To quantify the NEO threat to Earth, we present excerpts from the Binzel Committee report to the Astronomy and Astrophysics Survey Committee:
Realization of the fundamental role played by near-Earth object (NEO) impacts in shaping our planetary history is arguably the most profound legacy of twentieth century planetary exploration. Although concern over terrestrial impacts can be traced at least as far back as Halley (1705), modern recognition of the impact hazard has been most profoundly influenced by a broad range of interdisciplinary results, including:
Recognition that the NEO population is comprised by asteroids as well as comets can be traced to the photographic discoveries of the first known Earth-approaching asteroid (433 Eros) in 1898 and the first known Earth-crossing asteroid (1862 Apollo) in 1932. Eugene Shoemaker and Eleanor Helin initiated the first dedicated photographic survey for NEOs in 1973 while CCDs were introduced to NEO survey work by Tom Gehrels in the 1980s. Naturally, CCDs are the detectors of choice for all current search programs outlined in Appendix A.
The best estimates for the power law distribution of the NEO population are derived from both search statistics and the lunar cratering record with the results shown in Figure 1. To within a factor of two, there are an estimated 1500 NEOs greater than 1 km in size, of which fewer than 20% have been discovered or catalogued. Far fewer than 1% of the estimated ~150,000 objects larger than 100 m across have currently been discovered or catalogued.
Figure 1. Cumulative power law distribution for the estimated number of asteroidal NEOs larger than a given diameter. Active comets are not specifically included in this model, but are estimated to contribute an additional 10% to the flux at any given diameter. Dashed lines indicate the one-sigma uncertainty estimates about the mean. The hand-drawn curve at the bottom is the current estimate of survey completeness. Figure is from Rabinowitz et al. (1994).
Minimal erosion or geologic activity on the volcanic plains of the Moon, from which we have samples providing emplacement ages of 3 x 109 years, allow an integrated impact flux and flux rate to be calculated (Hartmann et al. 1981). The derived lunar impact flux rate is in good agreement with the scaled rate for the Earth's surface, which due to weathering retains a record of only the most recent events (Grieve 1991). These rates, in turn, are in good agreement with modeled estimates for the NEO population (Shoemaker 1983; Shoemaker et al. 1990; Rabinowitz et al. 1994) as presented in Figure 1. Chapman and Morrison (1994) used these findings to present the current best estimate for the NEO impact flux on Earth (Figure 2), scaled in terms of the kinetic energy of the impactor, where one megaton TNT corresponds to an energy of 4.3x1015 J.
Figure 2. Typical intervals between impacts equal to or larger than the indicated energies in megatons (MT). The depicted flux rate accounts for both asteroids and comets. The solid line is the "best estimate" from Shoemaker (1983) extrapolated forward (dashed line) to the Cretaceous-Tertiary (K/T) mass extinction. The flux rate is uncertain by a factor of 10 near 10-2 MT and by a factor of 3 for >100 MT. The equivalent asteroid diameter (assuming impact velocity of 20 km s-1 and 3 g cm-3 density) is shown along the top axis. Comets having equivalent kinetic energy would correspond to nucleus sizes about 1.5 times smaller, owing to their higher average impact velocities while accounting for their lower densities. Figure is from Chapman and Morrison (1994).
Impacts are an extreme case of low-probability / high-consequence effects of the astronomical environment to the Earth. Following the analysis published in Nature by Chapman and Morrison (1994), impact outcomes may be examined in categories corresponding to whether they result in none, regional, or global environmental damage.
Meteors and Bolides. The Earth experiences a constant barrage of impacts, adding an estimated mass of 105 kg per day. Objects give up most of their kinetic energy in the atmosphere and dissipate (as meteors) or likely fragment due to aerodynamic stress or explode (as bolides) if they encounter a column of atmosphere equal to or greater than their mass. Impacts having Hiroshima scale energies (~0.015 MT) occur annually and dissipate in the atmosphere. (Such events are routinely detected by surveillance satellites. Recently declassified results are reported by Tagliaferri et al. 1994.) The upper size limit for the Earth's atmospheric shield is in the range of 30-50 m (depending on object strength and density) for a typical entry velocity of 20 km s-1 (Chyba 1993). While dozens of impact events yield recoverable meteorite fragments each year, there is no authenticated human fatality from such meteorite falls, though a famous car-conking incident occurred in Peekskill, NY in October 1992. With the exception of rare (about 5%) events involving solid iron objects, objects smaller than 20-50m produce no substantial effects on the surface or to the environment of the Earth.
Locally or Regionally Destructive Events. If a 30-50 m meteoroid is able to penetrate to within ~10 km (or strike) the surface of the Earth, the kinetic energy imparted to the surface by the atmospheric shock wave or by direct impact can cause severe local damage in a manner analogous to a nuclear bomb, but without the coincident radiation or radioactive fallout. Civil defense studies (e.g. Glassone and Dolan 1977) suggest that damage scales by the energy to the 2/3 power. The 1908 Tunguska airburst event (Figure 3) and Meteor Crater Arizona provide important calibration. Tunguska involved a weak or modest strength >50 m impactor having an energy of 10-20 MT resulting in the devastation of >1000 km2 of Siberian forest. The 1 km Meteor Crater formed 50,000 years ago as the result of a smaller (~30 m) but higher density (iron) object reaching the surface. The average flux rate (Figure 2) suggests a Tunguska-sized impactor strikes the Earth on average every 2-3 centuries, corresponding to a 30 to 50% chance for such an event occurring somewhere on Earth during the next century. The largest impactor for which there is a ~1% chance of occurrence during the next century is in the size range of 250m (1,000 MT). Such an impact would cause regional environmental devastation through the formation a 3-5 km crater on land or a massive tsunami if off shore.
