Interested readers can find a thorough and very useful introduction to all aspects of asteroid science in the Asteroids III book (2002), which includes several chapters mentioned in the References of this paper.

2 Pebbles From Heaven

Asteroids are much more complicated bodies than purely large-scale replicas of typical pebbles. But even when speaking of simple pebbles, we should be careful. Pebbles are monolithic pieces of rock. Rocks are assemblages of minerals. Minerals are solid-state chemical com- pounds. If we tried to tell the story of any apparently insigni¯cant pebble we may ¯nd during a walk, we would need a long time. This would be a tale of heating and cooling, of pressure and erosion, of endless displacement, and water and wind. But we do not know any pebble grown in our Earth, that would be capable to tell us the tale of the formation of the Sun and of the planets from the original proto-planetary disk. In the most favorable cases, if we were lucky enough to choose as our pebble some rare zircon, we might be told the history of our newly-born planet up to 4.4 Gyrs ago (Blichert-Toft and Albarede 2008, and references therein). But if we want to be told the story up to a more distant past, when the original dust started to settle down and accrete into planetesimals, no rock collected on the Earth can help us. It is now widely recognized that our planet melted completely at the beginning of its existence. All this melted material leading to differentiation took place, with the heaviest elements falling toward the centre, and lighter ma- terial tending to float on the surface. Cooling down and solidification came after this. All the information that the original rocks could have about what was before, was lost at the melting epoch (see, for instance, Taylor 2001).

Since a long time we know that some small samples of rock that fall from the sky, have been witnesses of the processes that came before the melting of the Earth's material. These samples are included in some meteorites, objects that belong to the general population of minor bodies of our Solar System. Radiometric dating of these samples tell us the age of our Solar System: 4.66 billions of years (Taylor, 2001, and references therein). These small celestial bodies have survived, apparently not modified by any major physical evolution, since that time. And we know that most meteorites came originally from the asteroid main belt (Morbidelli et al. 2002). Comets are also thought to be primitive bodies, but their structures are fluffy, and hardly survive the passage through the terrestrial atmosphere in the case of an impact with our planet (Burbine et al. 2002). Asteroids are more compact.

Some small pieces of asteroids fallen on the Earth provide therefore some critical data to constrain the composition of the solar nebula at the epoch of its collapse. They are essential pieces of evidence concerning the chemical history in our Galaxy and possibly the origin of our planet and life on Earth (Joseph, 2009), because they tell us what was the relative abundance of the various elements at the epoch of birth of our Sun, in this region of our Galaxy. At the same time, some rare isotopic anomalies detected in some samples, provide some clues about the complicated events that took place at that time, with the likely explosion of one or more supernovae, which plausibly affected the collapse of the proto-solar nebula, and led to the insertion of anomalous material into the "primordial soup" (Taylor, 2001).

All this might look as an interesting, but now well-established body of knowledge. But asteroids can still amaze us: On October 7, 2008, a small asteroid orbiting along an impact path with the Earth was discovered just a few hours before hitting our planet. It was named 2008 TC₃ (Jenniskens et al. 2009). The reconstructed path, complemented by a few observations of the fireball produced by the impact with the atmosphere, led to identify the site of fall of the meteorite, in Sudan. The recovered meteorite was made of fragile, carbon-rich material, and its original parent asteroid was classified as a member of the so-called F taxonomic class. It was the first time that pieces of an object previously detected in the open space were later recollected on the ground after crossing the atmosphere.

Asteroids belonging to the F class are relatively rare and present some puzzling properties, which suggest some link with the comets (Belskaya et al. 2005; Cellino et al. 2001). Extinct comets are known to possibly exist among the asteroid population, particularly among the asteroids that are decoupled from the main belt, and orbit in the region of the terrestrial planets (Binzel et al. 2002). Conversely, another recent discovery has been that of some so-called main belt comets, bodies which were originally classified as normal asteroids in the outer regions of the main belt, around 3.3 Astronomical Units from the Sun, but have been later found to exhibit some cometary activity (Hsieh 2009). We are recognizing these years that the boundaries between different classes of minor bodies are often quite fuzzy.

