Planetary science

Tracking the martian climate

Article metrics

Like Earth, Mars has experienced long-term fluctuations in climatic conditions. The cause of certain fluctuations is now identified as variation in the planet's astronomical behaviour.

In 1971, the Mariner 9 orbiter revealed extensive layered deposits on Mars at each pole at latitudes above 75°. Later spacecraft provided more detailed views of these deposits, which occur in both the northern and southern hemispheres as broad domes up to 2.5 km thick (Fig. 1). Layers in Earth's ice caps have revealed a detailed record of the climatic history of the past several hundred thousand years, and the martian polar layering presumably carries similar information. This record might tell us about temporal changes in the abundance of water on Mars, and its distribution as liquid, ice and vapour. The inferred climate changes might also provide clues to the age of young features in more temperate martian latitudes, such as gullies on slopes1 and patterned ground, that may have been formed by water or ice. On page 375 of this issue2, Laskar et al. provide the first evidence of a specific correlation of the martian sequences with variations in climate driven by changes in the axis and the orbit of Mars.

Figure 1: Relief image of the north polar cap of Mars.

The cap is underlain by layers of ice and dust exposed in the spiral pattern of troughs, which Laskar et al.2 relate to astronomically driven climate change. The image is based upon topographic data returned by the laser altimeter on the Mars Global Surveyor spacecraft. The top of the layered deposits is about 2.5 km above the surrounding plains. The radial grooves north of 88° are an artefact of sparse data.

At the martian north pole, the layered deposits are covered with a perennial cap of water ice with dust mixed in3. At the south pole the perennial cap is much smaller than the layered deposits, and is composed of carbon dioxide ice probably underlain by water ice. A roughly spiral set of troughs and scarps incised into the underlying deposits at the north pole (Fig. 1) define a succession of nearly horizontal layers on the trough slopes facing the equator. The layers have been deposited by accumulation of ice and atmospheric dust in uncertain proportions, possibly incorporating CO2 in a caged form known as clathrates. The trough slopes have presumably been exposed by ablation of water ice within the layers by evaporation and removal of the dust by wind3.

Explanations for the polar layering have centred on the effects of insolation — the solar radiation received at the martian surface — in regulating the climate. Climate change, in turn, is expected to modulate both the rate at which layers accumulate and their composition, and possibly their removal or thinning by ablation and wind erosion3. Both Earth and Mars are subject to quasi-periodic variations in the characteristics of their spin axes (precession and obliquity) and their orbital eccentricity, due to gravitational interactions with other planets and the Sun4. On Earth the variations are modest, with obliquity varying through about 2.5° and eccentricity from 0.01 to 0.05, but they have been correlated with the major climate changes accompanying the terrestrial ice ages (the Milankovitch theory5). Because Mars lacks a large moon and is closer to the massive outer planets, it undergoes much more dramatic orbital variations: during the past 10 million years, obliquity has ranged from about 13° to as much as 47°, and eccentricity from 0.00 to 0.13. These variations occur in a complicated 'beat' pattern involving several frequencies (see Fig. 2 of Laskar and colleagues' paper on page 376).

It has been difficult to decipher the presumed climatic record contained in the martian layered deposits. One issue is the uncertain connection between the orbital variations and processes that could result in the accumulation and erosion of layers. The orbital effects on seasonal and latitudinal variations in insolation are well understood (see Fig. 2 of Laskar et al.). But we know only within broad limits how insolation is linked to conditions such as atmospheric pressure and circulation patterns; dust-storm intensity; cycling of water between the layered deposits, ice in loose sediment and dust on the surface, and groundwater; and rates of accumulation or ablation of ice from polar deposits. Which of the several component cycles of climate variation correlate with the layers has therefore remained unclear.

Imprecise knowledge of the thickness and sequence of exposed layers has also limited their interpretation. Images returned by Viking allowed resolution of the layers only down to about 10–20 m thickness. The increased resolution of the camera on Mars Global Surveyor has permitted measurements to a precision of a few tens of centimetres. But it was not until data were returned from the laser altimeter on this spacecraft that the steepness of the slopes exposing the layers could be reliably estimated, allowing more accurate calculations of layer thickness.

Laskar et al.2 have statistically matched observed vertical variations in layer brightness in a north polar scarp to calculated temporal variations in polar insolation. The authors use techniques similar to those used in terrestrial correlations between quasi-periodic astronomical effects and climate 'proxies' such as tree-ring thickness, isotope concentrations, and ice and sediment thickness. They find an intriguing match between variation in insolation and layer brightness when the top 350 m of layers exposed in a polar scarp is assumed to correspond to about 0.9 million years of accumulation, giving an average rate of accumulation of about 0.05 cm yr−1.

This estimated rate of accumulation is near the high end of previous estimates (which were based on calculations of the age of surface polar deposits from impact crater density, and on the rates of exchange of atmospheric dust and water vapour between the polar cap and the atmosphere). At 0.05 cm yr−1, the entire 2.5 km of layered deposits could have accumulated within the past 5 million years, astonishingly recently given the 4.5-billion-year age of the planet.

As Laskar et al. indicate, the figure of 5 million years could be consistent with the strong seasonal insolation associated with high obliquity that occurred before then: strong insolation could have discouraged layer deposition and ablated any earlier deposits. Possibly related to this, the north polar deposits are underlain by a sand-rich layer that was probably a dune field emplaced at some time when the polar cap was absent6. On the other hand, the layered deposits might be much older if deposition of new ice and dust between troughs roughly balanced rates of ablation on the equator-facing scarps3.

The proposed youth of the north polar layers raises other questions. The density of craters on the deposits at the south pole suggests a surface age of about 100 million years7. Why would the north polar deposits have been ablated during high-obliquity cycles while those at the south pole persisted? Where would the water in the northern deposits have gone if the cap has been episodically absent in recent geological time? If distributed globally, the amount of water released might have been enough to cover the planet to a depth of several metres, and might have contributed to the formation of young gullies and ice-related landforms in temperate and equatorial latitudes1.

Laskar et al.2 studied layering on one polar scarp. A rich record exists on many other scarps in both the north and south. Further studies must exploit these other sites to examine the consistency of depositional history among the scarps and provide a longer total record of climatic change. In the future, missions that land on the planet and drill into the layered deposits may provide definitive interpretations of the climate cycles on Mars.


  1. 1

    Malin, M. C. & Edgett, K. S. Science 288, 2330–2335 (2000).

  2. 2

    Laskar, J., Levrard, B. & Mustard, J. F. Nature 419, 375–377 (2002).

  3. 3

    Thomas, P. et al. in Mars (eds Kieffer, H. H. et al.) 767–795 (Univ. Arizona Press, Tucson, 1992).

  4. 4

    Kieffer, H. H. & Zent, A. P. in Mars (eds Kieffer, H. H. et al.) 1180–1218 (Univ. Arizona Press, Tucson, 1992).

  5. 5

    Imbrie, J. Icarus 50, 408–422 (1982).

  6. 6

    Byrne, S. & Murray, B. C. Am. Astron. Soc. Div. Planet. Sci. Abstr. 48, 48.10 (2001).

  7. 7

    Plaut, J. J. et al. Icarus 75, 357–377 (1988).

Download references

Author information

Correspondence to Alan D. Howard.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Howard, A. Tracking the martian climate. Nature 419, 350–351 (2002) doi:10.1038/419350a

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.