Planetary science

Seeing double in the Kuiper belt

A small fraction of Kuiper-belt objects are known to be accompanied by large moons. These double worlds may have formed in the earliest days of the Solar System through comparatively gentle gravitational encounters.

Beyond the orbit of the planet Neptune, in the deep freeze of the outermost region of the Solar System, there resides a vast population of icy worlds. The Kuiper belt is the source of many of the comets that periodically pass through the inner Solar System, and it offers planetary scientists a valuable laboratory for studying the accretion of planet-forming material and for understanding the evolution of the outer Solar System1. The formation of binary systems, objects with moons, in the Kuiper belt is the subject of much recent attention. Although a number of binaries have indeed been observed there, formation models that work for other small-body populations seem not to work as well or at all in the Kuiper belt. But on page 643 of this issue2, Goldreich, Lithwick and Sari propose a new mechanism for the formation of these binaries that is backed up by observational data.

Until the discovery of the first Kuiper-belt object (KBO) in 1992, Pluto — 2,300 km in diameter and composed of a mix of ice and rock — seemed to be an anomalous interloper beyond the realm of the gas-giant planets (Fig. 1). Today, over 600 new worlds are known to reside in the Kuiper belt, along with an estimated population of some 105 objects larger than 100 km in diameter. These objects now provide a natural context for Pluto as the largest member (so far) of a disk of icy minor planets, left over from the formation of the outer planets. Similarly, Pluto's large moon Charon seems a little less of an exception now that a few per cent of KBOs have been observed to have large satellites of their own.

Figure 1: Pluto and its satellite Charon.


The outermost planet of the Solar System resides in the Kuiper belt and is one of a number of objects there to have a satellite (left panel). How such binary systems formed in the Kuiper belt is not understood, but Goldreich et al.2 present a model of soft gravitational interactions that could account for their existence. Pluto is the most extensively studied object in the Kuiper belt. This true-colour image from ground-based observations (right panel) suggests that the planet's surface is dominated by frozen nitrogen, with traces of methane and carbon monoxide. A NASA mission to Pluto, and the Kuiper belt, will be launched in 2006.

The first KBO satellite was discovered only last year3, and since then another six binary systems have been observed4 through direct imaging, using both ground-based telescopes and the Hubble Space Telescope. The total number of satellite systems now stands at eight (including Pluto–Charon), out of 618 known KBOs. In the binaries discovered so far, the two components are similar in size but are relatively far apart, separated by some 60 to 1,000 times the radius of the larger object (with the exception of Pluto–Charon, where Charon's orbital radius is only about 17 times the radius of Pluto).

Just a decade ago, no one knew for sure whether minor planets such as KBOs and asteroids could even have satellites. Today, dozens of such systems are known in small-body populations throughout the Solar System. In addition to the diversity they bring to the bestiary of Solar System objects, the discovery and observation of KBO and asteroid satellites provides a wealth of information on the physical properties of these objects. From observations of the sizes and periods of satellite orbits, the total mass of the binaries and the individual component masses can be derived. In turn, determinations of the bulk densities of individual objects yield crucial insights into the composition and internal structure of these distant worlds.

The diverse history and evolution of minor planets and their satellites are further illuminated by models for the formation of binary systems5. In both the main asteroid belt (between Mars and Jupiter) and the Kuiper belt, mutual collisions play an important, if not dominant, role in shaping the properties of these objects. So mechanisms involving impacts are an obvious possibility for forming satellite systems: material ejected from cratering impacts might accrete into orbiting satellites; rotational fission of target objects might be induced by large oblique impacts; and debris from the break-up of a target body might form gravitationally bound pairs. Indeed, computer simulations of impacts between asteroids are proving successful in generating model satellite systems with properties that are consistent with those of some main-belt asteroid binaries6. Models for the formation of the tightly bound Pluto–Charon pair also seem to favour giant impacts similar to that thought to be responsible for the formation of our own Moon7.

