News & Views | Published:


The devil is in the dust

Naturevolume 424pages10081009 (2003) | Download Citation


Mars is a highly dynamic planet — at least as far as dust is concerned. A better knowledge of how dust is lofted into the atmosphere will help to untangle the complex evolutionary history of the planet's surface.

New laboratory work by Greeley et al.1 and numerical flow simulations by Toigo et al.2, both appearing in the Journal of Geophysical Research, provide insights into the role of 'dust devils' in injecting dust into the martian atmosphere.

Dust transport is a key to the long-term evolution of the martian surface, much of which has undergone episodes of erosion, deposition, burial and exhumation over the past four billion years3. Although flowing water is believed to be responsible for some of this surface change, lifting and redistribution of dust by wind has had an important role over that entire period, and may be the largest factor today. Suspended dust also influences atmospheric dynamics by absorbing solar radiation, efficiently heating the thin atmosphere and thereby contributing to the driving of large-scale winds. Atmospheric dust content has a seasonal maximum when Mars is closest to the Sun, but the maximum dust load varies from one seasonal cycle to the next. Occasionally, large dust storms develop in several areas simultaneously, spreading a dense pall over the planet and obscuring much of the surface for weeks at a time4.

Computer models of Mars' atmospheric general circulation do a reasonable job of simulating the dynamics of the dust-laden atmosphere5,6, but they are constrained by our limited understanding of how dust is lifted from the surface. The stress exerted on the surface by wind can lift dust particles into suspension. Because of the low atmospheric density, the minimum threshold wind speed for lifting is an order of magnitude larger on Mars than on Earth. Moreover, dust particles (of diameter less than 60 μm) require much stronger winds for lifting than do small, sand-size particles (roughly 60–200 μm), because inter-particle cohesion effectively locks the smaller dust particles into the surface. Dust is injected into the atmosphere when sand-size particles are lifted and hop or 'saltate' along the ground, knocking the dust particles into suspension7. This mechanism requires the coexistence of dust and sand-size particles, and martian surface wind speeds in excess of 30 m s−1. Such high wind speeds are relatively rare, and sand-size particles may not be very widespread. Nevertheless, large dust storms are remarkably common. More than 5,000 have been observed during two martian annual cycles in images obtained with the Mars Orbiter Camera on board the Mars Global Surveyor orbiter8.

Injection of dust into the martian atmosphere by a distinctly different phenomenon — dust devils — has also been widely observed3,9,10. Dust devils are common in arid regions on Earth during the hot season: strong daytime heating of the surface generates intense turbulence and convective plumes, and air converging into these plumes tends to conserve any initial angular momentum. If the plume and initial angular momentum are sufficiently intense, a spinning vortex develops that picks up dust to form the visible dust devil (Fig. 1a). Because of the intense daytime heating, such convectively driven 'boundary-layer vortices' on Mars can be much larger than on Earth, with diameters up to 1 km and heights up to 8 km. Surface streaks produced by dust devils are widespread3 (Fig. 1b). Most of these apparently form when the dust devil removes a thin layer of bright dust from a darker substrate. Subsequent sedimentation of dust can rapidly cover these streaks.

Figure 1: Evidence of dust devils: exhibits a and b.
Figure 1


a, In this picture, taken on 10 April 2001, the Sun is illuminating the surface of Mars from the left, so creating a dark shadow of the wispy dust devil which can be seen in the centre of the image. The faint line to the left is the track of the dust devil; its height, estimated from the shadow, was just over 1 km. b, Streaks on the martian surface, interpreted as dust-devil tracks. Such tracks often run across surface features such as dunes and boulders. They are usually dark (but not always, as in a), which is thought to be due to removal of bright dust, revealing darker, underlying material. The area depicted here is about 3 km by 5 km. Both images were taken by the Mars Orbiter Camera.

