The Raining Champions

A better understanding of the ground beneath our feet may be the key to improved seasonal weather forecasts, say an international team of environmental scientists in the journal Science today.

Researchers investigating the impact of soil moisture on rainfall have found that, across the globe, there are at least three hotspots where rainfall seems to be directly influenced by the amount of moisture in the soil, leading to the tantalising possibility of more accurate long-term weather forecasts.

Dr Christopher Taylor, from the Natural Environment Research Council‘s Centre for Ecology and Hydrology said: “We modelled soil moisture around the world and compared it with modelled rainfall. We found that in very wet or very dry areas the impact of changes in soil moisture on rainfall was negligible. But in the areas in between it was a different story. Here it seems, moisture and rainfall are connected.”

“Hope for accurate seasonal forecasts lie with computer models simulating the atmosphere’s response to slowly varying states of the ocean and land surfaces – these variables can be predicted weeks, or even months, in advance,” he added.

The teams, coordinated by Randal Koster at NASA’s Goddard Space Flight Center, have long known that the atmosphere responds to slow changes in the oceans, for example El Niño, but a great unknown was the atmosphere’s response to conditions on land.

Through a computer modelling experiment, they discovered a significant connection between rainfall and soil moisture in regions between very wet and very dry climates, such as: equatorial Africa just below the Sahara desert; the plains of North America; and across northern India. Less intense hotspots appear in South America, central Asia and China.

Rainfall can be predicted reasonably accurately a day in advance, even a few days in advance, but it has been difficult to forecast seasonally with any degree of accuracy.

Forecasts of this kind are invaluable to farmers who can use the information to plan growing seasons, schedule crop irrigation and reduce fertiliser use, thus reducing costs and improving profitability.

The project, partly by the Natural Environment Research Council, averaged out results from a dozen groups all performing the same highly controlled experiment. In this way, despite differences in the models’ results, areas of agreement in the results shone through.

Dr David Lawrence, who worked on the project for the Natural Environment Research Council’s Centre for Global Atmospheric Modelling, at the University of Reading said: “Routine monitoring of soil moisture in these hotspots may improve our long-term forecasting skills. There is still a great deal of uncertainty but the study shows that many independent models agree on these hotspots as the areas where soil moisture does affect rainfall.”