But for the large, widely separated KBO satellites, a collisional origin is hard to understand — unless the target objects were substantially smaller than predicted8. Collisions between large KBO targets and large projectiles (required to loft enough satellite-forming material into orbit) are so rare in the present-day Kuiper belt that this mechanism cannot explain the observed systems. Also, impact simulations tend to produce satellites with orbital separations of only a few times the primary's radius.

Of course, there may be many smaller, as yet undetected binary systems that were indeed formed by the same collisional mechanisms as found in the main asteroid belt. But more plausible mechanisms for the formation of large, widely separated binary pairs may involve close encounters between objects of comparable mass during primordial times, when the Kuiper belt was more populous and dynamically 'colder' than it is today.

One model9 proposes that under these conditions a physical collision can occur between two objects of comparable size while under the gravitational influence of a third object. The combined body resulting from that impact can remain bound to the third body, forming a satellite, if its speed after the collision is less than the speed needed to escape the gravity of the third body. The volume of space in which such collisions can occur increases with distance from the third body, so the number of satellites formed by this mechanism should increase with increasing orbital separation.

Goldreich and colleagues2 propose instead that transient binaries may form when two large objects approach near enough to each other that their mutual gravitational attraction is stronger than the disruptive tidal force of the distant Sun. The transient binary is then stabilized as energy is lost from the pair through numerous encounters from passing smaller objects ('dynamical friction') or by gravitational interaction with a nearby third large body. Further loss of orbital energy through continued dynamical friction 'hardens' the binaries, tending to reduce the orbital separations. So, if Goldreich et al. are right, in the present-day Kuiper belt the probability of finding satellites should be greater for smaller values of orbital radius. The model further predicts that the ongoing Hubble Space Telescope survey for Kuiper-belt binaries should find that about 5% of KBOs are accompanied by large satellites — and this is indeed roughly compatible with observations made to date.

Despite the seeming conflict between the predictions of these two models, and the fact that there remain some crucial details of timing to be worked out, both models offer plausible mechanisms for binary formation in the Kuiper belt, which may even have acted in concert in primordial times. As new KBO binaries are discovered, their properties — particularly the statistics of their orbital separations — will provide important constraints on these models and others. A spacecraft mission to Pluto–Charon and the Kuiper belt will provide wonderfully detailed information on the physical properties of at least one KBO binary and remains a top priority in outer Solar System exploration10. The exciting prospect of testing competing binary formation models with ongoing observations exemplifies a healthy state of affairs in exploring the frontiers of the Solar System.


  1. 1

    Farinella, P., Davis, D. R. & Stern, S. A. in Protostars and Planets IV (eds Mannings, V., Boss, A. P. & Russell, S. S.) 1255–1282 (Univ. Arizona Press, Tucson, 2002).

    Google Scholar 

  2. 2

    Goldreich, P., Lithwick, Y. & Sari, R. Nature 420, 643–646 (2002).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Veillet, C. et al. Nature 416, 711–713 (2002).

    ADS  CAS  Article  Google Scholar 

  4. 4

  5. 5

    Merline, W. J. et al. in Asteroids III (eds Bottke, W. F., Cellino, A., Paolicchi, P. & Binzel, R. P.) 289–312 (Univ. Arizona Press, Tucson, 2002).

    Google Scholar 

  6. 6

    Durda, D. D., Bottke, W. F., Asphaug, E. & Richardson, D. C. Bull. Am. Astron. Soc. 33, 1134 (2001).

    ADS  Google Scholar 

  7. 7

    Canup, R. M. & Asphaug, E. Nature 412, 708–712 (2001).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Stern, S. A. Astron. J. 124, 2300–2304 (2002).

    ADS  Article  Google Scholar 

  9. 9

    Weidenschilling, S. J. Icarus 160, 212–215 (2002).

    ADS  Article  Google Scholar 

  10. 10

    National Research Council New Frontiers in the Solar System: An Integrated Exploration Strategy (2002);

Download references

Author information



Corresponding author

Correspondence to Daniel D. Durda.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Durda, D. Seeing double in the Kuiper belt. Nature 420, 618–619 (2002).

Download citation

Further reading


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.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

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