Dust storms have been observed on the surface of Mars since the early 19th century when H. Flaugergues reported "yellowish clouds" in 1809 (Zurek, 1992). Previous observers as early as the late 17th century such as Giacomo Filippo Maraldi had noted temporal variations in the equatorial dark regions and the polar caps (Kieffer et al, 1992), and until modern times the light and dark regions were interpreted as land and sea areas, with the temporal changes thought to be seasonal variation in vegetation and polar ices (e.g., Percival Lowell ). While relatively regular observations of Mars have been made since about 1901, it wasn't until 1956 that the first major global storm was noted, and it was at this time that Kuiper (1957) suggested that the dark areas were lava fields and the seasonal wave of darkening was due to the redistribution of dust (Zurek, 1992).

Mariner 9 Image of Dust Storm (click to enlarge)

In stark contrast to the lunar-like landscapes seen by earlier flyby missions ( Mariners 4, 5, and 6)  Mariner 9 arrived at Mars in 1971 during the largest dust storm ever seen (a status retained until 2001); see image at left. This storm dramatically demonstrated how extensive, long-lived, and opaque Martian dust storms could be, far exceeding anything ever observed on Earth , with dust raised to 60 km or higher in the atmosphere (Greeley et al, 1992). Infrared data from Mariner 9 also revealed that the atmospheric thermal structure and circulation was significantly altered by solar heating of the dust (Zurek, 1992). Later, Viking lander and orbiter data showed dust opacity ( optical depth ) ranging from 0.5 to 9 over the landers, and tremendous variation in quantity and distribution of airborne dust, as 2 global storms came and went during the 1st and 4th Mars years of observations. Interestingly, wind velocities at the landers never exceeded 30 m/s and no surface material was observed to have moved, although brightness and contrast were altered after each storm, presumably due to dust settling from the atmosphere (Greeley et al, 1992).

In order to understand how this dust affects the atmosphere, it is critical to know the physical properties of the dust. Several studies (e.g., Toon et al, 1977; Wells et al, 1984) have undertaken this task, using orbiter and lander data to estimate dust parameters such as composition (igneous silicates, basalt, or their weathering products), particle size distributions (predominantly 1-10 microns), and sedimentation rates (~2x10 ^-3 g cm^-2 yr ^-1) as well as optical properties .

MGS-TES vs. MGCM (click to enlarge)

Another important area of research aimed at improving our understanding of these dust storms involves the use of global climate models (GCMs). Much progress has been made over the last 20 years in characterizing the general circulation of the atmosphere (e.g., Haberle et al, 1993 ). The figure from Leovy (2001) at the right shows a comparison of the Haberle et al model with data from the MGS thermal emission spectrometer , giving some indication of the accuracy of these models.

One model aimed at simulating the evolution of a Martian dust storm by Murphy et al (1995) provides some insight into how dust and the atmosphere interact during a storm. Using a 3D version of a Martian GCM with simultaneously evolving thermal, dynamical, and radiatively active dust fields, they artificially induced a dust supply in a confined latitude zone in the southern subtropics, letting the model run for 40 days or more. Several phenomenon quite similar to that observed on Mars resulted:

1. Hadley circulation and convection transported fine (~1 micron) dust up to 40 km altitude within a few days of storm onset, then spread it rapidly toward the poles and high elevations.
2. Traveling and stationary eddies transported dust to mid and high northern (winter) latitudes.
3. Northern (winter) hemisphere baroclinic waves weakened at onset, then intensified as the storm waned.
4. Thermal tidal amplitude increased at the onset of the storm.
5. Pressure response at Viking Lander 1 site agreed with observations.

While GCMs are far from perfect (e.g., none predicted the rapid cross-equatorial expansion observed in the 2001 event), they typically reproduce the overall circulation patterns quite accurately, and thus provide us with a level of understanding of the links between the global and regional wind patterns and the evolution of Martian dust storms which is not obtainable solely from observations.

Other links:
NASA-Ames Mars Atmospheric Modeling Group
NASA-Goddard GCM page
Atmospheric Optics Glossary

Introduction

Onset

Evolution

Cessation

Conclusions

References