Some computer models for global warming may be over-estimating the risk of drought, according to a study published on September 12 by the journal Nature. Several key models used in climate research that factor in warming trends suggest that droughts will intensify as world temperatures rise.
This is on the basis that dry soils have less moisture to suck up into the atmosphere, which reduces rainfall and thus causes even greater aridity.
But scientists are worried that these models are too large in scale and lack observational data, especially about what happens locally.
Seeking to plug the knowledge gap, a four-nation team led by Chris Taylor from Britain's Centre for Ecology and Hydrology looked at images from weather saellites which track the development of storm clouds across the globe.
The scientists say they were surprised when they matched where new storms appear alongside images of how wet the ground was.
They found that afternoon storms were likely to occur when soils were parched - not over soils that were moist. The apparent reason: drier soils create stronger warm winds called thermals, which boost the chance of rain.
"Both heat and moisture are critical ingredients for rain clouds to build up during the afternoon," Taylor explained.
"On sunny days the land heats the air, creating thermals which reach several kilometres (miles) up into the atmosphere. If the soil is dry, the thermals are stronger, and our new research shows that this makes rain more likely." The data trawl covered six continents, looking at surface soil moisture and rainfall patterns on daily and three-hourly time steps, with a resolution of 50 to 100 kilometers (31 to 62 miles), over a decade.