A Seasonal Rainfall Performance Probability Tool for Famine Early Warning Systems

This study reports on the development of a statistical tool which produces probabilistic outlooks of seasonal precipitation anomaly categories over Africa. Named the Seasonal Performance Probability (SPP), it quantitatively evaluates the probability of seasonal and sub-seasonal precipitation to finish at predefined percent of normal anomaly categories corresponding to below Average (<80% of Normal), Average (80-120% of Normal), and Above-Average (>120% of Normal) conditions. SPP is accomplished by applying Kernel Density Estimation (KDE) methods to compute smoothed, continuous density functions based on more than 30 years of historical precipitation data from the Africa Rainfall Climatology Version 2 (ARC2) dataset (Novella and Thiaw, 2013). Discussion of various KDE parameterizations tests to determine optimality of density estimates, and thus, performance of SPP for operational monitoring are presented. Verification results using Heidke Hit Proportion (HHP) scores and selected Africa monsoon case studies from 2006-2017 suggest that SPP provides reliable probabilistic outcomes of anomaly categories by early to mid-stages of rainfall seasons for major rainfall regions in east, west and southern Africa. SPP has been a useful tool in operational climate monitoring at CPC International desks, where it has helped to deliver early warning guidance for developing drought situations, and other related hydrometeorological climate anomalies. This is expected to promote better decision making in food security, planning and response objectives for USAID/FEWS-NET.