The above considerations should urge us to look at pebbles with more respect, mainly those which came from the space. But, as we will see below, asteroids are much more interesting than purely being samples of primitive material of our Solar System.

3 Asteroids as Interesting Celestial Bodies

It may be surprising to know that there is a lot to learn from the study of such small celestial objects as the asteroids, and several interesting open problems are present and challenge the imagination of the researchers. Gone are the days when the only information we had about these objects consisted of some short trace left on photographic plates. Yet, these old astrometric observations were important. They allowed the astronomers to compute the orbits of these bodies. In this way, many things were discovered. Aster- oids orbit mostly in the so-called main belt, between 2:1 and 3:3 AU, but many objects were found to have orbits at the same heliocentric distance of Jupiter, one group preceding the giant planet, another group following it.

These objects, the so-called Trojans, are beautiful examples of orbital sta- bility in the so-called three-body problem in Celestial Mechanics, and their origin and survival in this particular orbital con¯guration can give us some constraints on the origin and evolution of the outer Solar System (Marzari et al. 2002, and references therein). Another group of objects have orbits that bring them to sweep the region of the terrestrial planets. These bodies, the so-called near-Earth asteroids (NEA), have short lifetimes. Their fate is to wander until they eventually hit the Sun, or are perturbed and ejected into the outer Solar System. Another possible end-state is an impact with one of the terrestrial planets, including our own Earth (Gladman et al. 1997).

A steady supply of new objects is needed to explain the existence of a stable population of NEAs. The source region has been found to be the asteroid main belt (Morbidelli et al. 2002). There, the asteroids are not uniformly distributed, as shown in Figure 1. Forbidden strips of orbital semi-major axis exist, which correspond to zones of resonance with the orbital motion of (mostly) Jupiter. Even in the non-forbidden regions, several big clusters of objects are clearly visible if one plots the so-called proper elements, a kind of filtering and averaging of the orbital elements over long time scales, to remove the time-dependent oscillations of the elements due to planetary per- turbations (Knezevic et al. 2002). These clusters in the proper element plots are the so-called dynamical families, and are thought to be the outcomes of catastrophic collisions that led to the destruction of single parent bodies, and to the creation of clouds of fragments that can be nowadays identified (Zappala et al. 2002, and references therein).

Families are among the most evident products of the general process of collisional evolution of the asteroid population (Davis et al. 2002). Other, more indirect evidence comes from the existence of large asteroids having fast rotations and elongated shapes, as resulting from photometric data (the so-called light curves, describing the periodic variation of brightness due to the object rotation). These ob- jects are interpreted as having an overall equilibrium shape corresponding to their angular momentum, and assuming that they behave at least partly as fluid bodies (Farinella et al. 1981). This can be a reasonable approximation if these bodies are gravitational assemblages of much smaller pieces, interpreted as the outcomes of collisional disruption and re-accumulation of the fragments (Farinella et al. 1982). It is now well established that colli- sions have been the most important evolutionary process that has shaped the properties of the asteroid population, including the overall inventory and size distribution of main belt objects.

In recent years, it has been discovered that interesting and previously neglected mechanisms exist, which create a link between the physical prop- erties of small asteroids and their dynamical evolution, up to their final fate. In particular, the pressure of the thermal radiation emitted by the surface exposed to solar heating, produces a drift in orbital semi-major axis, which depends on many parameters, including the spin rate, the spin axis orien- tation, the object size and the thermal inertia of the surface. This so-called Yarkovsky effect (Bottke et al. 2006, and references therein), coupled with catastrophic disruptions caused by mutual collisions, seems to be the main engine that injects small asteroids into the main zones of orbital instability (resonances). From there, the objects quickly evolve into orbits which bring them out of the Solar System or in the region of the terrestrial planets, steadily replenishing the near-Earth population (Morbidelli et al., 2002).