Figure 3. Comparison of the >1000 km2 area of devastation resulting from the June 30, 1908 Tunguska airburst with the area of U.S. metropolitan areas. During the next century there is an estimated 1:2 to 1:3 chance of a similar energy impact. Figure prepared by John Pike and presented in Morrison (1992).
Events Having Global Environmental Consequences. The transition from increasingly severe regional environmental damage to global environmental damage likely resides in the energy range of 105 - 106 MT, corresponding to the impact of a 1-2 km asteroid or comet. Impacts within this size range occur with an average frequency of once per hundred-thousand to one-million years (Figure 2). Thus, there is a 1:1,000 to 1:10,000 chance of such an event during the next century. As shown in Figure 4, Chapman and Morrison (1994) consider that globally catastrophic effects conceivably may result from impact energies as low as 104 megatons, which are events that have a higher than 1:1,000 chance of occurring during the next century. The enormously more energetic Cretaceous-Tertiary mass extinction (Alvarez et al. 1980) appears to be a 50- 100 million year event, corresponding to a one-in-a-million chance during the next century. Significant temporal variations in the flux may occur due to "comet showers" or the break-up of a large Earth-crossing asteroid. However, the flux rate in Figure 2 assumes a randomly distributed population.
Figure 4. Estimated thresholds (shaded region) for impact energies and probabilities capable of triggering a global catastrophe, where an estimate of global fatalities is shown at the right. The lower threshold corresponds to a 1:1,000 chance of astronomically induced environmental effects during the next century. The dashed line indicates the potential local effects of tsunamis, whose century-timescale hazard chance may exceed 1 percent. Figure is from Chapman and Morrison (1994).
Toon et al. (1994) considered the global environmental perturbations that are likely to result from impacts and reached conclusions compatible with mean estimate from Chapman and Morrison (1994) in that global environmental consequences result from impact energies exceeding 105 MT, corresponding to the same 1:1,000 to 1:10,000 chance for such an event during the next century. The environmental effects considered by Toon et al. (1994) included those arising from the effects of dust and water injected into the stratosphere (Figure 5) and the depletion of ozone due to the formation of NO by the entering impactor and the re-entry of impact ejecta. Toon et al. predict environmental effects (manifested in part by significant global temperature changes) would persist for time periods ranging from months to years depending on the energy of the impactor and the precise circumstances of its trajectory, impact location, etc.
Figure 5. The astronomical effect on the environment of the Earth by NEO impacts is depicted in terms of changes in the transmission of solar radiation due to the increasing optical depth created by dust injected into the stratosphere. Open circles assume all of the generated dust mass is in the form of sub-micron particles, while filled circles correspond to 10% efficiency in the creation of sub-micron dust. Figure is from Toon et al. (1994)
The U. S. Congress and other international governmental agencies have come to recognize the reality of the NEO hazard and the potential benefits of ensured public safety that result from improved understanding and assessment of the NEO population. A concise set of statements from Congressional acts, the Executive Committee of the International Astronomical Union, and the Council of Europe are attached in Appendix B.
As shown in Figure 1, the current survey completeness of the NEO population falls off rapidly at sizes below 3-4 km. Harris (1998) has published the most extensive model calculations detailing the observational requirements for reaching higher levels of completeness. Figure 6 shows, for example, that a decade of all sky searching down to a magnitude limit of 21 is required to discover 90% of all 1 km NEOs. Current surveys (Appendix A) are falling far short of the sky coverage and limiting magnitude requirements to achieve completeness in discovering these large and potentially globally hazardous objects. At a magnitude limit of 24, Figure 6 shows that 90% discovery completeness could be achieved for 250 m objects, where this size range corresponds to the level of a 1% chance of impact during the next century having the potential for severe regional environmental damage. Due to the fast angular motion of these NEOs, exposures must be kept short (20 sec) to avoid trailing.
Figure 6. Fraction of completeness for discovery of the NEO population versus the limiting V magnitude of a search telescope system. The level of discovery completeness for each size range plotted is what would be accrued after a decade of effort. The model assumes "all sky" surveying of ~15,000 deg2 per month, which is typical of what can be performed at mid-latitudes with the Moon below the horizon, the Sun > 10 degrees below the horizon, and with a 20 degree avoidance of both the Galactic plane and the horizon. While this model is based on near-Earth asteroids, comparable completeness can be expected for short-period comets. Long-period comets remain a constant incomplete fraction (estimated at 10%) of the NEO hazard. Figure is reprinted from Harris (1998).
Search requirements, however, must extend beyond discovery, which is the "completeness" criterion used in Figure 6. A commensurate goal with discovery must be sufficient follow-up observations for a preliminary orbital characterization that is adequate to predict an object's location during future apparitions so that it becomes recoverable as a "known" body instead of being repeatedly "rediscovered". Effectively, this requires sky area coverage by a factor of 3 beyond the 15,000 square degrees per month used in the Harris (1998) model. Additionally, discovered objects recognized as having potentially hazardous orbits require substantial follow-up observations to yield the best possible predictions for future Earth encounters. Accounting for weather effects (single survey telescope) this requires surveying 15,000 square degrees every few nights, with exposure times less than 20 sec, to fainter than 24th V mag in several wavelength bands. This leads to the requirement of an effective aperture of about 7 meters for a 3-deg field.
