Long-Range Forecasting in Support of Operations in the Horn of Africa
Abstract
Over the past several decades, the Horn of Africa (HOA) has experienced recurring climate variations, including droughts and floods that have devastated the region's livelihoods and prompted increased investment in strategies to minimize the negative effects of climate variations. These preventative strategies include the enactment of early warning systems, such as the Famine Early Warning System Network, and military commands such as U.S. Africa Command. If these organizations are to be successful, they must account for the many climate factors that affect Africa, including seasonal climate variations and climate change. Thus, skillful long-range forecasts, especially of precipitation, have become increasingly valuable in planning the operations of these organizations. In this study, we focused on assessing the potential for predicting HOA precipitation rate (PR) during the October-November rain season at lead times of several seasons. We correlated HOA PR and remote climate variables, and discovered a strong potential for skillful long-range forecasts of HOA PR using sea surface temperatures (SST) near New Zealand, the Philippines, and Namibia as predictors. Our forecast methods included deterministic (tercile matching, linear regression, optimal climate normals) and probabilistic (composite analysis) methods. Our verification metrics showed a definite improvement in forecast skill over existing long-range forecasts based on long-term means, and indicated that our forecasting methods have the potential to improve the planning of military and non-military operations in the HOA.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 2010
- Accession Number
- ADA531554
Entities
People
- Benjamin D. Lemke
Organizations
- Naval Postgraduate School