Drought depiction, prediction and application: Implications for GDIS

The overall technical capabilities now exist to make real time, seasonal drought forecasts on a near global scale. By combining near real time observational data and seasonal forecasts from the new National Multi-model Ensemble (NMME) we plan to generate probabilistic predictions of multiple drought indicators and lead times of 1 to 9 months. Multiple drought indicators will be used since differential impacts of drought occur as a function of time scale. This paper examines some of the remaining technical challenges in making such global predictions while also examining the use of drought analysis and forecast information from an applied perspective. Examples of the latter include fire early warning in Kalimantan, food security outlooks in East Africa and research towards drought early warning in the agriculture sector in the Philippines and Sri Lanka.