Prediction of African rainfall and household food security

Thomas E. Downing1, Richard Washington2
1Environmental Change Unit, University of Oxford, 1a Mansfield Road, OX1 3TB Oxford, United Kingdom.
Tel. : +44-18 65 28 11 80, Fax: +44-18 65 28 11 81,
E-mail: tom.downing@edu.ox.ac.uk
2School of Geography, University of Oxford, Oxford, United Kingdom

Despite the pioneering efforts of Blanford and Walker, seasonal forecasts of rainfall have only become operational in the last ten years or so. With its marked record of droughts, high interannual variability of rainfall and its largely agricultural economy, Africa is a prime potential benefactor of seasonal forecasting. This paper reviews recent developments in both climate prediction and the use of climate forecasts in promoting household food security. We outline the process of compiling and distributing seasonal forecasts and operational seasonal forecasts for Africa. The utility of climate forecasts is described against the background of household vulnerability and strategies to cope with food insecurity.

Climate forecasts may indeed revolutionise resource management in Africa. Yet, their utility ultimately depends on the linkages between geophysical, economic and social science insights as well as global linkages between institutes that provide and distribute forecasts and those responsible for responses, ultimately the vulnerable households throughout the world.

The use of climate predictions is related to the vulnerability of targeted users, both in terms of direct effects and in terms of how easy it is for a given group to access climate predictions and respond to them. Building institutional capacity to provide medium-term climate forecasts to enhance adaptive resource management in Africa would be a major step forward both in achieving present development aims and in preparing for climate change. Research on how to disseminate information, and ensure it is applicable at a household level is crucial.

Would better forecasts have altered the outcomes of diverse African food crises over the past decade or longer? The answer depends on how the political economy would have adapted to widespread dissemination of forecasts (and other data on climate and production). Good forecasts, in each case, would not have been sufficient to ensure early responses, to bolster sustainable livelihoods and to prevent vulnerable populations from being displaced.