude 6. This is because of an instrumental defect: large earthquakes can cause enough shaking to force seismograph traces off-scale, resulting in a “saturation” in the maximum amplitude of deflection. To circumvent this saturation, researchers often rely on two additional measures of earthquake magnitude: body-wave magnitude mb, and surface-wave magnitude Ms.

Small, local events are characterized by mb, calculated using short-period P waves and reliable up to about magnitude 6.5, where it saturates everything larger. Large, distant events are characterized by Ms, calculated using the amplitude of surface waves with a 20-second period on a long-period vertical seismometer and reliable to about magnitude

8.5, where it too saturates all greater values. Ms values closely approximate mb values below magnitude 6. Perhaps the greatest attribute of MW is that is does not saturate; it can be used for all earthquakes, regardless of their energy output. How is the public to respond to earthquake reports that cite a variety of (potentially contradictory) earthquake magnitudes? Is the public or the press even aware of the existence of differing means to convey earthquake magnitude? There is no doubt that communicating the Richter Magnitude Scale (likewise the Torino Scale or any scientific device) to the masses is a difficult task. “It's a fairly common occurrence for visitors at the Geophysical Institute's seismology laboratory to ask if they can see the Richter scale as if it were a piece of equipment” (Gedney, 1985). Key federal agencies have identified this task and are working to ensure accuracy, consistency, and clarity in their reports of earthquake magnitude. In a recent memorandum to United States Geological Survey (USGS) personnel working in the Earthquake Hazards Program, John Filson (the Program’s Coordinator) outlined the USGS Earthquake Magnitude Policy: “The U.S. Geological Survey henceforth will be reporting moment magnitudes in reference to the size of earthquakes, although other estimates of magnitude may be used just after an earthquake when the moment magnitude may not yet be available. In public statements we should simply use the term “magnitude” without any adjectival modifiers” (United States Geological Survey, 2002). The Policy further informs USGS personnel: “If asked, more information can be provided. Typical additional information can include that the magnitude was estimated using an extension of the concept originally developed by Richter, and/or that there are several different methods for estimating the size of an earthquake, all of which are consistent with the Richter scale, and a description of the measurement technique used. However, most non- earth scientists are confused by this additional information, so it should be provided only if requested.” (United States Geological Survey, 2002) The policy then establishes a hierarchy of regional authoritative networks for information dissemination within their respective bounds, to ensure consistency in the message delivered to the public. The IAU could devise an analogous policy, perhaps titled: The Asteroid and Comet Impact Hazard Prediction Assessment Policy. Such a policy would similarly dictate what scale is to be used in evaluating the hazard, recommend ways in which researchers can temper scale values with additional information, and establish a protocol for information dissemination that outlines a hierarchy of authoritative networks. Modeled after the USGS policy, such an effort would effectively encourage accuracy, consistency, and clarity in reports of impact predictions. One potential barrier to this project stems from the fact that different methods for estimating the size of an earthquake are consistent with the Richter scale. The ‘machinery’ behind a magnitude value does not need to be revealed since the different methods used to obtain it yield nearly identical results. The same can not be said for the different methods for estimating the hazard presented by potential cosmic impactors. Creating a conversion method, or similar means that at least demonstrates to the public and press that the Torino and Palermo scales are compatible is a prerequisite to the policy described above.

Works Cited

American red Cross. (2002). Homeland Security Advisory System Recommendations.

Binzel, R.P. (2000). The Torino Impact Hazard Scale. Planetary and Space Science, 48: 297-303.

Chesley, S.R., Chodas, P.W., Milani, A., Valsecchi, G.B., Yeomans, D.K. (2001). Quantifying the risk posed by potential Earth impacts. In Asteroids 2001 Conference.

Cooke, William J. Application of International Space Station Debris Protection to Fundamental Problems in Asteroid Defense.

Gedney, Larry. (1985). The Richter Magnitude Scale. Alaska Science Forum, Article 701, February 4, 1985.

Jet Propulsion Laboratory. (2002) NEO Program: Impact Risks Website. http://neo.jpl.nasa.gov/risk/doc/palermo.html. Last updated: March 12, 2002.

Mileti, D. S., Fitzpatrick, C. (1991). Communication of Public Risk: Its Theory and Its Application. Sociological Practice Review, vol. 2, no. 1: 20-28.

Nigg, J. (1995). Risk Communication and Warning Systems. In Horlick-Jones, T., Amendola, A., and Casale, R., editors, Natural Risk and Civil Protection. E & FN Spon.

Partnership for Public Warning. (2002). Improving the Effectiveness of the Homeland Security Advisory System: A Report by The Workshop on Effective Hazard Warnings.

Perez, Joaquin. (2000). Personal e-mail communication.

Tate, Jonathan. CCNet.

United States Geological Survey. (2002). Reporting Earthquake Magnitudes: Memorandum to USGS Personnel working in the Earthquake Hazards Program. Download: http://www.ceri.memphis.edu/usgs/magnitude.shtml

Working Group on Natural Disaster Information Systems, Subcommittee on Natural Disaster Reduction. (2000). Effective Disaster Warnings.