News & Views | Published:

Planetary science

Rocks that go bump in the night

Naturevolume 417pages697698 (2002) | Download Citation


The planets were probably created by collisions between smaller rocky bodies over many millions of years. The identification of a recently formed asteroid family will tell us much about the dynamics of these collisions.

Between the orbits of Mars and Jupiter lie the remains of the early, violent history of the Solar System. These are asteroids, rocky bodies that range in diameter from a few hundred metres to just under 1,000 km. Over the past century, astronomers have realized that asteroids (and comets, their icy cousins originating further out in the Solar System) are the building blocks of planets. On page 720 of this issue, Nesvorný and colleagues1 report their discovery of an asteroid family that offers unprecedented insight into the dynamics of asteroid collisions — and hence into how the planets of the Solar System formed.

For planets to grow in size from their humble beginnings, they must be continually bombarded by smaller bodies for tens to hundreds of millions of years. This process still happens today: not only do these small bodies continue to strike the Earth, with effects ranging from spectacular fireballs to mass extinctions, but they continue to bump into each other as well. In the past these bumps were relatively gentle, which allowed the first protoplanets to form. But as giant planets such as Jupiter came into existence, the gravitational fields they generated stirred up the smaller bodies. This energy increase made them more likely, if they collided, to break apart than grow, with the result that only the largest protoplanets survived to form planets while the rest were ground down to small fragments.

Nesvorný and colleagues1 have uncovered the aftermath of one such energetic collision. What makes their discovery unique and exciting is that a large-scale break-up event has, for the first time, been precisely dated. Piecing together the fragments, Nesvorný et al. find that a 25-km asteroid in the Outer Main Belt between Mars and Jupiter broke up in collision 5.8 million years ago — a mere blink of an eye in the 4.5-billion-year lifetime of the Solar System.

Theorists studying planet formation would like nothing better than to smash two asteroids together and see what happens. Many questions could be answered. Under what conditions does an asteroid break apart? What happens to the debris? How do pre-existing fractures affect the outcome? Could a 'doomsday' asteroid threatening Earth be stopped by brute-force methods such as a nuclear blast? Of course, full-scale experiments are not possible, so planetary scientists rely on much smaller laboratory experiments as well as numerical simulations to predict collision outcomes2,3. But nature has provided a larger laboratory for us in the form of the ancient record of past collisions between asteroids. By studying this record, we can learn about the collision process.

One way to understand the physics of large-scale impacts among asteroids is to investigate asteroid families, groups of asteroids that share similar orbital elements and spectral properties4. Orbital elements describe the shape and orientation of the path that an object follows around the Sun, whereas spectral properties provide information about the composition of an object according to how it reflects sunlight at different wavelengths.

Asteroid families are formed during catastrophic collisions, as the fragments are launched away from the impact site at high speeds5. The nature of each break-up event and the velocity distribution of the ejecta depend critically on the incoming projectile's mass and velocity relative to the collision target. The ejection speed of the fragments acts as an impulse that gives each member a slightly different orbit from the parent body. In the absence of any further disturbance, these orbits would remain unchanged.

In reality, however, these objects are subject to dynamical evolution through several mechanisms: subsequent collisions between family members and other nearby asteroids, after the family has formed; gravitational perturbations produced by planets and large asteroids; and non-gravitational forces such as the Yarkovsky effect that slowly cause the orbits of smaller, kilometre-sized asteroids to change over time6. (The Yarkovsky effect changes the spin and orbit of a body by the asymmetric re-radiation of thermal energy absorbed from the sun7.) Together, these effects slowly muddle the memory of the family-forming impact, making it difficult both to identify the family and to piece together exactly what happened. In fact, the degree of dynamical diffusion can be used to estimate how long ago the original impact occurred, although it is not a precise measure.

Until now, most known asteroid families were thought to be the by-products of huge disruption events among bodies hundreds of kilometres in diameter8. Most of these families are estimated to be hundreds of millions, even billions, of years old (Fig. 1), so it is problematic to use their properties to constrain asteroid evolution models. Using a new database of orbital elements, however, Nesvorný and colleagues1 have identified a cluster of 39 small asteroids inside the Koronis asteroid family that were probably produced by the recent disruption of a 25-km asteroid.

Figure 1: Aged asteroid.
Figure 1


One of only a handful of asteroids that have been visited by spacecraft, Ida (pictured here) is a member of the Koronis family that formed over a billion years ago as a result of a catastrophic collision between two larger bodies. But the Karin asteroid cluster discovered by Nesvorný and colleagues1 is only a few million years old and may greatly improve our understanding of collisional dynamics and planet formation.

The youth of this cluster is suggested by two lines of evidence. First, the orbital elements of the cluster members are remarkably similar. Second, the researchers were able to determine the precise age of the break-up event by numerically integrating the orbital elements of 13 cluster members back through time. They found that particular elements (corresponding to the orbital orientation of each body) converged to a single value 5.8 ± 0.2 million years ago. Because this technique does not work on asteroid clusters that have suffered significant dynamical evolution, this is the first time that an asteroid break-up has been accurately dated.

The Karin cluster — as Nesvorný et al. have named it, after its largest member — is a compelling target for a space mission. The cluster is young enough that many erosional and weathering processes thought to occur on asteroid surfaces9 may not have had time to erase the tell-tale signatures of the break-up. So it might be possible to determine whether the family members are intact fragments or gravitational re-accumulations of smaller pieces. This new cluster will no doubt be the focus of attention for the asteroid community for some time. Meanwhile, the search for ever younger families will continue, in the hope of taking us closer to understanding the origins of our Solar System.


  1. 1

    Nesvorný, D., Bottke, W. F. Jr, Dones, L. & Levison, H. F. Nature 417, 720–722 (2002).

  2. 2

    Housen, K. R., Holsapple, K. A. & Voss, M. E. Nature 402, 155–157 (1999).

  3. 3

    Asphaug, E., Ostro, S. J., Hudson, R. S., Scheeres, D. J. & Benz, W. Nature 393, 437–440 (1998).

  4. 4

    Zappalà, V., Bendjoya, Ph., Cellino, A., Farinella, P. & Froeschle, C. Icarus 116, 291–314 (1995).

  5. 5

    Michel, P., Benz, W., Tanga, P. & Richardson, D. C. Science 294, 1696–1700 (2001).

  6. 6

    Bottke, W. F. Jr, Vokrouhlický, D., Brož, M., Nesvorný, D. & Morbidelli, A. Science 294, 1693–1696 (2001).

  7. 7

    Farinella, P., Vokrouhlický, D. & Hartmann, W. K. Icarus 132, 378–387 (1998).

  8. 8

    Durda, D. D., Greenberg, R. & Jedicke, R. Icarus 135, 431–440 (1998).

  9. 9

    Clark, B. E. et al. Meteor. Planet. Sci. 36, 1617–1637 (2001).

Download references

Author information


  1. the Department of Astronomy, University of Maryland at College Park, College Park, 20742, Maryland, USA

    • Derek C. Richardson


  1. Search for Derek C. Richardson in:

Corresponding author

Correspondence to Derek C. Richardson.

About this article

Publication history

Issue Date



By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing