Africa's fires mean less rain during the dry season

PASADENA, Calif., Aug. 6 (UPI) -- In any given year, half of all fires on Earth burned in Africa. Most are set by farmers to promote agricultural productivity and to clear land for planting.

But as a new study by NASA researchers shows, the fires are having distinct effects on the continent's weather. Smoke from Africa's agricultural fires suppress rain cloud formation and leads to less precipitation during the dry season.

Scientists at NASA's Jet Propulsion Laboratory discovered the correlation by comparing satellite imagery of burning and non-burning lands in tropical Africa, south of the Sahara Desert and north of the equator. Researchers compared the images with separate weather satellite data to see whether cloud formation over the course of day was influenced by the presence of agricultural fires and their thick plumes of smoke.

The researchers found that on days when satellite cameras captured smoky skies, fewer clouds formed and less rain fell.

"Less clouds and rainfall dry out the land and make it easier for farmers to ignite more fires, which data show they probably do," JPL scientist Michael Tosca said in a press release. "The added burning deepens and strengthens the effect and could lead to regional climate warming over time."

Researchers were able to confirm their findings using atmospheric computer models.

Tosca and his colleagues -- who published their recent findings in the journal Geophysical Research Letters -- say the soot rising from Africa's fires are especially dark and thick, the result of partially burned vegetation.

These large soot particles absorb sunlight heat up the air around them. When warm air rising from the sun-heated ground hits this thick layer of warm smoke, the air spreads out and convection -- the mixing and circulation of layers of air that encourages cloud formation -- is thwarted.

"We are able not only to show that the clouds decrease in the presence of aerosols, but that aerosols inhibit convection," Tosca said. "This effect is predicted by models, but it's really cool to see it in actual data."