Spectroscopic observations show that asteroids exhibit a variety of prop- erties, and belong to a number of different taxonomic classes. Different classes are thought to correspond to differences in mineralogical composi- tion of the surfaces. The relative abundance of objects belonging to different taxonomic classes varies as a function of heliocentric distance, and this is an important evidence of the existence of a general gradient in composi- tion of the original proto-planetary disk (Cellino, 2000). This general trend, however, is not as sharp as one might expect, and is partly hidden by a con- siderable amount of mixing. One of the most common taxonomic classes, which is more abundant in the outer belt, is believed to correspond to bodies having the same composition of the oldest and most primitive meteorites, the carbonaceous chondrites, which date back to the epoch of early forma- tion of our Solar System (Burbine et al. 2002). However, some apparent paradoxes are evident as one looks at the available data.

One of the major problems is the great difference between the properties of the big objects (1) Ceres and (4) Vesta. (1) Ceres is now classified as a dwarf planet, has a diameter of about 1,000 km, and exhibits a reflectance spectrum that is diagnostic of a primitive, and possibly hydrated, composition. As oppo- site, (4) Vesta, which has a diameter of about 500 km, exhibit evidence of a basaltic crust, which is interpreted as evidence of full melting and dif- ferentiation occurred in ancient times (Cellino et al. 2006). The internal heating that could produce the melting is thought to have been provided by the presence in the interior of short-lived radiogenic nuclei like Al₂₆.

The problem here is that, if it is admissible to accept the idea that Vesta had a thermal history in some way similar to that of the terrestrial planets, it is much more difficult to explain why and how Ceres did not experience the same kind of evolution. Being twice as large as Vesta, Ceres is expected to have accreted even earlier, and if the composition of the solid material in the asteroid belt at that epoch had been the same, it should have incorporated an amount of Al₂₆ more than sufficient to fully melt it. Apparently, this did not happen, and Ceres's composition is that of a primitive body that never experienced strong heating episodes. Today, the difference in heliocentric distance between Vesta and Ceres is about 0.4 AU (Vesta being closer to the Sun). The general gradient in composition of the asteroid population as a function of heliocentric distance does not suggest that the original com- position of Vesta and Ceres could be so different (Cellino, 2000). How to explain this apparent paradox? We hope to obtain some critical body of evidence from the results of the Dawn space probe, which is on its way to reach Vesta, and later Ceres, in the next years (Cellino et al. 2006).

Another long-debated problem has been for a long time that of the par- ent bodies of the so-called Ordinary Chondrites, the most common class of meteorites that fall on the Earth. The genetic link between the most populous taxonomic class of asteroids orbiting in the inner part of the main belt has been long questioned due to some apparent inconsistencies in the re°ectance spectra. The conundrum has been apparently solved when the Galileo space probe visited the asteroid (243) Ida and found that the ma- terial excavated around the most recent impact craters closely mimics the spectral properties of Ordinary Chondrites. The explanation was then that a process of space weathering due to the effect of exposure of the surfaces to solar wind, cosmic rays, micro-meteor impacts, progressively modifies the spectral properties of the surfaces of these asteroids (Chapman 1996). Very recent results, however, point out that space weathering mechanisms might act over very short time scales, much shorter than the dynamical timescales needed to move asteroids from the main belt to the region of the terrestrial planets (Vernazza et al. 2009). If confirmed, this might be a new problem for current models.

Collisions, non-gravitational dynamical effects, apparent inconsistencies in the thermal histories of the bodies. We are only scratching the surface of the problems that are today open in asteroid science. A major limit of our present knowledge is the fact that we do not have at disposal reliable values of mass and average density, but for a handful of objects explored in situ. It is certainly hard to do Astrophysics without knowledge of masses and densities of the objects. This kind of information, however, will be hopefully available in the next decade, based on the data provided by the Gaia mission of the European Space Agency, scheduled for launch in early 2012 (Mignard et al. 2007).