It has long been suspected that dust is more easily lifted by dust devils than by wind stress and saltation alone7. The new experiments verify this and demonstrate the lifting mechanism. Greeley et al.1 have generated laboratory vortices whose scaled pressure and flow fields closely match those observed in terrestrial dust devils and in vortices observed at the Mars Pathfinder landing site. They investigated thresholds for particle lifting for a wide range of particle diameters and densities, and a range of vortex diameters, at pressures ranging from terrestrial to martian. In addition to lifting by the wind-stress mechanism, low pressure in the laboratory vortex core induces anomalous upward pressure on particles that counters inter-particle cohesion. This 'vacuum-cleaner effect' acts most efficiently on the smallest particles, so that dust is lifted about as easily as sand-size particles.

If low dust-devil core pressure works in the same way when scaled up to martian convective vortex conditions, these experiments show that the wind-speed threshold for dust lifting by convective vortices would be much lower than for wind stress alone, and that saltation by sand-size particles would not be required. This may explain the surprising ubiquity of dust devils and wind streaks on Mars. Whether a similar vacuum-cleaner effect caused by intense small vortices within larger dust storms has a role in lifting dust is not known, but this possibility may help to explain the large number and wide distribution of the large dust storms.

Computer flow-simulation models can also provide insights into the dynamics of boundary-layer vortices. By adapting such a model to martian conditions, Toigo et al.2 have been able to simulate the evolution of convective vortices with and without strong background wind. These vortices closely resemble the vortices seen on Earth and Mars, as well as the laboratory vortices produced by Greeley and colleagues. They form distinct coherent structures, where intense updraughts enhance rotating motion through convergence of the horizontal flow. Although the simulated Mars vortex wind and pressure anomalies do not yet reach the threshold for dust lifting found by Greeley et al., such computer simulations promise to be an important tool for interpreting laboratory and field observations of dust devils. Integration of the laboratory dust-lifting results with computer vortex models may soon permit reliable simulation of the complete process of dust injection into the atmosphere by dust devils. Together, these new results1,2 significantly advance our ability to simulate dust lifting, and promise to contribute to our understanding of martian surface evolution.


  1. 1

    Greeley, R et al. J. Geophys. Res. 108, doi:10.1029/2002JE001987 (2003).

  2. 2

    Toigo, A., Richardson, M., Ewald, S. & Gierasch, P. J. Geophys. Res. 108, doi:10.1029/2002JE002002 (2003).

  3. 3

    Malin, M. & Edgett, K. J. Geophys. Res. 106, 23429–23570 (2001).

  4. 4

    Zurek, R. et al. in Mars (eds Kieffer, H. H., Jakosky, B. M., Snyder, C. W. & Matthews, M. S.) 835–933 (Univ. Arizona Press, Tucson, 1992).

  5. 5

    Leovy, C. Nature 412, 245–249 (2001).

  6. 6

    Murphy, J. et al. J. Geophys. Res. 100, 26357–26376 (1995).

  7. 7

    Greeley, R., Lancaster, N., Lee, S. & Thomas, P. in Mars (eds Kieffer, H. H., Jakosky, B. M., Snyder, C. W. & Matthews, M. S.) 730–766 (Univ. Arizona Press, Tucson, 1992).

  8. 8

    Cantor, B. 6th Int. Mars Conf. Pasadena, California, Abstr. 3166 (Lunar & Planetary Inst., Houston, Texas, 2003).

  9. 9

    Thomas, P. & Gierasch, P. Science 230, 175–177 (1985).

  10. 10

    Metzger, S., Carr, J., Johnson, J., Parker, T. & Lemmon, M. Geophys. Res. Lett. 26, 2781–2784 (2002).

Download references

Author information


  1. Professor Emeritus of Atmospheric Sciences, University of Washington, Seattle, 98105, Washington, USA

    • Conway B. Leovy


  1. Search for Conway B. Leovy in:

Corresponding author

Correspondence to Conway B. Leovy.

About this article

Publication history

Issue Date


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.

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