Solar system

Sandcastles in space

Analysis of a kilometre-sized, near-Earth asteroid shows that forces weaker than the weight of a penny can keep it from falling apart. This has implications for understanding the evolution of the Solar System. See Letter p.174

Our logical concepts for how asteroids should behave have taken another knock, as evidenced in a paper by Rozitis et al.1 on page 174 of this issue. The researchers establish that a kilometre-sized, near-Earth asteroid known as (29075) 1950 DA is covered with sandy regolith (the surface covering of an asteroid) and spins so fast — one revolution every 2.12 hours — that gravity alone cannot hold this material to its surface. This places the asteroid in a surreal state in which an astronaut could easily scoop up a sample from its surface, yet would have to hold on to the asteroid to avoid being flung off.

Rozitis and colleagues show that for this rubble-pile asteroid (the body has a porosity of roughly 51%) to stay in one piece, it must have cohesive strength — just not very much. On the basis of the density, size and shape of 1950 DA, the authors find that the asteroid requires a cohesive strength of at least 64 pascals to hold all of its rubble-pile components together: similar to the pressure that a penny exerts on the palm of your hand.

This strength is consistent with, but much more precisely determined than, similar levels of cohesive strength that have been deduced for rubble-pile asteroids on the basis of spin-rate and size statistics of asteroids2 and on the inferred strength, size and spin rate of the active asteroid P/2013 R3 (ref. 3). This asteroid was recently observed to comprise several chunks that are slowly escaping from each other, probably owing to rotational disruption4. A model for how to generate such a modest level of strength in geophysical bodies has been hypothesized2, and achieves this through 'dry cohesion' arising from van der Waals forces between components of a rubble pile (Fig. 1). In this theory, the finest grains (potentially as small as 1–10 micrometres) in a rubble pile that are present in sufficient numbers to connect all larger grains provide a very weak cement that can hold the body together — a fairy-dust version of a sandcastle.

Figure 1: Cohesive forces in regolith.

Computer simulations2 of two metre-sized boulders (pink spheres) with loosely packed centimetre-sized regolith (green and blue particles) between them. The whole system is under self-gravitational attraction, and to determine its strength, the boulders are pulled apart with an increasing force. Panels a and d show the initial configuration, with the pulling force exactly equal to the gravitational attraction. Panels b, e, and c, f show the system response for equal forces beyond the gravitational limit. If the regolith has no cohesive strength (panels ac), it immediately separates from the boulders once they are pulled with a force greater than their gravitational attraction, which leaves the regolith behind to aggregate under its own self-gravity and provides no extra strength to the system. If there are cohesive van der Waals forces between the regolith particles (panels df), the particles serve as a glue and strengthen the bond between the boulders. The level of cohesion required to hold rubble-pile asteroid (29075) 1950 DA together, found by Rozitis et al.1 to be 64 pascals, can be generated by a loosely packed regolith with particles as fine as roughly 10 micrometres2.

Although this image of fairy-castle asteroids is entertaining, the implications of these measurements are far-reaching. A defining feature of the rubble pile 1950 DA is that it is globally in a microgravity environment — the centrifugal forces from its rapid spin rate are nearly balanced by its gravitational attraction, with the difference between them being a tiny fraction of Earth's gravity. In such a regime, weak van der Waals forces can dominate5. The evident stability of such a strange body as 1950 DA exposes our ignorance of how the geophysics of asteroids works in the microgravity regime, with its current state being difficult to reconcile with classical views of how rubble-pile bodies form from catastrophically disrupted parent bodies. Although Rozitis et al. lay out a plausible story for the current state of 1950 DA, the development of a complete theory of microgravity geophysics could have significant consequences, beyond this single case, for our evolving understanding of asteroids and the Solar System.

For asteroids, the larger implications of such a weakly cohesive material — for example, the dissipation of energy in their interiors6, the shedding of material from their surfaces7 and the creation of binary asteroid systems through the fissioning of rapidly rotating rubble piles8,9 — have yet to be fully explored and understood. Going beyond asteroids, many different bodies and environments in the past and present Solar System lie in microgravity regimes similar to that of 1950 DA, where inertial, gravitational and weak molecular forces may be simultaneously relevant. The effects of the interplay of these forces in, for instance, the creation and destruction of transient structures in planetary ring systems and the accretion of grains in protoplanetary disks all become ripe topics for investigation motivated by this example.

