Published online 26 March 2009 | Nature | doi:10.1038/news.2009.194


Settling of dust warms tropical Atlantic

A decline in aerosols could account for most of the rise in regional ocean temperatures.

Satellite picture of a dust stormA satellite image of a dust storm off the coast of West Africa.SeaWiFS Project, NASA/Goddard Space Flight Center, ORBIMAGE

Recent ocean warming believed to be fuelling Atlantic hurricanes may be largely the result of a reprieve from volcanoes and dust storms.

Sea surface temperatures in the tropical North Atlantic have climbed markedly since 1980, rising nearly 0.25 ºC per decade. The warming waters have been linked to increased hurricane activity, but climatologists differ on whether the temperature rise is due mainly to global warming or a natural cycle in ocean circulation.

Now a new study1 points to a different cause: a decline in the cooling influence of both sulphates spewed by volcanoes and dust kicked up by African winds. These particles, called aerosols, reflect sunlight and prevent it from penetrating the ocean. But researchers hadn't previously described in detail how fluctuating aerosol concentrations over decades influence ocean temperatures in the tropical North Atlantic, says lead author Amato Evan, a climate scientist at the University of Wisconsin-Madison.

Evan and his colleagues used satellite measurements of aerosols from 1982-2007, feeding them into a simple model of the tropical North Atlantic to estimate their effects on sea surface temperature. They found that varying amounts of particles over the sea could explain 69% of the warming trend over those years.

Aerosols from volcanic activity and African dust storms cooled the region in the 1980s and early 1990s, paving the way for a big rise in sea temperatures as the dust subsided later. Evan says about half the trend was down to the eruption of two tropical volcanoes early in the study period — El Chichón in Mexico in 1982 and Mount Pinatubo in the Phillipines in 1991. The decline of dust storms in the African Sahara and Sahel, possibly because a brutal Sahel drought passed its peak, accounts for another fifth of the warming.

Dust up

It makes sense that a lightening aerosol load is acting to warm the tropical North Atlantic, says oceanographer Greg Foltz of the University of Washington in Seattle. Volcanic eruptions are known to cool the climate worldwide, and the recent decrease in dust coming from Africa was also reported in previous studies. "If you put them together, it's not that surprising that it would have [this] effect," says Foltz. But it's impressive, he says, that the aerosol-driven model can reproduce so much of the ocean temperature variability.

Tim Barnett of the Scripps Institution of Oceanography in San Diego, an expert on how human-caused climate change is affecting oceans, agrees that the large estimated impact of aerosols is remarkable but says it should be considered speculative. More research is needed, he says, to determine how aerosols affect the total amount of heat flowing through the ocean.

Says Foltz, "This will add more fuel to the debate as to what's causing the increase in sea surface temperature."


Evan is now looking into how volcanic aerosols affect the ocean elsewhere, but he says the large impact of dust storms is probably a quirk of this particular ocean basin. Africa is the world's largest dust source, sending up 260-1600 million tonnes each year, much of it blown westward by trade winds.

What is unclear is how future climate change may influence Africa's dustiness. Changes in wind, rainfall and vegetation could all play a role. If the Atlantic dust cover is halved, a possibility one study has suggested for a world with doubled CO2 levels, Evan and his colleagues say these waters could warm up to 0.4 ºC more than global climate models have projected. But models can't yet offer a confident view of how such interactions would play out, says Evan: "Dust could be a huge positive or negative feedback on climate change, but we don't know." 

  • References

    1. Evan, A. T., Vimont, D. J., Heidinger, A. K., Kossin, J. P. & Bannartz, R. Science doi:10.1126/science.1167404 (2009).
Commenting is now closed.