Gaia will be the most powerful tool ever developed for remote sensing, in terms of astrometric accuracy. The measurement of the tiny de°ections caused by mutual close encounters involving the largest main belt asteroids, is expected to produce reliable mass measurements for about 100 objects. Coupled with size measurements, directly obtained by Gaia signals, average densities will be also obtained for the same asteroids, which will belong to a variety of taxonomic classes. This will be a first, fundamental step to get essential information about the overall composition and plausible internal structure of a large variety of main belt asteroids. Obtaining reliable clues concerning the internal structure is currently the Holy Grail of asteroid science. Not only this is an essential information to understand these elusive bodies, but it is also a necessary pre-requisite to develop credible systems of defense and mitigation of the damages from impacts of interplanetary objects with the Earth.

4. Asteroid Impact Hazards

The topic of catastrophic impacts with celestial bodies is in some respect a dangerous one from the point of view of being able to provide correct scientific information to non-specialists. The danger is to feed non-rationale and unmotivated beliefs, and to provide information that can be misunderstood and used to support a general demand of catastrophism that has nothing to do with science.

The particular characteristic of the asteroid hazard is that of being related to events that are extremely rare, but that may produce extraordinarily heavy consequences when they occur. Since the concept of risk is related to both the probability of the occurrence of a certain phenomenon, as well as to the likely consequences of such event, the asteroid hazard can reach some level of risk that is comparable with that of other natural events that are much more common, including disasters such as floods and earthquakes.

Figure 8: Different models of the cumulative absolute magnitude distribution of near-Earth asteroids. The corresponding diameters, corresponding impact energy, and frequencies of impact are also shown. Data updated to 2003.
Origin of the plot: NASA.

Figure 8 and that date within, has been presented by a number of authors at different International meetings, and provides a nice summary of our knowledge of the near-Earth asteroid population as a function of the absolute magnitude. The same data are also converted into approximated sizes and corresponding impact energies in case of impact. The impact frequencies corresponding at different sizes are also shown. Figure 8 shows the situation in 2003, but not much has changed since then, apart from a large increase in the number of near-Earth objects discovered by sky surveys.

One key-point when interpreting Figure 8 is the uncertainty in our knowledge of the size distribution of NEAs, due to the fact that the population is fully known only down to a certain completeness limit in magnitude (size). At smaller sizes, an extrapolation is needed, and different models are possible. A general consensus is that the current NEA population should include about 1,000 objects larger than 1 km. This size is conventionally considered to mark the threshold between global and local catastrophes in terms of corresponding effects on the biosphere. At smaller sizes the impact energy decreases sensibly, but the number of possible impactors increases much, as shown by the predictions of different models plotted in Figure 8.

In any case, it is certain that the Earth has been hit many times by wandering asteroids during its history. Impact craters tend to disappear over long timescales due to erosion and tectonic phenomena, yet many craters are still clearly visible on the Earth surface, as shown in some Figures in what follows. It is believed that at least in some cases the impacts may have been highly energetic, with profound consequences on the terrestrial biosphere.

To make a well known example, the complete eradication of the dinosaurs 65 million years ago is believed by many scientists to have been a consequence of the impact of an asteroid was about 6-miles in size (10-km) (Alvarez, 2008). The explosion was probably the equivalent of about 200 million megatons of dynamite (or hundreds of nuclear bombs). However, not just the blast, but the prolonged effects on the atmosphere and climate, all contributed to the mass extinctions which followed. The hypothesis that an asteroid impact led to the extinction of the dinosaurs is not universally accepted, a competing theory being that of an episode of paroxysmic volcanic activity in the Deccan region in India. However, the point is that the impact with a 10-km asteroid certainly delivers an amount of energy that is more than sufficient to have catastrophic consequences on the biosphere, and events of this kind have certainly occurred during the Earth's history, independent on the exact event which caused the dinosaurs' extinction.

Some scientists even believe that an asteroid impact may have split apart the giant super-continent, Pangea, 250 mya (Joseph 2000), whereas others have proposed that 65 mya a 25-mile-wide (40-kilometer-wide) asteroid slammed into what today is India, causing it to crack and shatter, propelling part of the subcontinent into Africa and leaving the Seychelles islands behind (Chatterjee et al., 2009).