Coming back to near-Earth asteroids, this result and the underlying theory also have ramifications for the exploration of small asteroids such as 1950 DA, currently a topic of great interest to national space agencies and a few private corporations. Small amounts of cohesion in an asteroid's regolith can enable its surface to become 'perched', just waiting for a meteorite impact (or passing astronaut) to destabilize it — similar to avalanches on Earth. The global strength of such rubble-pile asteroids held together by these weak forces is also unclear. How often might avalanches consume the entire body, causing it to split and disassemble? Recent observations of active asteroids seem to indicate that such natural outcomes might not be that rare4,7.

The ability for human or robotic interactions to create such global changes to a small asteroid suggests an intriguing vision of geophysical laboratories in space. Given that small, near-Earth asteroids are accessible using spacecraft, it becomes possible to do controlled geophysical experiments on these bodies that result in global and locally measurable changes. This would allow us to probe the geophysics of microgravity aggregates in their natural environments, and to do so at scales that cannot be recreated on Earth or in Earth's orbit, at the cost of a modest planetary-science mission.

Independent of whether we choose to take advantage of such natural laboratories in the near future, humanity might eventually have no choice, because 1950 DA is due to pay an uncomfortably close visit to Earth. The asteroid is one of the most potentially hazardous known, with a 1 in 4,000 chance of impacting the Earth in the year 2880 (ref. 10). Such an impact could have planet-wide consequences owing to the asteroid's size. Among the many proposed methods for deflecting this hazard is to run a massive spacecraft into it at high speed, or to set off a nuclear blast in close proximity11. However, for this weakly bound body, we should wonder whether such an attempt would make it crumble and fall apart like a sandcastle that has been baked in the sunshine.

Whether the impact from such a disaggregated asteroid would pose a larger threat to Earth has been a matter of debate in the scientific community. Whereas a single asteroid packs a larger punch, the shotgun spray from a disaggregated body may hit multiple sites across the globe. For a rapid rotator such as 1950 DA, however, this is not a relevant question. Once released from each other, the speed of the body's components relative to the asteroid's centre of mass would range from tens of centimetres per second if it split in half, to up to 50 cm s−1 for material that might break off its surface. These speeds are much greater than most mitigation techniques could deliver to the parent asteroid. This would cause the components to drift relative to the initial impact trajectory by more than one Earth radius in less than a year — sparing humanity from having to resolve such a delicate question.


  1. 1

    Rozitis, B., MacLennan, E. & Emery, J. P. Nature 512, 174–176 (2014).

    CAS  Article  ADS  Google Scholar 

  2. 2

    Sánchez, P. & Scheeres, D. J. Meteorit. Planet. Sci. 49, 788–811 (2014).

    Article  ADS  Google Scholar 

  3. 3

    Hirabayashi, M., Scheeres, D. J., Sánchez, D. P. & Gabriel, T. Astrophys. J. Lett. 789, L12 (2014).

    Article  ADS  Google Scholar 

  4. 4

    Jewitt, D. et al. Astrophys. J. Lett. 784, L8 (2014).

    Article  ADS  Google Scholar 

  5. 5

    Scheeres, D. J., Hartzell, C. M., Sánchez, P. & Swift, M. Icarus 210, 968–984 (2010).

    Article  ADS  Google Scholar 

  6. 6

    Goldreich, P. & Sari, R. Astrophys. J. 691, 54 (2009).

    Article  ADS  Google Scholar 

  7. 7

    Jewitt, D., Agarwal, J., Weaver, H., Mutchler, M. & Larson, S. Astrophys. J. Lett. 778, L21 (2013).

    Article  ADS  Google Scholar 

  8. 8

    Walsh, K. J., Richardson, D. C. & Michel, P. Nature 454, 188–191 (2008).

    CAS  Article  ADS  Google Scholar 

  9. 9

    Jacobson, S. A. & Scheeres, D. J. Icarus 214, 161–178 (2011).

    Article  ADS  Google Scholar 

  10. 10

    Farnocchia, D. & Chesley, S. R. Icarus 229, 321–327 (2014).

    Article  ADS  Google Scholar 

  11. 11

    Ahrens, T. J. & Harris, A. W. Nature 360, 429–433 (1992).

    Article  ADS  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Daniel J. Scheeres.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Scheeres, D. Sandcastles in space. Nature 512, 139–140 (2014).

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.


Quick links

Sign up for the Nature Briefing newsletter for a daily update on COVID-19 science.
Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing