Dust: heating or cooling?

In the left image, thin martian clouds are vis...
Mars: with and without dust storms in 2001. Image via Wikipedia

Once every 13 years on average dust blots out most of the surface of Mars turning it into an orange ball. The last such planet-encircling dust storm occurred in 2001, but lesser storms spring up on a seasonal basis. Yet Martian seasons have very different weather from terrestrial ones because of the greater eccentricity of Mars’s orbit, as well as the fact that its ‘weather’ doesn’t involve water. When Mars is closest to the Sun solar heating is 20% greater than the average, for both hemispheres. The approach to that perihelion marks the start of the dust season which last a half the Martian year. Unsurprisingly, the sedimentary process that dominates Mars nowadays is the whipping up and deposition of sand and dust, though in the distant past catastrophic floods – probably when subsurface ice melted – sculpted a volcanic landscape pockmarked with impact craters up to several thousand kilometres across. Waterlain sediments on early Mars filled, at least in part, many of the earlier craters and probably blanketed the bulk of its northern hemisphere that is the lowest part of the planet and now devoid of large craters. Erosion and sedimentation since that eventful first billion years has largely been aeolian. Some areas having spectacular dunes of many shapes and sizes, whereas more rugged surfaces show streamlined linear ridges, or yardangs (http://earth-pages.co.uk/2011/05/08/winds-of-change/), formed by sand blasting. Most of the dust on Mars is raised by high winds in the thin atmosphere sweeping the great plains and basins, and, by virtue of Stokes’s law, the grains are very much smaller than on Earth.

The dustiest times on Earth, which might have blotted out sizeable areas from alien astronomers, in the last million years have been glacial maxima, roughly every 100 ka with the latest 20 ka ago. Layering in the Antarctic ice core records such dust-dominated frigid periods very precisely. Less intricate records formed away from the maximum extent of ice sheets as layers of fine sediment known as loess, whose thickness variations match other proxy records of palaeoclimate nicely. Loess, either in place or redeposited in alluvium by rivers, forms the most fertile soil known – when the climate is warm and moist. The vast cereal production of lowland China and the prairies of North America coincides with loess: it may seem strange but a large proportion of 7 billion living humans survive partly because of dust storms during glacial periods of the past.

Being derived from rock-forming minerals dust carries with it a diverse range of chemical elements, including a critical nutrient common on land but in short supply in ocean water far offshore: iron in the form of oxide and hydroxide coatings on dust particles – the dust coating your car after rain often has a yellow or pinkish hue because of its iron content. Even when the well-known ‘fertilizer’ elements potassium, nitrogen and phosphorus are abundant in surface ocean water, they can not encourage algal phytoplankton to multiply without iron. Today the most remote parts of the oceans have little living in their surface layers because of this iron deficiency. Yet oceanographers and climatologists are pretty sure that this wasn’t always the case. They are confident simply because reducing the amount of atmospheric carbon dioxide and its greenhouse effect to levels that would encourage climate cooling and glacial epochs needed more carbon to be buried on the ocean floors than happens nowadays, and lifeless ocean centres would not help in that.

Dust plume off the Sahara desert over the nort...
Saharan dust carried over the Atlantic Ocean by a tropical cyclone. Image via Wikipedia

At present, the greatest source of atmospheric dust is the Sahara Desert (bartholoet, J. 2012. Swept from Africa to the Sahara. Scientific American, v. 306 (February 2012), p. 34-39). Largely derived from palaeolakes dating from a Holocene pluvial episode, Saharan dust accounts for more than half the two billion metric tonnes of particulate atmospheric aerosols dispersed over the Earth each year. Located in the SE trade-wind belt, the Sahara vents dust clouds across the Atlantic Ocean, most to fall there and contribute dissolved material to the mid-ocean near-surface biome but an estimated 40 million t reaches the Amazon basin, contributing to fertilising the otherwise highly leached tropical rain-forest soils. While over the ocean the high albedo of dust adds a cooling effect to the otherwise absorbent sea surface. Over land the fine particles help nucleate water droplets in clouds and hence encourages rainfall. The climatic functions of clouds and dusts are probably the least known factors in the climatic system, a mere 5% uncertainty in their climatic forcing may mean the difference between unremitting global warming ahead or sufficient cooling by reflection of solar radiation to compensate for the cumulative effects of industrial CO2 emissions.

Recording amounts of dust from marine sediments quantitatively is very difficult and impossible in terrestrial sediments, but superb records tied accurately to time at annual precision exist in ice sheets. Low dust levels in Greenland and Antarctic ice tally well with the so-called ‘Medieval Climate Anomaly’ (a warm period) whereas through the 13th to 19th centuries (the ‘Little Ice Age’) more dust than average circulated in the atmosphere. Crucially, for climate change in the industrial era, there has been a massive spike in dust reaching near-polar latitudes since the close of the 18th century during the period associated with signs of global warming: a counterintuitive relationship, but one that is difficult to interpret. The additional dust may well be a result of massive changes in land use across the planet following industrialised agricultural practices and growing population. There are several  questions: does the additional dust also reflect global warming with which it is correlated, i.e. evaporation of the huge former lakes in the Sahara (e.g. Lake Chad); is the dust preventing additional greenhouse warming that would have taken place had the atmosphere been clearer; is it even the ‘wrong kind of dust’, which may well reflect short-wave solar radiation away but also absorbs the longer wavelength thermal radiation emitted by the Earth’s surface, i.e. an aerosol form of greenhouse warming. Needless to say, neither clouds nor dust can be factored into climate prediction models with much confidence.

Dust tied to climate

This TOMS image shows a record-setting Asian d...
Dust moving in April 2001 from arid areas in Central Asia and North Africa to the oceans. From NASA's Nimbus-7 satellite. Image via Wikipedia

At present the central areas of the oceans are wet deserts; too depleted in nutrients to support the photosynthesising base of a significant foot chain. Oddly, even when commonly known nutrients are brought to the ocean surface far from land by deep-sourced upwellings the effect on near-surface biomass is far from that expected. The key factor that is missing is dissolved divalent iron that acts as a minor nutrient for phytoplankton: even in deep ocean waters any such ferrous iron is quickly oxidised and precipitated as trivalent ferric compounds. One of the suggested means of geoengineering away any future climatic warming is to seed the far-off oceans reaches with soluble iron in the hope of triggering massive planktonic blooms, dead organisms sinking to be buried along with the their carbon content in the ocean-floor oozes. Retrospectively, it has been suggested that the slight mismatch between changes in atmospheric CO2 concentration and climate changes may be linked to fluctuating availability of iron dissolved from dust in ocean-surface waters, but so far that hypothesis has not been robustly tested. It is well known, however, that global cooling is accompanied by drying of continental climates and thereby an increase in the delivery of dust, even to polar ice caps where cores have shown dustiness to fluctuate with temperature.

Recently an ocean-floor sediment core from around 42° S has revealed a high-resolution record of the deposition of dust and iron at that location over the last 4 Ma (Martinez-Garcia, A. et al. 2011. Southern Ocean dust-climate coupling over the past 4 million years. Nature, v. 476, p. 312-315). In it one proxy for dust is the amount of organic compounds known as n-alkanes that are a major component of the waxes shed from plant leaves. Others are iron, titanium and thorium concentrations in the ooze. Dust proxies tally with land-ice volumes shown by the fluctuating d18O measured in bottom-dwelling foraminifera found as fossils in the core to form a convincing link between dust and climate over the Southern Ocean. Those proxies also match nicely the record of dust delivered to Antarctica that emerged from the 0.8 Ma Dome C ice core that was extracted and analysed by the EPICA consortium. The record shows boosts in iron and dust deposition at 2.7 Ma, when ice first took hold of northern high latitudes, and at 1.25 Ma when larger ice sheets began to develop and climate shifts switched to 100 ka cyclicity. Although the match between marine and glacial dust accumulation in the latter part of this mid-Pleistocene Transition is an important step forward in palaeoclimatology, it is a surprise that the new ocean-floor data is not plotted with the record of atmospheric CO2 in Antarctic ice bubbles: if there was a clear relationship that would have iced the cake.